phantom torso

share this!

May 28, 2009

The Phantom Torso Returns

Explore further

Feedback to editors

Mouse embryos grown in space for first time: Japan researchers

21 minutes ago

Saturday Citations: Mars limnology, phage immunology, quantum technology. Plus: The mushrooms are coming

20 hours ago

Scientists develop new method to create stable, efficient next-gen solar cells

Study uncovers giant fluctuation-enhanced phonon magnetic moments in a polar antiferromagnet

21 hours ago

Researchers reveal hidden sensory mechanism of hair follicles

Oct 27, 2023

Underwater robot finds new circulation pattern in Antarctic ice shelf

Meltwater flowing beneath Antarctic glaciers may be accelerating their retreat

Scratch my back, I'll scratch yours: Baboons wily cooperators

Earliest evidence of flip flops in the Middle Stone Age

Bringing a shark to a knife fight: 7,000-year-old shark-tooth knives discovered in Indonesia

Relevant physicsforums posts, does the universe prefer certain kinds of matter based on the laws in place.

Oct 26, 2023

Star cluster in the Milky Way appears to be as old as the Universe

Oct 25, 2023

Help identifying this Tumbling object in a night sky photo from Namibia

How feasible is home radio astronomy.

Oct 24, 2023

Sample and Return of Asteroid Bennu - live on Oct 20, 2020 at 1720 GMT

Our beautiful universe - photos and videos.

Oct 23, 2023

More from Astronomy and Astrophysics

Related Stories

How safe is travel to mars.

Oct 24, 2006

News from a space phantom

Oct 20, 2005

Odd list of body parts: which parts are most sensitive to solar flares?

Oct 27, 2005

How much radiation awaits lunar colonists? A new NASA mission aims to find out

Sep 9, 2005

Space radiation threats to astronauts addressed in federal research study

Oct 25, 2006

Brain stem cells sensitive to space radiation

Dec 11, 2007

Recommended for you

Two groups look at the economic viability of mining asteroids

Study shows need for ITU to tighten regulations for low orbit satellites as filing numbers grow

Oct 18, 2023

Signatures of the Space Age: Spacecraft metals left in the wake of humanity's path to the stars

Oct 16, 2023

US astronaut gets used to Earth after record-setting 371 days in space

Oct 14, 2023

'Ring of fire' eclipse brings cheers and shouts of joy as it moves across the Americas

Psyche spacecraft launched to mysterious and rare metal asteroid in first mission of its kind

Oct 13, 2023

Let us know if there is a problem with our content

Use this form if you have come across a typo, inaccuracy or would like to send an edit request for the content on this page. For general inquiries, please use our contact form . For general feedback, use the public comments section below (please adhere to guidelines ).

Please select the most appropriate category to facilitate processing of your request

Thank you for taking time to provide your feedback to the editors.

Your feedback is important to us. However, we do not guarantee individual replies due to the high volume of messages.

E-mail the story

Your email address is used only to let the recipient know who sent the email. Neither your address nor the recipient's address will be used for any other purpose. The information you enter will appear in your e-mail message and is not retained by Phys.org in any form.

Newsletter sign up

Get weekly and/or daily updates delivered to your inbox. You can unsubscribe at any time and we'll never share your details to third parties.

More information Privacy policy

Donate and enjoy an ad-free experience

We keep our content available to everyone. Consider supporting Science X's mission by getting a premium account.

E-mail newsletter

3 Dimensional Torso Phantom

• Description

COMPLETE WITH REMOVABLE ORGANS

The CIRS Anthropomorphic Torso Phantom is designed to provide an accurate simulation of an average male torso for medical imaging applications. The removable organs enable flexibility in the placement of TLD’s, contrast agents, etc. The epoxy materials used to fabricate the phantom provide optimal tissue simulation in the 40 keV to 20 MeV energy range.

The phantom simulates the physical density and linear attenuation of actual tissue to within 2 percent in the diagnostic energy range.

Each phantom contains removable organs. Included organs are lungs, heart, liver, pancreas, kidney, and spleen. The lower portion of the phantom contains a removable soft bolus material simulating a mix of 50 percent adipose and 50 percent muscle tissue.

This insert is used to maintain the position of the organs when the phantom is placed upright. For ease of removal, the bolus is enveloped in a screen-bag. Simulated muscle material layers the rib cage and vertebral column.

The exterior envelope simulates a mix of 30 percent adipose and 70 percent muscle tissue. The phantom is sealed at the bottom by an acrylic plate. Water or blood mimicking fluid can be used to fill all the interstitial voids.

• Removable lungs, heart, liver, pancreas, kidney and spleen
• Flexibility in TLD placement
• Optimal tissue simulation in the diagnostic energy range and up to 20 MeV
• Physical density and linear attenuation within 2 percent of actual tissue
• Interstitial voids fillable with water or blood-mimicking fluid
• Includes foam-lined carry case, user guide and 60-month warranty

3 Dimensional Torso Phantom: Data Sheet

Model 801-P

Model 715 Series

• Ultrasound FAQs

NOTICE: This system contains Controlled Unclassified Information (CUI)

NTRS - NASA Technical Reports Server

As of October 27, 2023, NASA STI Services will no longer have an embargo for accepted manuscripts. For more information visit NTRS News .

Available Downloads

Related records.

Where next?

Explore related content

Phantom Torso in HRF section of Destiny module

Nasa washington, dc, united states.

ISS002-E-6080 (2 May 2001) --- The Phantom Torso, seen here in the Human Research Facility (HRF) section of the Destiny/U.S. laboratory on the International Space Station (ISS), is designed to measure the effects of radiation on organs inside the body by using a torso that is similar to those used to train radiologists on Earth. The torso is equivalent in height and weight to an average adult male. It contains radiation detectors that will measure, in real-time, how much radiation the brain, thyroid, stomach, colon, and heart and lung area receive on a daily basis. The data will be used to determine how the body reacts to and shields its internal organs from radiation, which will be important for longer duration space flights. The experiment was delivered to the orbiting outpost during by the STS-100/6A crew in April 2001. Dr. Gautam Badhwar, NASA JSC , Houston , TX, is the principal investigator for this experiment. A digital still camera was used to record this image.

• Title: Phantom Torso in HRF section of Destiny module
• Date Created: 2001-05-02
• Rights: JSC
• Album: mgwhite

Get the app

Explore museums and play with Art Transfer, Pocket Galleries, Art Selfie, and more

A sensorized human torso phantom

Affiliation.

• 1 Automatic Control Laboratory, ETH Zürich, Switzerland.
• PMID: 15544299

Force-torque measuring input devices can significantly enhance the performance of classical simulation environments that are, for example, based on pure passive phantoms. Such devices allow not only the determination of force/torque amplitude and direction but also the contact point on the phantom. The force/torque information can be displayed visually or acoustically, drive a realistic graphical animation environment or it can be saved and compared with a haptic library comprising the force/torque history of any medical specialist. In this paper the technical principle is exemplified by an interactive human torso. A plastic phantom model of a human torso is instrumented with a 6-degree-of-freedom force/torque sensor, thus, allowing an intuitive and interactive use for education of human anatomy.

Publication types

• Research Support, Non-U.S. Gov't
• Models, Anatomic*
• User-Computer Interface

• Our Purpose
• Our Science and Technology
• Join Our Team
• Partner with Us
• Community and Education

The Lawrence Livermore Realistic Torso Phantom

In 1970, the Energy Research and Development Administration’s (now DOE) Intercalibration Committee for Low-Energy Photon Measurements established a forum to compare data and methods of plutonium whole-body counting – the methods for the measurement of radioactivity within the human body. For a variety of reasons, actinides, like plutonium 239, had proven especially difficult for health physicists to detect and quantify. In 1974, after an international, multilaboratory study, the committee concluded that the calibration phantoms (manikins) in use at the time were mostly inadequate and not designed to detect X-ray emitters in the human lung. As such, the need for a new phantom was warranted, and Livermore’s Hazards Control Department was selected to construct three realistic torso-only manikins, molded from human cadavers.

Livermore researchers began by gathering data (height, weight, and chest circumference) from more than 500 male employees at the Livermore and Los Alamos laboratories, to build a physical profile for the “average” male radiation worker. Based on the data, Lab researchers selected a cadaver of like dimensions from the University of California, San Francisco (UCSF) Anatomy Department from which molds of the torso and organ cavity were made.

After the molds were completed, the rib cage was extracted from the cadaver and cleaned. The remaining soft tissue was removed using a colony of dermestid beetles from the University of California, Berkeley. Lab researchers next affixed the rib cage to a dissolvable cast of the organ cavity to form an inner mold. This rib cage-affixed inner-mold was suspended in position inside of a hollow silicone mold, made from the model of the torso, and the space between was filled with tissue-equivalent (TE) plastic. The organ cavity mold was then dissolved, leaving the rib cage imbedded in the torso casting, surrounding the organ cavity.

Once the polyurethane torso shell with imbedded rib cage was completed, the simulated organs, molded from the cadaver’s organs, were inserted into the cavity, and a series of removable chest plates of TE material were fabricated to simulate a wide range of sizes. Two additional rib cages were obtained from the UCSF Anatomy Department and prepared in a similar manner to construct the second and third phantoms.

Upon completion, Livermore sent its three fabricated phantoms to laboratories in the United States, Europe, and Canada as part of an intercomparison program to compile and exchange data necessary to upgrade calibration and whole-body counting techniques. As the program proceeded, it became clear that each of the laboratories involved required use of the phantoms for longer periods of time than allotted. In addition, several facilities expressed interest in acquiring phantoms as permanent parts of their calibration programs. As such, Livermore agreed to fabricate a second set of torso manikins, and 16 second-generation phantoms were made. The new phantoms were nearly identical to the first three, except that a simulated rib cage replaced the real bone used in the original phantoms.

Following the development of the first and second-generation phantoms, and their subsequent commercial production, the validity of the Livermore phantom was tested in various international intercomparison studies. The tests conducted validated the Lab’s phantom as a highly realistic calibration tool.

The Livermore realistic torso phantom technology was transferred to the public sector in the early 1980s and is still commercially sold.

Pictured: Rib cage attached to the soluble organ cavity mold and positioned inside the hollow torso silicone mold.

• Core Competencies
• Demographics
• Unique Facilities
• Lab Directed Research and Development
• Institutional Initiatives
• Research Integrity
• S&T Highlights
• Recognition
• Director's Awards
• Journal Covers
• A Look Back
• LLNL Institutes/Centers
• Current and Former Lawrence Fellows
• Lawrence Fellowship: Learn more and apply
• Career Development & Network Activities
• Life Around the Lab
• Lawrence Livermore Postdoc Association
• Research SLAM!
• Internally Sponsored Internships
• Externally Sponsored Internships
• Veteran and Military Programs
• Virtual Discovery Center
• Virtual Tours and Class Visits
• Fun with Science
• Science on Saturday
• Biotech Summer Experience
• STEM with Smartphones
• Application Process
• Program Activities
• Teacher Research Academy
• Physics with Phones Introduction and Links
• Understanding Motion
• Exploring Acceleration
• Exploring Friction & Mechanics
• The Physics of Rotational Motion
• Moment of Inertia
• Investigating Impulse and Momentum
• The Science of Collisions
• Measuring the Pressure Around Us
• A Foundation for Understanding Waves
• Mechanical Waves and Sound
• Experimenting with Electromagnetic Waves
• Exploring Magnetic Fields in the World Around You
• Making Digital Measurements and Quantifying Uncertainty
• Ambassador Lecture Series
• Faculty Sabbatical Program
• Science Education
• Postdoc Opportunities
• Student Opportunities
• About Us: Academic Engagement Office
• Innovation and Partnerships Office
• Research Library

Welcome to the new NASA website. We’re working to continuously improve your web experience. If you don't find what you are looking for, give feedback .

Suggested Searches

• Climate Change
• Expedition 64
• Mars perseverance
• SpaceX Crew-2
• International Space Station
• View All Topics A-Z

Humans in Space

Earth & climate, the solar system, the universe, aeronautics, learning resources, news & events.

Langley Aeronautics Research Directorate Leadership

Body Wise Health Talks

NASA Technologies Receive Multiple Nods in TIME Inventions of 2023

• Search All NASA Missions
• A to Z List of Missions
• Upcoming Launches and Landings
• Spaceships and Rockets
• Communicating with Missions
• James Webb Space Telescope
• Hubble Space Telescope
• Why Go to Space
• Commercial Space
• Destinations
• Living in Space
• Explore Earth Science
• Earth, Our Planet
• Earth Science in Action
• Earth Multimedia
• Earth Science Researchers
• Pluto & Dwarf Planets
• Asteroids, Comets & Meteors
• The Kuiper Belt
• The Oort Cloud
• Skywatching
• The Search for Life in the Universe
• Black Holes
• The Big Bang
• Dark Energy & Dark Matter
• Earth Science
• Planetary Science
• Astrophysics & Space Science
• The Sun & Heliophysics
• Biological & Physical Sciences
• Lunar Science
• Citizen Science
• Astromaterials
• Aeronautics Research
• Human Space Travel Research
• Science in the Air
• NASA Aircraft
• Flight Innovation
• Supersonic Flight
• Air Traffic Solutions
• Green Aviation Tech
• Drones & You
• Technology Transfer & Spinoffs
• Space Travel Technology
• Technology Living in Space
• Manufacturing and Materials
• Science Instruments
• For Kids and Students
• For Educators
• For Colleges and Universities
• For Professionals
• Science for Everyone
• Requests for Exhibits, Artifacts, or Speakers
• STEM Engagement at NASA
• NASA's Impacts
• Centers and Facilities
• Directorates
• Organizations
• People of NASA
• Internships
• Our History
• Doing Business with NASA
• Get Involved
• Aeronáutica
• Ciencias Terrestres
• Sistema Solar
• All NASA News
• Video Series on NASA+
• Newsletters
• Social Media
• Media Resources
• Upcoming Launches & Landings
• Sounds and Ringtones
• Interactives
• STEM Multimedia

Progress Continues Toward NASA’s Boeing Crew Flight Test to Station

Hubble captures a galactic dance.

IXPE Untangles Theories Surrounding Historic Supernova Remnant

Misión récord de astronauta ayuda a planificar viajes al espacio profundo

Space Station Science Highlights: Week of May 15, 2023

Earth Observation From the Space Station

Powerful NASA-ISRO Earth Observing Satellite Coming Together in India

US-German Satellites Show California Water Gains After Record Winter

NASA Delivers First Flight Hardware to ESA for Lunar Pathfinder

NASA’s Curiosity Captures Martian Morning, Afternoon in New ‘Postcard’

NASA Scientists Make First Observation of a Polar Cyclone on Uranus

Webb, DART Missions Win AIAA 2023 Premier Awards

NASA’s Spitzer, TESS Find Potentially Volcano-Covered Earth-Size World

Government and Industry Collaboration Leads to First Air Taxi Delivery 

Joe Piotrowski, Jr.: One Engineer’s (Flight) Path to Airborne Science

Forget Movie Magic, NASA Armstrong has the Real Thing

NASA Tech Lets Scientists See Snow-Water Through the Trees

Primary Instrument for NASA’s Roman Completed, Begins Tests

STELLA: NASA’s DIY Educational Gadget for Measuring Plant Health

NASA Announces Teams for 2024 Student Launch Challenge

Scientific Highlights of NASA Astronaut Frank Rubio’s Year in Space

Historic Wind Tunnel Facility Testing NASA’s Mars Ascent Vehicle Rocket

New NASA Report Looks at Societal Considerations for Artemis

Ciencia destacada del año en el espacio del astronauta Frank Rubio

Pruebas de la NASA con maniquí de Artemis I aportan información para futuras misiones tripuladas

Space station 20th: sts-100 brings canadian robotic arm to the space station, johnson space center.

In April 2001, as the Expedition 2 crew of Yuri V. Usachev of Roscosmos and NASA astronauts James S. Voss and Susan J. Helms completed their first month aboard the International Space Station (ISS), the space shuttle Endeavour arrived at the space station on the STS-100 mission to install the Canadarm2 robotic system. During their first month in space, the Expedition 2 crew continued to commission the orbiting laboratory and conducted early research. They monitored the departure of a Progress cargo resupply vehicle and relocated their Soyuz spacecraft during a short flight around the space station. In addition to installing Canadarm2, the seven-member STS-100 crew, representing four of the ISS Program’s partner agencies, also brought new facilities to expand the research capabilities of the growing space station.

Following the departure from the space station of the space shuttle Discovery on March 19, 2001, Expedition 2 crew members Usachev, Voss, and Helms began their long-duration mission in earnest. They continued the commissioning of the Destiny U.S. laboratory module begun by the Expedition 1 crew. Helms conducted sessions with the Middeck Active Control Experiment-II instrument, first started during Expedition 1. On April 16, they monitored the departure of the Progress M44 cargo resupply vehicle from the Zvezda Service Module’s aft port. Two days later, Usachev, Voss, and Helms donned their Sokol launch and entry suits and boarded their Soyuz TM31 spacecraft. With Usachev at the controls, they undocked from the Zarya module’s Earth-facing port and flew to the newly vacated Zvezda aft port, redocking after a 30-minute flight. The maneuver freed up a docking port for the next cargo resupply ship. Then they prepared for the arrival of the next space shuttle mission.

The ninth space shuttle assembly and resupply mission to the ISS, STS-100, began on the afternoon of April 19, 2001, with the launch of space shuttle Endeavour from Launch Pad 39A at NASA’s Kennedy Space Center (KSC) in Florida. STS-100 included the most internationally diverse shuttle crew, with four of the ISS partnership organizations represented – Commander Kent V. Rominger and Pilot Jeffrey S. Phillips of NASA, and Mission Specialists Chris A. Hadfield representing the Canadian Space Agency, John L. Phillips and Scott E. Parazynski of NASA, Umberto Guidoni of the European Space Agency, and Yuri V. Lonchakov of Roscosmos. Their mission to install the Canadarm2 robotic on the space station involved the most complex series of robotic tasks of any shuttle mission up to that point. Less than two days after launch, Rominger guided Endeavour to a smooth docking with the station at the Pressurized Mating Adapter-2, or PMA-2, located on the forward end of the U.S. laboratory module Destiny. Because the shuttle reduced its internal pressure for the spacewalk the next day, the hatches between the station and the shuttle wouldn’t be opened for two more days. The crews exchanged a few items using the PMA-2 as an airlock.

The day after the docking, Ashby used the Shuttle’s Canadarm Remote Manipulator System, first flown on STS-2 in 1981, to lift the pallet carrying the Canadarm2 and a UHF antenna out of the shuttle’s payload bay and transfer it to a temporary location on the Destiny module. Hadfield and Parazynski began the mission’s first spacewalk. Parazynski first installed the UHF antenna on the outside of Destiny, then he and Hadfield removed eight bolts that held the Canadarm2 in its pallet for launch. They manually unfold its arms, and connected communications cables between the station and the pallet as Helms and Voss tested the connections from the robotics workstation inside Destiny. The two spacewalkers returned to the shuttle’s airlock after spending 7 hours 10 minutes outside.

On the fifth day of the mission, the crews opened the hatches between the shuttle and the station and warmly greeted each other. Using the shuttle’s robotic arm, Guidoni and Parazynski lifted the 20,000-pound Raffaello Multi-Purpose Logistics Module (MPLM) out of the payload bay and transferred it to the Unity module’s Earth-facing berthing port. Using the robotics workstation, Helms and Voss activated Canadarm2 and maneuvered it for the first time, commanding it to attach to a grapple fixture on Destiny’s exterior. At the end of the day, the crews returned to their respective spacecraft and closed the hatches in preparation for the following day’s second spacewalk.

The next day, Guidoni opened the hatches to the berthed MPLM, enabling the transfer of 6,000 pounds of cargo, including two science racks, to the space station. The Payload Operations Integration Center (POIC) at NASA’s Marshall Space Flight Center in Huntsville, Alabama, monitored the activation and operations of the two Expedite the Processing of Payloads for Space Station (EXPRESS) racks, and the transfer of active experiments into one of the facilities. The POIC has served as the nerve center for all NASA research activities aboard the ISS for more than 20 years. Meanwhile, Hadfield and Parazynski ventured outside for their second spacewalk. They began the 7-hour 40-minute excursion by installing cables to connect the Canadarm2 to its grapple fixture on Destiny’s exterior. Helms and Voss activated the robotic arm, now a permanent fixture on the space station, and as Hadfield and Parazynski watched, they commanded it to lift the pallet off the station and hold it in a temporary parked position. The spacewalkers completed their excursion after installing a new avionics component and removing unneeded equipment from the station’s exterior. At the end of the day, the crews reopened the hatches between the station and the shuttle.

During the next three days, transfers from the MPLM continued and crew members transferred several powered active experiments from the shuttle’s middeck to EXPRESS Rack 1. The active experiments included a plant growth chamber, two protein crystallization facilities, and a commercial bioprocessing facility. Other experiments transferred and installed in Destiny included a phantom torso, a life-size model of a human upper body embedded with radiation detectors, part of a suite of radiation experiments that included the Bonner Ball Neutron Detector delivered to the station during the STS-102 mission. The crew members also loaded the MPLM with equipment no longer needed on station for return to Earth. The shuttle crew members also conducted maintenance tasks on the space station, including repairing the treadmill.

On the mission’s tenth day, using the shuttle’s robotic arm, Hadfield reached up and grappled the pallet and then Helms released it from the station’s arm, completing a historic “Canadian handshake” in space. Hadfield returned the now empty pallet to the shuttle’s payload bay. With all the transfers completed, the crews closed the hatches to the MPLM and Parazynski returned it to the payload bay using the shuttle’s robotic arm. Rominger fired Endeavour’s thrusters to raise the space station’s orbit. Then it was time for the two crews to say their goodbyes after eight days of joint operations, and they closed the hatches between the two spacecraft for the last time.

The following day, with Ashby at the controls, Endeavour undocked from the space station. He flew Endeavour around the station as the crew photographed it, with the newly installed Canadarm2 clearly visible. With inclement weather at the primary landing site at KSC on May 1, Rominger brought Endeavour home for a smooth landing at NASA’s Dryden Flight Research Center, now NASA’s Armstrong Flight Research Center at Edwards Air Force Base in California. The mission to install the Canadarm2 onto the space station lasted 11 days 21 hours 30 minutes.

Enjoy the crew-narrated video about the STS-100 mission.

To be continued…

Additive Manufacturing of Human Torso Phantom for Microwave Imaging

Ieee account.

• Change Username/Password
• Update Address

Purchase Details

• Payment Options
• Order History
• View Purchased Documents

Profile Information

• Communications Preferences
• Profession and Education
• Technical Interests
• US & Canada: +1 800 678 4333
• Worldwide: +1 732 981 0060
• Contact & Support
• About IEEE Xplore
• Accessibility
• Terms of Use
• Nondiscrimination Policy
• Privacy & Opting Out of Cookies

A not-for-profit organization, IEEE is the world's largest technical professional organization dedicated to advancing technology for the benefit of humanity. © Copyright 2023 IEEE - All rights reserved. Use of this web site signifies your agreement to the terms and conditions.

• Original research
• Open access
• Published: 03 May 2022

3D printed anthropomorphic left ventricular myocardial phantom for nuclear medicine imaging applications

• Janos Kiss   ORCID: orcid.org/0000-0002-4379-8409 1 ,
• Laszlo Balkay 2 ,
• Kornel Kukuts 3 ,
• Marton Miko 2 ,
• Attila Forgacs 3 , 4 ,
• Gyorgy Trencsenyi 2 &
• Aron K. Krizsan 3  

EJNMMI Physics volume  9 , Article number:  34 ( 2022 ) Cite this article

2252 Accesses

2 Citations

4 Altmetric

Metrics details

Anthropomorphic torso phantoms, including a cardiac insert, are frequently used to investigate the imaging performance of SPECT and PET systems. These phantom solutions are generally featuring a simple anatomical representation of the heart. 3D printing technology paves the way to create cardiac phantoms with more complex volume definition. This study aimed to describe how a fillable left ventricular myocardium (LVm) phantom can be manufactured using geometry extracted from a patient image.

The LVm of a healthy subject was segmented from 18 F-FDG attenuation corrected PET image set. Two types of phantoms were created and 3D printed using polyethylene terephthalate glycol (PETG) material: one representing the original healthy LVm, and the other mimicking myocardium with a perfusion defect. The accuracy of the LVm phantom production was investigated by high-resolution CT scanning of 3 identical replicas. 99m Tc SPECT acquisitions using local cardiac protocol were performed, without additional scattering media (“in air” measurements) for both phantom types. Furthermore, the healthy LVm phantom was inserted in the commercially available DataSpectrum Anthropomorphic Torso Phantom (“in torso” measurement) and measured with hot background and hot liver insert.

Phantoms were easy to fill without any air-bubbles or leakage, were found to be reproducible and fully compatible with the torso phantom. Seventeen segments polar map analysis of the "in air” measurements revealed that a significant deficit in the distribution appeared where it was expected. 59% of polar map segments had less than 5% deviation for the "in torso” and "in air” measurement comparison. Excluding the deficit area, neither comparison had more than a 12.4% deviation. All the three polar maps showed similar apex and apical region values for all configurations.

Conclusions

Fillable anthropomorphic 3D printed phantom of LVm can be produced with high precision and reproducibility. The 3D printed LVm phantoms were found to be suitable for SPECT image quality tests during different imaging scenarios. The flexibility of the 3D printing process presented in this study provides scalable and anthropomorphic image quality phantoms in nuclear cardiology imaging.

Performance measurements and optimization of nuclear medicine imaging systems involve the use of different phantoms to mimic human activity distributions [ 1 , 2 , 3 ]. Accurate anthropomorphic phantoms have been introduced to reveal quantitative inaccuracies and to detect the presence of image artefacts caused by inappropriate acquisition, reconstruction, and image processing [ 4 , 5 , 6 , 7 , 8 ]. Several of these phantoms are commercially available, generally with fixed size and geometry. 3D printing technology including direct ink writing [ 9 ], fused deposition modelling (FDM) [ 10 , 11 , 12 , 13 ], digital light processing (DLP) [ 14 ] or stereo-lithography (SLA) [ 15 ] offers large variety of possibilities to design custom-made geometrical and anthropomorphic phantoms [ 16 , 17 , 18 , 19 , 20 , 21 , 22 , 23 ]. A systematic review by Filippou et al. concludes that 3D printing methods can complete or replace commercially available phantoms in the fields of CT, MRI, PET, SPECT, US, and mammography imaging [ 24 ]. Several studies reveal 3D printed phantom solutions for nuclear medicine applications using real patient imaging data, including fillable multicompartmental torso in quantitative imaging analysis for 90 Y-DOTATATE radiopeptide therapy [ 25 ], fillable kidney phantom for 177 Lu SPECT reconstruction optimization [ 26 ], as well as tumor phantom set of various shapes for testing comparison of PET radiomics features in a multi-center approach [ 27 ]. Focusing on cardiac phantom solutions, Matsutomo et al. designed and printed a set of specific inserts to simulate different ischemic levels to complete the commercially available Myocardial SPECT Phantom HL (Kyoto Kagaku Co., Ltd., Kyoto, Japan) [ 28 ]. Grice et al. introduced a left ventricle (LV) cardiac phantom with simplified wall geometry containing low perfusion lesions within a non-anthropomorphic background container, printed from polylactic acid (PLA) material [ 29 ]. The endeavor of creating new cardiac phantoms is encouraged by clinically relevant, but unanswered methodological questions. The lack of geometrically appropriate cardiac phantom prevents the investigation of image artefacts and image processing failures attributed to the inhomogeneity of the cardiac wall thickness. 3D printing technology makes it possible to create the complex geometry of a real heart, which is not feasible with traditional manufacturing methods. This study aimed to determine whether creating a fillable, anatomically accurate 3D printed left ventricle myocardium (LVm) phantom segmented from a PET image volume of a real patient is feasible. The suitability of polyethylene terephthalate glycol (PETG) plastic for anthropomorphic LVm phantom production is demonstrated for the first time. The phantom insert was designed to be compatible with the commercially available Anthropomorphic Torso Phantom (DataSpectrum Co., Durham, NC). CT images are presented to confirm the reproducibility of 3D printing and phantom preparation. Furthermore, SPECT measurements were performed to demonstrate that the proposed phantoms give a complementary solution to the currently available phantoms in the field of nuclear cardiology imaging.

Materials and methods

Phantom design.

Input image data for the phantom design were extracted from an 18 F-FDG PET/CT study of a patient (age: 67 years; weight: 63 kg) without known coronary artery disease. Whole body PET/CT acquisition was performed on a GE Discovery MI system, using the local patient examination protocol (injected dose 220 MBq, 1.5 min acquisition time per bed position, 30% overlap between bed positions). Q.Clear reconstruction was applied with 384 × 384 matrix resulting in 1.82 × 1.82 × 2.79 mm voxel size (Fig.  1 upper row). The local ethics committee approved the use of patient data in this study.

Three orthogonal views of reconstructed 18 F-FDG PET/CT image (upper row) and the 3D phantom design (bottom row)

For image segmentation process, we used the 3D Slicer software with basic functionality [ 30 ]. Images were cropped near to the area of the heart. Segmentation was done by applying the Otsu method with minimum and maximum threshold values obtained by visual inspection [ 31 ]. Irrelevant segmented voxels were deleted manually using the Erase tool. The 3D mask was saved in Standard Tessellation Language (STL) format with LPS (Left, Posterior, Superior) coordinate system and size scale of 1.0. The exported model was post-processing using Autodesk Meshmixer (Autodesk Inc., San Rafael, California, USA). The Plane Cut tool was used to make a plane surface on the left ventricle model from the direction of the left atrium perpendicular to the apex. A 0.5 mm offset distance was defined to create the model hollow, while solid accuracy and mesh density parameters were set to 512 in the Hollow tool. The size of the plain surface of the model was increased with the Extrude tool to make a 10 mm wide solid pedestal. In addition, with the Hollow and Extrude tools, a bubble trap was created. A phantom holder was also created in Trimble SketchUp Pro 2020 (Trimble Inc., Sunnyvale, California, USA) based on the distance and size of the pedestal holes of the commercially available Biodex Cardiac insert. This holder was merged with the previously created plane surface of the cardiac model in Meshmixer using Boolean Union method (Fig.  1 ). As a last step, two filling holes were designed, one of them through the bubble trap. Finally, with our primary purpose, two types of phantom models were designed: one as a representation of the healthy LVm with 190 ± 1 ml fillable volume (Fig.  1 lower row), and another mimicking transmural perfusion defective myocardium with a 20 × 30 mm oval solid plastic cold part (Fig.  2 ). The latter model has 165 ± 1 ml total fillable volume. These two phantoms will be referred to in the following as LVm healthy and LVm defective phantom.

Multi-sectional image of the real 3D printed LVm defective phantom

• 3D printing

For slicing and creating the print plan, the Repetier-Host (Hot-World GmbH & Co. KG, Willich, Germany) software was used. The model was laid flat on its pedestal. Slicing parameters were the followings: 100% infill density; shell thickness was 0.4 mm with 0.2 mm layer height. No adhesion or support was generated for the 40 mm/s print speed. Retraction and cooling were enabled. Phantoms were printed using an Anet A8 FDM 3D printer (Anet Technology Co., Ltd., Shenzhen, People's Republic of China), build volume 220 × 220 × 240 mm, Marlin firmware, 0.4 mm nozzle diameter, with 3DJAKE PETG transparent filament. PETG thermoplastic was used for watertight and durability reasons and to avoid significant stringing, which is a well-known phenomenon in the case of other printing materials (e.g., PLA) [ 32 ]. Print bed and nozzle temperature were set to the mid-value of the manufacturer's recommended temperature ranges: 70 °C and 240 °C, respectively. The total 3D printing time was approximately 6 h for each model. The 0.4 mm nozzle diameter and 0.4 mm shell thickness print parameter give 0.4 mm real wall thickness to the printed phantoms. A few times, the phantom had clearly visible separate layers on the outer apex side after printing, in these cases, we used a soldering iron to melt them together. To prevent any further leakage between layers, Prisma Color Acrylic spray was applied to the outer surface on the printed phantoms as a finishing process. To assure as less leaking as possible, M5 size screws were 3D printed to tightly fit in the phantom filling holes (Fig.  3 ).

Photographs of the real 3D printed LVm healthy phantom before ( a ) and after ( b ) a red food-dye diluted water filling

The described steps required to create LVm phantom models are shown in Fig.  4 . Our 3D phantom model is available in STL format in the supplementary material.

Flowchart of steps required to create the LVm phantom models. Input data were an 18 F-FDG PET/CT study of a patient without known coronary artery disease. Three different software were used for model construction, and an additional program was applied to create the printing plan at different stages of the manufacturing process. Finished models were printed with an Anet A8 FDM 3D printer

Printing reproducibility, leaking test

Reproducibility of the phantom production was demonstrated with three separate printing series. Size, including the diameter of the pedestal and filling holes, was measured with a sliding calliper. Spiral CT scans with 120 kV, 120 mA x-ray settings, and voxel size of 0.625 × 0.703 × 0.703 mm were performed and evaluated on healthy phantoms filled with water to measure the accuracy of reproducibility. For the leakage test, watertight fillings were checked at least two times for each phantom (Fig.  3 ).

Phantom SPECT/CT measurements

99m Tc- water solution was mixed with red food-dye for better visual detection of bubbles and leakage. Decay corrected activity concentrations calculated to the acquisition start can be seen in Table 1 .

Measurements and reconstructions

Both LVm healthy and LVm defective phantoms were measured without additional scattering media (referring to as "in air” measurement). The LVm healthy insert was placed into the Anthropomorphic Torso Phantom for a second acquisition (referring to as "in torso” measurement). All imaging acquisitions were performed with identical acquisition parameters on a NaI(Tl) detector-based AnyScan® DUO FLEX SPECT/CT system (Mediso Medical Imaging Systems, Budapest, Hungary) equipped with Low energy High-Resolution (LEHR) parallel hole collimator. The different phantom arrangements on the SPECT/CT scanner bed can be seen in Fig.  5 . The routine clinical patient protocol for myocardial perfusion was selected, including the following parameters: 90 degrees scan arc, 64 projections, 128 × 128 matrix size, 140.5 keV energy with 20% window width, body contouring, and step and shoot mode. Additionally, a CT scan with 120 kV, 50 mA x-ray settings, and a voxel size of 2.50 × 0.977 × 0.977 mm was performed for attenuation correction purposes.

"In air” measurement positioning of LVm phantom on the SPECT/CT scanner bed ( a ), and the “in torso” setup with the Anthropomorphic Torso Phantom, when the LVm phantom is placed at the location of the original heart insert ( b )

Data processing

Data were processed by the Mediso InterView™ XP application. Default Cardiac Perfusion Image reconstruction of Tera-Tomo™ 3D SPECT-Q was applied on the acquired raw data. Image size of 128 × 128 with 5.91 mm cubic reconstruction voxel size, 32 number of iterations and 4 subset size was used with CT-based attenuation and scatter correction. Polar maps with 17 segments were created for all three measurements and were applied to reveal similarities and differences. For this process, reorientations were performed by a medical expert physician with six years of experience. Polar maps of "in air" measurements of the healthy and defective myocardium phantoms were compared, to demonstrate how the defect alters the internal distribution of the radioactive solution inside the phantom. Since the "in torso” measurement represents more realistic scattering and attenuation conditions, the polar map of the "in air" measurement of the healthy myocardium phantom was compared to the polar map of the "in torso” measurement. Polar map graphs from the calculated percentage differences were also created for these evaluations.

After rigid registration of the high-resolution CT images of LVm healthy models, the phantoms present identical geometry within tight tolerances in shape (Fig.  6 ) and fillable volume.

Representative sagittal view of the registered CT images of the three identical LVm healthy phantoms

The measured mean filled volume was 189.4 ml ± 1.4 ml, including the volume of the bubble trap. The printed LVm phantoms were easily refillable and were closed tightly, without any air bubbles or observable leakage during all of the presented measurements. Additionally, two phantoms were filled and stored for three months at room temperature, and no leakage or evaporation was detected. Reconstructed SPECT images of "in air” measurements reveal accurate uptake volumes (Fig.  7 ). Differences between defective and healthy phantom images were clearly visible on the sagittal views (Fig.  7 a, b) as well as on the 3D rendered image (Fig.  7 c, d). The defect appeared where it was planned during the phantom design.

Uptake patterns of reconstructed SPECT images of LVm healthy and defective phantoms measured "in air” on three orthogonal views ( a , b ) and 3D rendered images of the two phantom realizations ( c , d )

Printed phantoms were compatible with the Anthropomorphic Torso Phantom to be assembled at the cardiac region. The reconstructed SPECT images revealed that the activity distribution of the LVm healthy phantom could be visualized in detail (Fig.  8 ).

Reconstructed SPECT image of the LVm healthy phantom inserted in the Anthropomorphic Torso Phantom

Clear differences were found while analyzing the resulted polar maps of the three measurements of the "in air” and "in torso” arrangements (Fig.  9 ).

Original (first column) and 17 segments (second column) polar maps of the three measurements

The polar map segment with the highest signal was found to be the basal anterolateral for LVm healthy and LVm defective "in air” measurements. On the other hand, the originally high signal mid-inferior region on the LVm healthy model was decreased significantly due to the artificial defect on the LVm defective model.

Polar map segment differences of the LVm healthy phantom measurements (first and last rows in Fig.  9 ) could originate from at least two sources. The radiopharmaceutical activity decayed compared to the "in air” case; therefore, the overall signal yield was expectedly lower. Moreover, the liver and the background in the torso phantom contained image distortions due to the spillover effect. All three polar maps have similar apex and apical region values.

The detailed relative perfusion percentage values of each region for all three measurements are summarized in Table 2 , together with the relative percentage difference of measurement comparisons.

In the relative % difference columns (column IV. and V.), negative value means deterioration, while a positive represents an improved region. The values of the LVm healthy—LVm defective phantom comparison (column IV.), are in the range between − 24.7% and 10.6%, and 11 of the 17 segments have less than 5% value. The LVm healthy phantom "in air”–"in torso” comparison (column V.) has values between − 12.4%, and 10.2%, and 10 of the 17 segments have values less than 5%. The relative percentage differences were also depicted on differential polar maps (Fig.  10 ) based on column IV. and V. values of Table 2 . Each color indicates 5 percentage steps. At the LVm healthy versus LVm defective phantom comparison, the deviations were higher than 5% deterioration concentrated on the four inferior regions where the artificial defect was designed. Two regions showed an improved signal ratio of more than 5%. Improvement and relapse regions in the case of LVm healthy phantom "in air” versus "in torso” comparison did not come from the nature of our phantom. As the concerning graph shows, the deviation is located in the basal edge regions, while the values are still around 5% except for the basal anterolateral region. This result and the 10.2% improvement in the apical region can be attributed to the uncertainty of the manual heart reorientation.

Polar maps of the relative percentage differences for different LVm phantom measurements. Left panel: results of the LVm healthy phantom "in air” vs. "in torso” comparison. Right panel: results of the LVm healthy vs. LVm defective phantom comparison

While several conventional plastic phantoms are available to test the image quality and reliability of nuclear cardiology applications with SPECT [ 8 , 33 , 34 , 35 ], they still have some anatomical and size limitations. 3D printing technology has gained wide attention recently for creating anthropomorphic phantoms, due to its cost-effectiveness, fast production capability and the possibility for advanced and customized design in almost any shape even for nuclear cardiology applications [ 28 , 29 ]. In this work, two anatomically accurate LV myocardial phantom inserts were created from a real patient 18 F-FDG PET/CT study image set (Fig.  1 upper row). One represents the original healthy LV myocardium (Fig.  1 lower row), and the other includes an artificially added myocardium deficit (Fig.  2 ). Three LVm healthy phantoms were 3D printed to verify that there are no significant alterations in geometry (Fig.  6 .) and fillable volume (189.4 ml ± 1.4 ml). These phantom inserts were planned to be convenient and complementary solutions to the commercially available plastic phantoms used in nuclear cardiology. Bubbles in the myocardium volume of the LV phantoms could affect the distribution of the radioactive solution. The LV insert of the Anthropomorphic Torso Phantom has no bubble trap, while the 3D printed LVm inserts were designed to include one for bubble-free filling of the LV wall. Therefore, the imaging of our phantoms was not affected by the presence of bubbles in the artificial myocardium volume. Moreover, the conventional LV insert is available at a certain size in a geometrically simple shape [ 33 ]; however, our 3D printed LVm insert is scalable in size and results in a more realistic uptake pattern of the LV myocardial perfusion SPECT image (Fig.  7 ). The latter has particularly high significance in the case of testing optimal settings for image reconstruction algorithms to avoid artifacts. The LV myocardium wall has a significantly different cross-sectional diameter at the apex than other regions, and the iterative image reconstruction tends to reach accurate activity levels at different iteration numbers for the apex than to the LV walls [ 36 ] even in case of a geometrically simple LV phantom. This is more prominent when we consider the real anatomy of the LV with non-uniform wall thickness. Therefore, our phantom design is a good advocate to the geometrically simple LV phantoms to find optimal iteration number for a certain image reconstruction. In addition, the reduction in left ventricular apical tracer uptake called apical thinning or false apical defect [ 37 ] is frequently observed in myocardial perfusion imaging both in the field of PET [ 38 ] and SPECT imaging [ 39 , 40 ]. Among many potential causes, the diminished activity at the apex can be attributed to real anatomy [ 41 ] combined with the partial volume effect, as it is visible in our phantom model as well (Fig.  6 ). Another commercially available phantom called the Kyoto HL cardiac torso phantom (Kyoto Kagaku Co. Ltd., Kyoto, Japan) was used by Yoneyama et al. to test image reconstruction resolution recovery solutions to overcome ejection fraction (EF) limitations in case of pediatric patients [ 42 ]. However, with our method, two small size hearts can be printed from normal gated PET image sets in end-systole and end-diastole phases, and the EF measurement accuracy of different reconstruction methods can be tested [ 43 ]. It has to be mentioned that the commercially available AGATE phantom [ 8 ] can mimic simple heart motion and is compatible with the Anthropomorphic Torso Phantom. Therefore, gated SPECT acquisitions and EF calculation are possible; however, this phantom is also available only in adult patient size. The anatomically correct design of the LV myocardium is also important when comparing hybrid or ellipsoid sampling of polar map generation [ 44 ]. A set of printed phantoms with different clinically representative cases could be used to perform a comparison of existing nuclear cardiology software, since considerable differences are present in their performance [ 45 ]. We performed a representative set of SPECT image acquisitions using the 3D printed phantom inserts. Both LVm healthy and LVm perfusion defect phantoms were filled with 99m Tc, and SPECT acquisitions were performed on an AnyScan® DUO FLEX SPECT/CT system "in air”, without any scattering media and in the Anthropomorphic Torso Phantom including hot background and hot liver insert. The printed LVm models remained intact throughout the experiments, and the inserted 99m Tc radioactive solution did not dissolve into the torso phantom background chamber. Seventeen segments polar map analysis of SPECT images revealed that by comparing the LVm defective model to the LVm healthy one, a significant deficit in the radiopharmaceutical distribution appeared where it was expected (Fig.  9 ). The design process enables flexibility in placing the perfusion deficit with different numbers and shapes within the fillable wall of the phantom. Including the expected low perfusion segments at the deficit area, around 65% of the polar map segments had less than 5% deviation. When comparing the LVm healthy model, the "in torso” and "in air” measurements, 59% of all polar map segments had less than 5% deviation (Fig.  10 ). Concerning only the segments excluding the deficit area, neither comparison revealed more than 12.4% deviation, which difference could have originated most probably from phantom positioning error, the applied reconstruction method, and the well-known spill-over effect, especially in the case of the "in torso” phantom measurement. Beyond the flexibility and applicability of our method, this study has several limitations. We used an 18 F-FDG PET image set to create the phantom model; however, a more realistic model can be created with currently available PET myocardial perfusion traces such as 82 Rb-chloride, 13  N-ammonia or 18 F-flurpiridaz. We presented 3D printed phantoms in one LV size from a non-gated PET data of a healthy male patient. It would be beneficial to demonstrate phantom studies using healthy females and pediatric or even heart disease images as input data. On the other hand, the variety of these printable LVm phantoms should be limited to provide a few standardized shapes available to be downloaded and printed in any nuclear cardiology laboratory. The LV deficit designed and printed in the phantom was completely solid, representing scar burden. However, a fillable defect could be printed within the LV wall, and lower activity concentration can be inserted in that chamber to mimic ischemia. In this study, only the LV was segmented; however, the anthropomorphic nature of the phantom could be emphasized with a model including the right ventricle as well.

In this study, we proved that creating a fillable, anthropomorphic 3D printed phantom of the LV myocardium segmented from a real patient PET image volume is possible. SPECT images were acquired in different imaging scenarios proving the usefulness of the printed LVm phantoms. The flexibility of the 3D printing process presented in this study provides scalable and anthropomorphic image quality phantoms in nuclear cardiology imaging.

Availability of data and materials

Our phantom inserts are available in STL format in the supplementary material.

Abbreviations

Digital light processing

Ejection Fraction

Fused deposition modelling

Low Energy High-Resolution

Left, Posterior, Superior

Left ventricle

Left ventricle myocardium

Polyethylene terephthalate glycol

Polylactic acid

Stereo-lithography

Standard Tessellation Language

Ryu H, Meikle SR, Willowson KP, Eslick EM, Bailey DL. Performance evaluation of quantitative SPECT/CT using NEMA NU 2 PET methodology. Phys Med Biol. 2019;64:145017.

Article   CAS   Google Scholar  

Attarwala AA, Hardiansyah D, Romanó C, Jiménez-Franco LD, Roscher M, Wängler B, et al. Performance assessment of the ALBIRA II pre-clinical SPECT S102 system for 99mTc imaging. Ann Nucl Med. 2021;35:111–20. https://doi.org/10.1007/s12149-020-01547-7 .

Article   CAS   PubMed   Google Scholar  

Khorshidi A. Assessment of SPECT images using UHRFB and other low-energy collimators in brain study by Hoffman phantom and manufactured defects. Eur Phys J Plus. 2020;135:1–19. https://doi.org/10.1140/epjp/s13360-020-00238-6 .

Article   Google Scholar  

Blaire T, Bailliez A, Ben Bouallegue F, Bellevre D, Agostini D, Manrique A. First assessment of simultaneous dual isotope (123I/99mTc) cardiac SPECT on two different CZT cameras: a phantom study. J Nucl Cardiol. 2018;25:1692–704.

Timmins R, Ruddy TD, Wells RG. Patient position alters attenuation effects in multipinhole cardiac SPECT. Med Phys. 2015;42:1233–40.

Zhang D, Ghaly M, Mok GSP. Interpolated CT for attenuation correction on respiratory gating cardiac SPECT/CT—a simulation study. Med Phys. 2019;46:2621–8.

Veress AI, Fung GSK, Lee TS, Tsui BMW, Kicska GA, Segars WP, et al. The direct incorporation of perfusion defect information to define ischemia and infarction in a finite element model of the left ventricle. J Biomech Eng. 2015;137:1–10.

Visser JJN, Sokole EB, Verberne HJ, Habraken JBA, Van De Stadt HJF, Jaspers JEN, et al. A realistic 3-D gated cardiac phantom for quality control of gated myocardial perfusion SPET: the Amsterdam gated (AGATE) cardiac phantom. Eur J Nucl Med Mol Imaging. 2004;31:222–8.

Kim S, Oh J, Jeong D, Park W, Bae J. Consistent and reproducible direct ink writing of eutectic gallium-indium for high-quality soft sensors. Soft Robot. 2018;5:601–12.

Abdullah KA, McEntee MF, Reed W, Kench PL. Development of an organ-specific insert phantom generated using a 3D printer for investigations of cardiac computed tomography protocols. J Med Radiat Sci. 2018;65:175–83.

Okkalidis N, Chatzigeorgiou C, Okkalides D. Assessment of 11 available materials with custom three-dimensional-printing patterns for the simulation of muscle, fat, and lung hounsfield units in patient-specific phantoms. J Eng Sci Med Diagn Ther. 2018;1:1–7.

Google Scholar  

Vyavahare S, Teraiya S, Panghal D, Kumar S. Fused deposition modelling: a review. Rapid Prototyp J. 2020;26:176–201.

Hong D, Lee S, Kim GB, Lee SM, Kim N, Seo JB. Development of a CT imaging phantom of anthromorphic lung using fused deposition modeling 3D printing. Medicine (United States). 2020;99:e18617.

Zhang J, Hu Q, Wang S, Tao J, Gou M. Digital light processing based three-dimensional printing for medical applications. Int J Bioprinting. 2020;6:12–27.

Robinson SS, Aubin CA, Wallin TJ, Gharaie S, Xu PA, Wang K, et al. Stereolithography for personalized left atrial appendage occluders. Adv Mater Technol. 2018;3:1–9.

Ramírez-Nava GJ, Santos-Cuevas CL, Chairez-Oria I, Rioja-Guerrero E, Oroz-Duarte J. Tomographic 99mTc radioactivity quantification in three-dimensional printed polymeric phantoms with bioinspired geometries. Radiat Phys Chem. 2020;177:109130. https://doi.org/10.1016/j.radphyschem.2020.109130 .

Gear JI, Cummings C, Craig AJ, Divoli A, Long CDC, Tapner M, et al. Abdo-Man: a 3D-printed anthropomorphic phantom for validating quantitative SIRT. EJNMMI Phys. 2016. https://doi.org/10.1186/s40658-016-0151-6 .

Article   PubMed   PubMed Central   Google Scholar  

Anwari V, Lai A, Ursani A, Rego K, Karasfi B, Sajja S, et al. 3D printed CT-based abdominal structure mannequin for enabling research. 3D Print Med. 2020. https://doi.org/10.1186/s41205-020-0056-9 .

Alqahtani MS, Lees JE, Bugby SL, Samara-Ratna P, Ng AH, Perkins AC. Design and implementation of a prototype head and neck phantom for the performance evaluation of gamma imaging systems. EJNMMI Phys. 2017. https://doi.org/10.1186/s40658-017-0186-3 .

Woliner-van der Weg W, Deden LN, Meeuwis APW, Koenrades M, Peeters LHC, Kuipers H, et al. A 3D-printed anatomical pancreas and kidney phantom for optimizing SPECT/CT reconstruction settings in beta cell imaging using 111In-exendin. EJNMMI Phys. 2016. https://doi.org/10.1186/s40658-016-0165-0 .

Gear JI, Cummings C, Sullivan J, Cooper-Rayner N, Downs P, Murray I. Radioactive 3D printing for the production of molecular imaging phantoms. Phys Med Biol. 2020. https://doi.org/10.1088/1361-6560/aba40e .

Article   PubMed   Google Scholar  

Läppchen T, Meier LP, Fürstner M, Prenosil GA, Krause T, Rominger A, et al. 3D printing of radioactive phantoms for nuclear medicine imaging. EJNMMI Phys. 2020. https://doi.org/10.1186/s40658-020-00292-0 .

Robinson AP, Tipping J, Cullen DM, Hamilton D, Brown R, Flynn A, et al. Organ-specific spect activity calibration using 3d printed phantoms for molecular radiotherapy dosimetry. EJNMMI Phys. 2016;3:1–11. https://doi.org/10.1186/s40658-016-0148-1 .

Filippou V, Tsoumpas C. Recent advances on the development of phantoms using 3D printing for imaging with CT, MRI, PET, SPECT, and ultrasound. Med Phys. 2018;45:e740–60.

Gear JI, Long C, Rushforth D, Chittenden SJ, Cummings C, Flux GD. Development of patient-specific molecular imaging phantoms using a 3D printer. Med Phys. 2014;41:1–4.

Tran-Gia J, Lassmann M. Optimizing image quantification for 177Lu SPECT/CT based on a 3D printed 2-compartment kidney phantom. J Nucl Med. 2018;59:616–24.

Pfaehler E, Beukinga RJ, de Jong JR, Slart RHJA, Slump CH, Dierckx RAJO, et al. Repeatability of 18F-FDG PET radiomic features: a phantom study to explore sensitivity to image reconstruction settings, noise, and delineation method. Med Phys. 2019;46:665–78.

Matsutomo N, Seki H, Hishikawa M, Motegi K, Yamamoto T. Technical Note: development of an ischemic defect model insert attachable to a commercially available myocardial phantom. Med Phys. 2020;47:4340–7.

Green S, Grice J. Technical note: 3D-printed phantom for dedicated cardiac protocols and geometries in nuclear medicine. Med Phys. 2021. https://doi.org/10.1002/mp.15406 .

Fedorov A, Beichel R, Kalpathy-Cramer J, Finet J, Fillion-Robin JC, Pujol S, et al. 3D Slicer as an image computing platform for the Quantitative Imaging Network. Magn Reson Imaging. 2012;30:1323–41. https://doi.org/10.1016/j.mri.2012.05.001 .

Liu D, Yu J. Otsu method and K-means. In: Proceedings of 2009 9th Int Conf Hybrid Intell Syst HIS 2009, vol. 1. 2009. p. 344–9.

Paraskevoudis K, Karayannis P, Koumoulos EP. Real-time 3D printing remote defect detection (stringing) with computer vision and artificial intelligence. Processes. 2020;8:1464.

Imbert L, Poussier S, Franken PR, Songy B, Verger A, Morel O, et al. Compared performance of high-sensitivity cameras dedicated to myocardial perfusion SPECT: a comprehensive analysis of phantom and human images. J Nucl Med. 2012;53:1897–903.

Liu CJ, Cheng JS, Chen YC, Huang YH, Yen RF. A performance comparison of novel cadmium–zinc–telluride camera and conventional SPECT/CT using anthropomorphic torso phantom and water bags to simulate soft tissue and breast attenuation. Ann Nucl Med. 2015;29:342–50. https://doi.org/10.1007/s12149-015-0952-z .

Kobayashi H, Momose M, Kanaya S, Kondo C, Kusakabe K, Mitsuhashi N. Scatter correction by two-window method standardizes cardiac I-123 MIBG uptake in various gamma camera systems. Ann Nucl Med. 2003;17:309–13.

Okuda K, Nakajima K, Yoneyama H, Shibutani T, Onoguchi M, Matsuo S, et al. Impact of iterative reconstruction with resolution recovery in myocardial perfusion SPECT: phantom and clinical studies. Sci Rep. 2019;9:1–9. https://doi.org/10.1038/s41598-019-56097-4 .

Purser NJ, Armstrong IS, Williams HA, Tonge CM, Lawson RS. Apical thinning: real or artefact? Nucl Med Commun. 2008;29:382–9.

Steffen DA, Giannopoulos AA, Grossmann M, Messerli M, Schwyzer M, Gräni C, et al. “Apical thinning”: relations between myocardial wall thickness and apical left ventricular tracer uptake as assessed with positron emission tomography myocardial perfusion imaging. J Nucl Cardiol. 2020;27:452–60.

Denisova NV, Ansheles AA. A study of false apical defects in myocardial perfusion imaging with SPECT/CT. Biomed Phys Eng Express. 2018. https://doi.org/10.1088/2057-1976/aae414 .

Okuda K, Nakajima K, Matsuo S, Kondo C, Sarai M, Horiguchi Y, et al. Creation and characterization of normal myocardial perfusion imaging databases using the IQ·SPECT system. J Nucl Cardiol. 2018;25:1328–37.

Johnson KM, Johnson HE, Dowe DA. Left ventricular apical thinning as normal anatomy. J Comput Assist Tomogr. 2009;33:334–7.

Yoneyama H, Nakajima K, Okuda K, Matsuo S, Onoguchi M, Kinuya S, et al. Reducing the small-heart effect in pediatric gated myocardial perfusion single-photon emission computed tomography. J Nucl Cardiol. 2017;24:1378–88.

Germano G, Kavanagh PB. Ready, shoot, aim? Summary justice for small hearts in nuclear cardiology. J Nucl Cardiol. 2017;24:1389–92.

Germano G, Kavanagh PB, Waechter P, Areeda J, Van Kriekinge S, Sharir T, et al. A new algorithm for the quantitation of myocardial perfusion. SPECT I: technical principles and reproducibility. J Nucl Med. 2000;41:712–9.

CAS   PubMed   Google Scholar  

Johansson L, Lomsky M, Marving J, Ohlsson M, Svensson SE, Edenbrandt L. Diagnostic evaluation of three cardiac software packages using a consecutive group of patients. EJNMMI Res. 2011;1:1–7.

Download references

Acknowledgements

Not applicable.

Open access funding provided by University of Debrecen. The research was supported by the Thematic Excellence Programme (TKP2020-NKA-04) of the Ministry for Innovation and Technology in Hungary.

Author information

Authors and affiliations.

Division of Radiology and Imaging Science, Department of Medical Imaging, Faculty of Medicine, University of Debrecen, Nagyerdei krt. 98., Debrecen, 4032, Hungary

Division of Nuclear Medicine and Translational Imaging, Department of Medical Imaging, Faculty of Medicine, University of Debrecen, Nagyerdei krt. 98., Debrecen, 4032, Hungary

Laszlo Balkay, Marton Miko & Gyorgy Trencsenyi

ScanoMed Nuclear Medicine Centers, Nagyerdei krt. 98., Debrecen, 4032, Hungary

Kornel Kukuts, Attila Forgacs & Aron K. Krizsan

Mediso Ltd., Laborc Utca 3., Budapest, 1037, Hungary

Attila Forgacs

You can also search for this author in PubMed   Google Scholar

Contributions

AKK proposed the original idea, helped to find the right model design, and helped the polar map generation. LB, GT and AF helped to plan the measurements and analyzed the results from technical aspects. KK helped to complete the phantom measurements. MM analyzed the results from medical aspects. JK segmented, CAD designed, and 3D printed the presented LVm models as well as contributed to the measurements and wrote the manuscript draft. All authors read and approved the final manuscript.

Corresponding author

Correspondence to Janos Kiss .

Ethics declarations

Ethics approval and consent to participate.

The use of the 18 F-FDG PET/CT data of the corresponding anonym patient in this study was approved by the local ethics committee.

Consent for publication

Competing interests.

This study was performed as part of the first author’s PhD project. The other authors declare that they have no competing interests.

Additional information

Publisher's note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ .

Reprints and Permissions

About this article

Cite this article.

Kiss, J., Balkay, L., Kukuts, K. et al. 3D printed anthropomorphic left ventricular myocardial phantom for nuclear medicine imaging applications. EJNMMI Phys 9 , 34 (2022). https://doi.org/10.1186/s40658-022-00461-3

Download citation

Received : 14 October 2021

Accepted : 20 April 2022

Published : 03 May 2022

DOI : https://doi.org/10.1186/s40658-022-00461-3

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Technetium-99m

The U.S. Department of Energy’s Radiological and Environmental Science Laboratory (RESL) recently produced the world’s first lung phantom that contains the entire Thorium-232 decay series.

The lung phantom will be used to calibrate and test radiation detection systems used by the Department and other agencies throughout the federal government.

How Are the Phantoms Created?

Phantoms are replicas of human organs that interact with x-rays and gamma rays the same way real organs do in people. They’re made by pouring a mixture of polyurethane and calcium carbonate into silicone rubber molds, which are based on plaster casts of a cadaver’s organs. Lawrence Livermore National Laboratory previously had the only Thorium-232 lung phantom in the world prior to RESL, however it was not in equilibrium . It takes around 50 years for Thorium-232 to break down or decay into a series of other elements before it can reach equilibrium—or when the radioactivity of all these elements remains constant. Having the decay series in equilibrium allows researchers to know exactly how much Thorium-232 is in the body or lungs without measuring for it directly.

Thorium-232 emits alpha radiation that isn’t easily detectible because it is absorbed by body tissue. The other elements in the decay series emit gamma radiation, which can be detected and used to inform calculations of the amount of Thorium-232 in the body. RESL incorporated the entire Thorium-232 decay series into its lung phantom by using 60-year-old thorium oxide material stored at Idaho National Laboratory before verifying the results. The whole process took less than a year and is currently the only one of its kind in the world to reach equilibrium. "Thorium is three times more abundant than uranium in the Earth’s crust and can also be used as nuclear fuel,” said Guy Backstrom, the director of RESL . “As the clean energy revolution takes place across America, the responsible development of all energy resources will help ensure our continued leadership in clean energy. RESL’s world-class scientists make sure that the measurements made to monitor the workers that are developing these energy systems remain safe.”

Background on RESL

RESL is a government-owned, government-operated laboratory in DOE’s Idaho Operations Office. The lab focuses on analytical chemistry as well as radiation calibration and measurements. RESL operates quality assurance programs to help confirm that DOE operations are completed in a safe and environmentally responsible manner. They confirm the quality and stability of laboratory measurement systems throughout DOE to ensure the reliability of the data being used to make decisions. RESL is currently working to make whole-body phantoms containing the entire Thorium-232 decay series using similar materials.

26th Southern Biomedical Engineering Conference SBEC 2010, April 30 - May 2, 2010, College Park, Maryland, USA pp 414–417 Cite as

A Tissue Equivalent Phantom of the Human Torso for in vivo Biocompatible Communications

• David M. Peterson 5 , 6 ,
• Walker Turner 5 ,
• Kevin Pham 5 ,
• Hong Yu 5 ,
• Rizwan Bashirullah 5 ,
• Neil Euliano 6 &
• Jeffery R. Fitzsimmons 5  
• Conference paper

2035 Accesses

Part of the IFMBE Proceedings book series (IFMBE,volume 32)

A tissue equivalent phantom (TEQ) was designed and constructed for in vivo biocompatible communication systems operating from 902-928 MHZ (Industrial, Scientific and Medical (ISM) band). The tissue equivalent phantom was designed by first noting the permittivity and conductivity of various tissues in the human torso using the FCC website http://www.fcc.gov/oet/rfsafety/dielectric.html, then by mixing the appropriate amounts of TX-151 (a polysaccharide gel), distilled water, sodium chloride and sucrose until the different regions of the phantom matched the parameters of the human torso Initial values were recorded based on previous work at lower frequencies and determined empirically at 915MHz. Computer modeling studies of human tissue were performed over the 902-928MHz band using a finite difference time domain computer modeling program (xFDTD, RECOM). Comparative analysis was conducted to determine the performance of the phantom. The phantom allows for testing and evaluation of very small antenna devices designed for in vivo diagnostics and monitoring.

• Vector Network Analyzer
• Specific Absorption Rate
• Very High Frequency
• Human Body Model
• Radio Frequency Energy

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

This is a preview of subscription content, access via your institution .

Buying options

• Available as PDF
• Read on any device
• Instant download
• Own it forever
• Compact, lightweight edition
• Dispatched in 3 to 5 business days
• Free shipping worldwide - see info

Tax calculation will be finalised at checkout

Purchases are for personal use only

Unable to display preview.  Download preview PDF.

Author information

Authors and affiliations.

University of Florida, Gainesville, FL, USA

David M. Peterson, Walker Turner, Kevin Pham, Hong Yu, Rizwan Bashirullah & Jeffery R. Fitzsimmons

Convergent Engineering, Gainesville, FL, USA

David M. Peterson & Neil Euliano

You can also search for this author in PubMed   Google Scholar

Editor information

Editors and affiliations.

Dept. Bioengineering, University of Maryland, Glenn L. Martin Hall 2181, 20742, College Park, Maryland, USA

Keith E. Herold

Biomed Research Foundation, 20832, Olney, USA

Jafar Vossoughi

Fischell Dept. of Bioengineering, University of Maryland, 20742, College Park, USA

William E. Bentley

Rights and permissions

Reprints and Permissions

Copyright information

© 2010 Springer-Verlag Berlin Heidelberg

About this paper

Cite this paper.

Peterson, D.M. et al. (2010). A Tissue Equivalent Phantom of the Human Torso for in vivo Biocompatible Communications. In: Herold, K.E., Vossoughi, J., Bentley, W.E. (eds) 26th Southern Biomedical Engineering Conference SBEC 2010, April 30 - May 2, 2010, College Park, Maryland, USA. IFMBE Proceedings, vol 32. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-14998-6_105

Download citation

DOI : https://doi.org/10.1007/978-3-642-14998-6_105

Publisher Name : Springer, Berlin, Heidelberg

Print ISBN : 978-3-642-14997-9

Online ISBN : 978-3-642-14998-6

eBook Packages : Engineering Engineering (R0)

Share this paper

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Find a journal
• Publish with us

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings
• Advanced Search
• Journal List
• v.233(1); 2018 Jul

Deformable torso phantoms of Chinese adults for personalized anatomy modelling

Hongkai wang.

1 Department of Biomedical Engineering, Faculty of Electronic Information and Electrical Engineering, Dalian University of Technology, Dalian, Liaoning, China

Xiaobang Sun

2 Department of Information Technology, University of Jyväskylä, Jyväskylä, Finland

Tongning Wu

3 China Academy of Industry and Communications Technology, Beijing, China

Congsheng Li

Zhonghua chen, meiying liao, zhaofeng chen.

4 Institute of Digital Medicine, Third Military Medical University, Chongqing, China

5 The Affiliated Cancer Hospital of Hainan Medical College, Haikou, Hainan, China

Hongcheng Shi

6 Department of Nuclear Medicine, Zhongshan Hospital, Fudan University, Shanghai, China

7 Department of Nuclear Medicine, Peking Union Medical College Hospital, Beijing, China

Yanjun Zhang

8 Department of Nuclear Medicine, the First Affiliated Hospital of Dalian Medical University, Dalian, Liaoning, China

9 Trauma Department of Orthopaedics, the First Affiliated Hospital of Dalian Medical University, Dalian, Liaoning, China

Shaoxiang Zhang

Changjian liu.

In recent years, there has been increasing demand for personalized anatomy modelling for medical and industrial applications, such as ergonomics device development, clinical radiological exposure simulation, biomechanics analysis, and 3D animation character design. In this study, we constructed deformable torso phantoms that can be deformed to match the personal anatomy of Chinese male and female adults. The phantoms were created based on a training set of 79 trunk computed tomography ( CT ) images (41 males and 38 females) from normal Chinese subjects. Major torso organs were segmented from the CT images, and the statistical shape model ( SSM ) approach was used to learn the inter‐subject anatomical variations. To match the personal anatomy, the phantoms were registered to individual body surface scans or medical images using the active shape model method. The constructed SSM demonstrated anatomical variations in body height, fat quantity, respiratory status, organ geometry, male muscle size, and female breast size. The masses of the deformed phantom organs were consistent with Chinese population organ mass ranges. To validate the performance of personal anatomy modelling, the phantoms were registered to the body surface scan and CT images. The registration accuracy measured from 22 test CT images showed a median Dice coefficient over 0.85, a median volume recovery coefficient ( RC vlm ) between 0.85 and 1.1, and a median averaged surface distance ( ASD ) < 1.5 mm. We hope these phantoms can serve as computational tools for personalized anatomy modelling for the research community.

Introduction

Digital human phantoms are widely used in anatomy‐based medical simulations, anatomy education, and industrial design. As analysed in several reviews (Zaidi & Xu, 2007 ; Zaidi & Tsui, 2009 ; Xu, 2014 ), a new direction of digital human phantom research is to model individual human anatomy, which is the basic requirement of person‐specific applications, such as clinical radiological dose calculation (Alziar & Bonniaud, 2009 ; Divoli et al. 2009 ; Marcatili et al. 2015 ; Momennezhad et al. 2016 ), electromagnetic exposure evaluation (Conil et al. 2008 ; Wu et al. 2011 ; Li et al. 2017 ), patient‐specific biomechanics simulation (Bijar et al. 2016 ; Miller, 2016 ; Wittek et al. 2016 ), anatomy education (Blum et al. 2012 ), and animation character design (Ali‐Hamadi et al. 2013 ). Hereafter we will first briefly review the history of digital human phantoms and the state‐of‐the‐art of personal anatomy modelling and then propose our idea of deformable human phantoms that can match individual human anatomy based on statistical learning from a large training set of real‐subject computed tomography (CT) images.

The earliest digital human phantoms were the ‘stylized phantoms’ composed of simple geometric primitives of spheres, cylinders, and slabs (ICRP, 1959 ). Later, more anatomically realistic phantoms were constructed by segmenting medical images [CT or magnetic resonance (MR)] of the reference human subjects (Gibbs et al. 1984 ; Williams et al. 1986 ). Beginning in the 1990s, visible human project (VHP) datasets were generated in several countries (Ackerman, 1998 ; Zhang et al. 2004 ; Park et al. 2006 ) using cryosectioning images of human cadavers, leading to the creation of a series of highly detailed phantoms, including the VIP‐man, voxel‐man, Chinese adult anatomical models, and the VHK‐man (Schiemann et al. 2000 ; Xu et al. 2000 ; Sang et al. 2008 ; Wu et al. 2011 ). In the 21st century, more phantoms were constructed using boundary representation (BREP), e.g. non‐uniform rational B‐spline surface (NURBS) or polygonal meshes. Due to the convenience of modelling shape deformation, BREP was used to adjust the organ shape and size of existing phantoms to match the reference anthropometry values (e.g. body weight, height, and organ volumes), resulting in the ‘reference human phantoms’ of average anthropometry parameters, such as the RPI‐AM/AF phantoms (Zhang et al. 2009 ), FASH/MASH phantoms (Cassola et al. 2010 ; Kramer et al. 2010 ), and the PSRK‐man for Koreans (Kim et al. 2011 ).

Later, to meet the requirements of population‐orientated simulation, anatomical phantoms representing different percentiles of height, weight, and body mass index (BMI) were created from multi‐subject medical images or by adapting the anatomy of existing reference human phantoms. The Foundation for Research on Information Technologies in Society (IT'IS) in Zurich, Switzerland used MR images of volunteers to create a ‘virtual family’ (Christ et al. 2010 ) and ‘virtual population’ library (Gosselin et al. 2014 ) of different genders, ages, heights, and weights. The RPI‐AM/AF phantoms were adjusted to create phantoms of different BMIs (Ding et al. 2012 ) and different breast cup sizes (Hegenbart et al. 2008 ). Farah et al. ( 2010 ) from IRSN of France also constructed 34 female torso phantoms of different chest girths and breast cup sizes. The FASH/MASH phantoms were deformed to create adult and paediatric phantoms matching different percentiles of heights and weights (Cassola et al. 2011 ). The 4D XCAT phantoms (Stabin et al. 2012 ) were also adapted to create phantoms of adults, children, and pregnant females according to reference organ mass values (ICRP, 2002 ). The University of Florida (UF) group created a library of UF family phantoms (Geyer et al. 2014 ) by segmenting CT and MR images of adults, children, and newborns, followed by the adjustment of organ meshes to match the reference organ masses. Segars et al. ( 2013 ) morphed the 4D XCAT phantoms to match CT datasets of 58 adults and 10 children (Li et al. 2008 ; Segars & Tsui, 2009 ). Broggio et al. ( 2011 ) from IRSN of France selected 25 representative body surface scans from the CAESAR database (Robinette et al. 2002 ) and fit the internal organs into each body surface by adjusting the organ sizes according to the literature equations.

As reflected from the above history, a trend in digital human phantom development is to model increasingly specific anatomical subtypes, ultimately for the individual person. To realize individualized anatomy modelling, some studies have constructed deformable phantoms to match individual human anatomy. The concept of the deformable phantom was proposed during the early development of BREP phantoms. A well‐known example is the 4D NURBS‐based cardiac‐torso (NCAT) phantom developed by Segars et al. ( 2001 ). By deforming the NUBS surface of torso anatomy, this phantom models respiration motion and heart beating and also emulates inter‐subject variations in body height, chest measurements, diaphragm position, heart size, position, and orientation. Based on a similar strategy, Segars et al. ( 2010 ) later developed the 4D extended cardiac‐torso (XCAT) phantom of the entire body, including more structures and finer timing resolution of cardiac/respiratory motions than the NCAT phantoms. Na et al. ( 2010 ) at the Rensselaer Polytechnic Institute (RPI) developed deformable adult human phantoms that can adapt the body height, weight, and organ volumes. Their adaptation of phantom anatomy was realized by adjusting the organ meshes of the RPI‐AM/AF phantoms (Zhang et al. 2009 ) to match the anthropometry values of different population percentiles.

For personalized anatomy modelling, a limitation of the existing deformable phantoms is that they model inter‐subject variations at a relatively coarse level (e.g. body weight, height, and fat quantity). Although the 4D NCAT and XCAT phantoms enable the adjustment of organ size, position, and orientation, they do not model the realistic shape variations between different individuals. Therefore, it is desirable to create deformable phantoms by learning realistic anatomical variations from a large training set of real‐subject medical images.

The objective of this study was to construct digital human phantoms that can be deformed to match the anatomy of different individuals. To learn realistic inter‐subject anatomical variations, we used the statistical shape modelling (SSM) technique, which has been successfully used for anatomical variation modelling (Heimann & Meinzer, 2009 ) of human organs (Mofrad et al. 2010 ) and small animals (Wang et al. 2012 , 2015 ). The training images used in this study include 79 health‐screening positron emission tomography/computed tomography (PET/CT) images from four central hospitals in China. Because the health‐screening images mainly cover the torso region, this study constructs torso phantoms for male and female adults, which are named the DCHT‐M and DCHT‐F (Deformable Chinese Torso Male and Female, respectively) phantoms. To match the deformable torso phantoms with individuals, we used the active shape model (ASM) approach (Heimann & Meinzer, 2009 ), which was commonly adopted to register SSM with patient data. In this study, individual body surface scan and torso CT image were used as the targets of phantom matching, obtaining the estimation of individual subject trunk anatomy.

The workflow of phantom construction is illustrated in Fig.  1 . The male and female phantoms were constructed using the same workflow.

The workflow of phantom construction. (A) The low‐dose torso CT images of the training subjects. (B) Surface rendering of the segmented type I organs. (C) The template mesh of the reference human model and the torso region cut from the template. (D) Type I and II organs rendered in different colours. (E) Type I organs registered to the training subjects. (F) Type II organs mapped via the registered type I organs. (G) The constructed phantom.

Data collection

Because it is difficult to recruit a large number of healthy volunteers for CT or MR acquisition, we collected medical images already stored in the hospital database. In past decades, thousands of Chinese people have received PET/CT scans for early cancer screening (Tong, 2016 ), although most of them were diagnosed as asymptomatic. The PET/CT scans of asymptomatic subjects within the ages of 20–80 and weights of 40–120 kg were collected as the training images of this study. We used only the CT images for phantom construction, leaving the PET images for future metabolism modelling research. Figure  1 A shows typical low‐dose CT images of the health‐screening PET/CT scans. We collected 79 PET/CT images of normal Chinese adults from four central hospitals in the northeast, southeast, and central areas of China. Table  1 lists the number of collected images for different genders and ages. All the images cover the body region from the neck to the upper thigh, including the entire pelvis. The CT pixel sizes ranged from 0.59 to 1.37 mm, and the inter‐slice spacing was between 1.25 and 3.00 mm. The CT scanner settings were 100–140 kV for tube potential and 28–298 mA for tube current.

The number of collected images for different genders and ages

Ethics statement

This study was performed under the ethical approval from Dalian University of Technology Ethics Committees. No patient identification information has been used in this research or presented in this paper.

Organ segmentation

Due to the low X‐ray dose used for typical PET/CT acquisition, the CT image contrast only facilitates the segmentation of bones and major trunk organs. All the organ structures were segmented using semi‐automatic methods, followed by the proofreading of a radiologist with over 10 years of working experience.

The whole body, skeleton, and lungs were segmented using the thresholding method, followed by manual correction using the mitk software (Wolf et al. 2005 ). The segmentation results were further polished by morphological closing and hole‐filling procedures. Based on the whole skeleton segmentation, each individual bone was separated using the interactive graph cuts method (Boykov & Jolly, 2000 ). The limb bones were excluded, as they are not fully covered by the CT scan. Internal soft organs, including the pericardium, liver, spleen, kidneys, pancreas, aorta, inferior vena cava, and torso cavity, were semi‐automatically segmented using the contour interpolation tool of mitk software. The segmentation of skeletal muscles and subcutaneous fat were constrained between the surfaces of the torso cavity and the skin using thresholding with a manually adjusted threshold for each subject. All the segmented organs were converted into triangular surface meshes (Fig.  1 B) using the marching cubes algorithm (Lorensen & Cline, 1987 ).

Template mesh registration

Due to the imperfect image contrast of low‐dose CT, not all the organs can be segmented from the CT images. We refer to the segmented organs as ‘type I organs’ and the unsegmented organs as ‘type II organs’. To compensate for the missing type II organs, a 3D template model of complete human anatomy (Fig.  1 C) was registered to each training subject. The template models for the male and female were purchased from the TurboSquid web store (Turbosquid, 2017 ), from which the torso structures of both genders were cut out (Fig.  1 C). Table  2 lists all the anatomical structures contained in the torso region. The meshes of type I template organs (Fig.  1 D) were registered to the segmented organs using the robust point matching (RPM) method (Chui & Rangarajan, 2003 ). Afterwards, the meshes of type II template organs were mapped to the individual subjects via thin‐plate‐spline (TPS) interpolation method. The control points ( P C ) of the TPS interpolation were selected as the type I organ vertices within 10 mm to the type II organ surfaces. After the template registration, the motion vectors of the control points ( V C ) were calculated as the difference between their registered positions ( P ~ C ) and the original positions (i.e. V C = P ~ C − P C ). From V C , we interpolated the motion vectors ( V II ) for all the type II organ vertices ( P II ), using the TPS interpolation method (Bookstein, 1996 ), and then obtained the mapped positions of the type II organ vertices as P ~ II = P II + V II .

List of organ structures included in the phantom (type I organs are marked with bold and italic font)

For the registration of type I organs, the RPM method sometimes generates unsatisfactory results for complex‐shaped organs (e.g. the vertebrae). We developed a marker‐based version of the RPM method for these difficult cases. A user interface was programmed for manually specifying an arbitrary number of landmark pairs on the two meshes to be registered. Each landmark is duplicated n d times and added to the point cloud being registered. In this way, the RPM method simultaneously matches the mesh vertices and the landmarks. The parameter n d serves as a weighting factor for the landmarks. It should be large enough to guide correct registration and small enough to tolerate potential user bias. An empirical value of n d  = 10 was used in this study, which yielded visually correct registration and an averaged surface distance (i.e. the average distance between the closest vertices of the two meshes) below 0.1 mm. To ensure the proper registration of individual vertebrae, six landmarks were specified for each vertebra at the vertebra body centre, the two superior articular facets, the two transverse costal facets, and the tip of the spinous process. Figure  1 E demonstrates the typical template registration results of type I organs. Figure  1 F shows the mapped type II organs together with the registered type I organs.

Construction of the statistical shape model

As a result of the template mesh registration, each vertex of the template mesh were mapped to the corresponding anatomical locations of different training subjects. The registered template meshes were used to represent the organ shapes of individual training subjects, so that different subjects have the same number of mesh vertices. The inter‐subject anatomical variations were modelled as the changes in corresponding mesh vertex coordinates between the training subjects. The statistical shape model (SSM) method was used to construct the deformable phantoms (Fig.  1 F). Prior to the construction of the SSMs, the generalized Procrustes analysis (GPA) (Bookstein, 1996 ) should be applied to remove the inter‐subject differences of translation, rotation, and scaling, such that only the shape difference remains. Unlike the conventional GPA method, we did not remove the scaling differences because body size variation is an important feature of human anatomy.

After the GPA step, the 3D vertex coordinates of all the registered template organs were concatenated to form the shape vector of each training subject. Let X i = ( x i ,1 ,  y i ,1 ,  z i ,1 ,  x i ,2 ,  y i ,2 ,  z i ,2 , …,  x i , N ,  y i , N ,  z i , N ) be the shape vector of the i th training subject, where ( x i,j , y i,j , z i,j ) denotes the 3D coordinate of the j th vertex of the subject i . N is the total vertex number of all the organs in the template mesh, which is 363 916 and 43 539 for the male and female phantoms, respectively. Because both the vertex number N and the order of vertices in the shape vector were inherited from the original template mesh, every training subject had the same number of elements (i.e. 3 N ) and the same order of vertex arrangement in the shape vector.

Principal component analysis (PCA) was applied to construct the SSM based on the shape vectors. First, mean shape vector X ¯ of all the training subjects was calculated, and the shape vector of each training subject was centralized by subtracting X ¯ from X i , i.e. X i ~ = X i − X ¯ , where X ~ i was the centralized shape vector. Let Q = [ X ~ 1 , X ~ 2 … , X ~ k ] be the matrix containing the shape vectors of k training subjects and each subject corresponds to one column of Q . The size of Q was 3 N  ×  k , where k  ≪ 3 N , as the number of training subjects was much less than the number of template mesh vertices. The value of k was 41 and 38 for the male and female phantoms, respectively. Conventionally, PCA performs eigendecomposition of the covariance matrix QQ T , but in our study the size of QQ T (3 N  × 3 N ) was too large for a direct eigendecomposition. Instead, we performed eigendecomposition of Q T Q (size k  ×  k ) and then left‐multiplied Q by the resultant eigenvectors to obtain the same eigenvectors as direct eigendecomposition of QQ T .

The resulting eigenvectors {∅ i } represent the modes of shape variation, and the eigenvalues {λ i } are the corresponding variances of different modes. In this paper, we will frequently use the term ‘mode’ to represent the eigenvectors and use ‘mode i ’ to represent ∅ i . The shape variation modes are ordered by their variances (i.e. λ 1  ≥ λ 2 ··· ≥ λ n ) such that mode 1 corresponds to the largest variance, mode 2 corresponds to the second largest variance, and so on. The variance percentage ratio of mode i is computed as 100 % × λ i / ( ∑ j = 1 n λ j ) .

The SSM was represented as the mean shape vector plus the linear combinations of different shape variation modes (Heimann & Meinzer, 2009 ):

where X is a shape instance generated by the SSM represented as a shape vector ( x 1 , y 1 , z 1 , x 2 , y 2 , z 2 , …, x k , y k , z k ) T containing the 3D coordinates of k mesh vertices, X ¯ is the mean shape vector of all training subjects, {∅ i } is the shape variation modes obt a ined via PCA based on the shape vectors of all training subjects, and { a i } is the shape coefficient serving as the weight of the variation modes. Different values of { a i } will result in different instances of torso anatomy.

Because the phantom shape is controlled by the real‐valued coefficients { a i }, it is possible to generate infinite numbers of shape instances of the population by adjusting the coefficient values. When the coefficient values are adjusted continuously, one can observe continuous deformation of the phantom. This is why the phantoms are called ‘deformable’. The phantom deformation can be achieved in real‐time, as the computation is trivial for modern personal desktop computers.

Personalized anatomy modelling

By adjusting the SSM shape coefficients { a i }, the phantom can be registered to the medial image (e.g. CT and MR) or body surface scan data of an individual person. The registered phantom provides the model of entire torso anatomy which is useful for personalized medical treatment, electromagnetic simulation, ergonomics device design, animation character creation, etc. To realize phantom registration, we used the point set registration approach to match the phantom vertices with the point set of an individual person. For the surface scan data, the individual point set was acquired by scanning the external body surface using a 3D surface scanner. For the medical images, the individual point set included the surface points of the high‐contrast organs, which were segmented using the semi‐automated method described in the Organ segmentation section. The marching cubes algorithm (Lorensen & Cline, 1987 ) was used to extract the organ surface points from the segmented regions.

The phantom registration included two steps: an initial rigid alignment of the SSM mean shape and a subsequent deformable registration of the SSM. The iterative closest point registration (ICP) strategy (Besl & McKay, 1992 ) was used in both steps. In each iteration of ICP, the closest target points to the phantom vertices were searched, and a spatial transform T was computed to bring the phantom vertices closer to the searched target points. For the initial alignment, T was a rigid transform composed of 3D translation and rotation. For the deformable registration, T included not only the rigid components but also an isotropic scaling factor and a nonlinear shape deformation controlled by the SSM shape coefficients { a i }. The optimal values of the rigid components, the scaling factor, and the shape coefficients { a i } were computed using the ASM approach (Heimann & Meinzer, 2009 ).

We conducted two types of experiments: a free deformation test and a personalized anatomy modelling test. The free deformation test deformed the phantoms by freely adjusting each shape coefficient { a i } to observe the anatomical meaning of each variation mode (i.e. each eigenvector of the SSM). The masses of the freely deformed phantom organs were compared with the reference organ masses of the Chinese population to verify whether the deforming organs remain in a plausible mass range. The personalized anatomy modelling test registered the phantoms with individual Chinese subjects to quantify the accuracy of personal anatomy modelling.

Free deformation test

To observe the anatomical variation of each mode, we separately adjusted each shape coefficient a i while keeping the other coefficients at zero. The value of each a i was adjusted in the range of [ − 3 λ i , 3 λ i ] , which is the plausible deformation range commonly used in the literature.

Figure  2 shows the anatomical variations corresponding to the two largest variation modes of both genders. The two largest modes were related to global‐scale changes of body height (Fig.  2 A) and fat quantity (Fig.  2 B). The body height changes corresponded to 34.3 and 31.0% of the total variation for males and females, respectively. The fat quantity changes corresponded to 15.0 and 13.3% of the total variation for males and females, respectively. Notably, mode 2 (i.e. eigenvector ∅ 2 ) revealed different fat accumulation patterns for different genders. The male phantom showed a change in abdomen girth, especially from the lateral view. In contrast, the female phantom showed a change in subcutaneous fat thickness. This finding coincides with the phenomenon that men are prone to store visceral fat, while women tend to accumulate subcutaneous fat (Lovejoy & Sainsbury, 2009 ; O'Sullivan, 2009 ). This gender‐specific fat accumulation pattern was not modelled by previous human phantoms.

Variation modes related to the changes of (A) body height and (B) fat amount. For each mode, the deformed phantoms corresponding to different ai values are rendered. In the top row, the phantoms are rendered with semi‐transparent skin and muscles to illustrate internal organs; In the bottom row, the phantoms are rendered with opaque skin to demonstrate outer body shape. The variance percentage ratio of each mode is marked besides the mode name.

Figure  3 demonstrates the variation modes for internal organs. As the phantoms of both genders have similar variation patterns, we only show the results of the male phantom. Figure  3 A demonstrates the internal organ deformation related to the respiratory process, including the change in lung volume, the rotation of ribs, and the motion of abdominal organs. These variations are reflected from mode 3 of DCHT‐M and mode 4 of DCHT‐F, corresponding to 7.7 and 6.5% of the total variation for each gender, respectively. We can see that the thoracic and abdominal organs all move in accordance with each other. During the inhalation process, the bottom of the lungs expands behind the back of the liver. Meanwhile, the liver and kidneys are pushed downwards. At first glance, it seems unreasonable that the SSM can learn respiratory motions, as all the training subjects were holding their breath during the CT acquisition. However, because different subjects were holding their breath at different levels, the SSM learned the variation in breath‐holding levels, resulting in a deformation pattern similar to respiratory motion.

Variation modes of (A) respiratory motion and (B) abdominal organ geometry.

Figure  3 B shows the variations in abdominal organ positions, orientations, and shapes. These variations are found in mode 7 of DCHT‐M and mode 4 of DCHT‐F, corresponding to 3.2 and 6.5% of the total variation in the male and female phantoms, respectively. It can be observed that when the left kidney moves downwards, the spleen also moves in the same direction. Similarly, when the liver becomes smaller, the right kidney moves upwards to stay close to the bottom of the liver. Such synchronized motions and deformations between adjacent abdominal organs are essential for precise modelling of personal abdominal anatomy.

Figure  4 displays the gender‐specific variations, including the changes in male muscle size and female breast size, which are 1.8 and 4.2% of the total variation for the male and female phantoms, respectively. Mode 10 of the male phantom reveals the muscle size variation (Fig.  4 A). We can observe the increase in the latissimus dorsi and pectoralis major as a 10 decreases from 3 λ 10 to − 3 λ 10 . For the female phantom, the variation in breast size is presented by mode 6 (Fig.  4 B). Notably, the growth in breast size and increments of the waist girth are correlated. This is a natural phenomenon, as female subjects tend to grow both breast size and subcutaneous fat in order to rear children. It should be mentioned that the size changes in male muscle and female breast are not very visually significant because the training set does not include particularly muscular males or big‐breasted females.

Gender‐specific variation of (A) male muscle size and (B) female breast size.

Summarizing the above observations, SSM successfully learned several principal modes reflecting realistic inter‐subject anatomical variations. There are still some modes without an obvious anatomical explanation. SSM is a mathematical method; without taking account of anatomical knowledge, it cannot guarantee that every mode has actual anatomical meaning. Nevertheless, for the purpose of personalized anatomical modelling, all modes contribute to the phantom registration whether or not they have apparent anatomical meaning.

To evaluate how well the organ masses of the DCHT‐M/F phantoms agree with large population statistics, we compared the organ masses of the DCHT‐M/F phantoms with the statistics of Chinese organ masses in the IAEA 1998 publication (Kawamura et al. 1998 ), which is based on over 20 000 Chinese subjects (age > 20 years) collected from 1950 to 1990. The z ‐score was used to measure the deviation of the phantom from the population mean value:

where x is the organ mass of the phantom, μ is the mean of the population, and σ is the standard deviation of the population. To estimate the phantom organ mass, we multiplied the phantom organ volume by the reference organ densities in the literature (ICRP, 2009 ). Because the organ volumes change with the value of a i , we varied a i of each mode within the range [ − 3 λ i , 3 λ i ] ( i  =   1,2,3) and computed the corresponding z ‐score ranges. The results are plotted in Fig.  5 . We plotted only the first three variation modes, which accounted for 50.0 and 57.3% of the total variation for males and females, respectively. The other modes have much narrower z ‐score ranges than the first three modes and therefore are not plotted in order to retain figure clarity.

The z ‐scores comparing the phantom organ masses with the reference values from a large population survey ( IAEA 1998). Different variation modes of the phantoms are plotted with different colours. The circles mark the z ‐scores of a i   =   0 (i.e. the mean shape of the phantom). The error bars mark the z ‐score ranges for a i ∈ [ − 3 λ i , 3 λ i ] . The z ‐score levels of 1.96 and −1.96 (which correspond to P ‐value 0.05) are plotted as horizontal lines.

In this test, a normal distribution of population statistics was assumed. Therefore, if the z ‐score value falls out of the range [−1.96, 1.96], the probability that the phantom organ mass belongs to the population distribution is <5% ( P  <   0.05). As shown in Fig.  5 , the mean shapes ( a i  = 0) of all of the soft organs and bone parts had z ‐scores within the range of −1.96 to 1.96 (i.e. P  >   0.05), and more than half of them fell between −1 and 1 (corresponding to P  >   0.32). When the shape coefficients varied between [ − 3 λ i , 3 λ i ] , the z ‐score ranges of most organs remained within −1.96 to 1.96. These results imply a reasonable agreement between the phantom organ masses and the population statistics data. It can also be observed that the phantoms slightly overestimated the masses of the kidneys, sternum, and sacrum, and underestimated the masses of the heart, pancreas, and clavicles. One possible reason for this difference is that the phantom training set is much smaller than the large population samples, and potential bias may exist in our small sample set. If more training data are used, less bias can be expected.

Personalized anatomy modelling test

We tested the ability of the phantoms to model individual anatomy by deforming them to match personal body surface scans or torso CT images of different individual subjects. Figure  6 A illustrates the phantom registration results with the surface point clouds of a male and a female subject. The male surface data were acquired for one of the authors (aged 25 years) using a hand‐held 3D surface scanner to simulate the application of animation character design from the real‐subject surface scan. The female surface data were created using the human zbuilder software (Human‐zBuilder, 2017 ) to mimic an old woman in an intra‐operative surgery situation. As shown in Fig.  6 A, despite the interference of the shirt on the male surface and the loose skin of the female surface, the registered phantoms were properly matched to both subjects, giving an estimation of internal organ structures. Because the surface scan data do not include internal organs, it is impossible to quantify the accuracy of organ estimation. The surface data registration experiment only proves the feasibility of internal organ estimation based on outer body shape. Quantitative assessment of internal organ accuracy is given by the following CT‐based registration experiment.

Personalized anatomy modelling results. (A) Phantom registration with body surface scan. (B) Phantom registration with CT images. The results are displayed as organ contours overlaid onto representative sagittal and coronal slices of the target CT images. The top and bottom rows show the results for male and female subjects, respectively. The left and right columns show thin and fat subjects, respectively. (C) Box plots of organ registration accuracy of the 22 test CT images as measured by Dice, RC vlm and ASD , respectively.

For the CT‐based experiment, the phantoms were registered to 22 CT images of healthy subjects not included in the training images, including 13 males (ages between 31 and 88 years and body weights between 51 and 93 kg) and nine females (ages between 35 and 77 years and body weights between 41 and 60 kg). The acquisition protocol of the test CT images is the same as that described in section Data collection above. To provide ground truth organ regions, major trunk organs (including skin, whole skeleton, lungs, whole heart, liver, spleen, and kidneys) were segmented from the torso CT images by a human expert, and the surface meshes of these organs were extracted using the marching cubes method. The purpose of this experiment was to evaluate the performance of the phantoms for modelling complete torso anatomy given the boundaries of a few high‐contrast organs.

Figure  6 B shows the registration results for four representative test subjects of different genders and body fat quantities. The registered phantoms are displayed as coloured contours overlaid on the target CT images. Different coronal and sagittal sections of the torso are displayed to give complete observation of the registration results. For most major organs, the registered phantoms were well‐matched, with minor inconsistency of the organ edges. For quantitative evaluation of the CT‐registration results, the registration accuracy was measured via three metrics: the Dice coefficient (Dice), recovery coefficient of organ volume (RC vlm ), and averaged surface distance (ASD):

where the Dice coefficient measures the overlapping ratio between the registered phantom organ region (R P ) and the target subject organ region (R I ); |·| is the region volume; ∩ denotes the overlapping parts of two regions; RC vlm measures the ratio of the registered phantom organ volume over the target subject organ volume; ASD measures the averaged surface distance between the surface of registered phantom organ and the surface of target subject organ; n P and n T stand for the number of surface mesh vertices in the phantom organ and test subject organ, respectively; d i P → T is the minimum distance from the i th vertex of the phantom organ to the surface of the target subject organ; and d j T → P is the minimum distance from j th vertex of the target subject organ to the surface of the phantom organ.

Figure  6 C reports the results of Dice, RC vlm , and ASD for the 22 test CT images. Because the skeleton and skin of the target subjects include limbs, whereas the phantom does not, it is unreasonable to evaluate Dice and RC vlm for the skeleton and body region. Therefore, the skeleton and body region (skin) only show the ASD results. Figure  6 C shows that most organs have a median Dice > 0.9, a median RC vlm between 0.85 and 1.5, and a median ASD below 1.5 mm. The heart and kidneys have RC vlm values almost equal to 1, indicating good volume estimations for these organs. The ASDs of all organs are between 1 and 1.5 mm, which is roughly the pixel size of the training images for constructing the phantoms, meaning that the accuracy of anatomical modelling is limited by the spatial resolution of the training images.

Phantom construction

To collect enough training images of normal subjects, we chose the health‐screening PET/CT images as the training data. Due to the low X‐ray dose used for PET/CT acquisition, only major trunk organs and bones could be segmented from the CT images. Ideally, we should use medical images with better soft‐tissue contrast, such as diagnostic CT, contrast‐enhanced CT or MRI images. However, these images are generally acquired for local diseased organs and we are trying to collect enough diagnostic CT images covering the entire torso with minimal gross organ defects. To compensate for the missing organs, we morphed a 3D template of complete anatomy to each training subject using the already segmented organs as the matching target. Similarly, Segars et al. ( 2013 ) also morphed their XCAT phantoms to CT images to create a phantom library. Such morphing takes advantage of the anatomical dependency between the segmented and unsegmented organs, yielding a reasonable estimation of the missing anatomy. It should be noted that Segars et al. used a smooth invertible transformation (i.e. the multichannel large deformation diffeomorphic mapping), whereas we used the thin‐plate‐spline (TPS) transform to map the unsegmented organs. Although the purposes were similar, Segars's method produced smoother organ deformation but was much slower than ours (6–8 h vs. a few minutes for each training subject). Nevertheless, we will consider learning from Segars et al. and use diffeomorphic transforms in future research because better phantom quality deserves the additional construction time.

To construct the phantoms, two requisite steps are the semi‐automatic organ segmentation and the marker‐based template registration. These two steps are the most time‐consuming and labour‐intensive parts of the workflow. We used a semi‐automated approach because it allows the radiologist to proofread and correct the segmentation results. To speed up the processing of more training data, an automatic organ segmentation method with less labour‐intensive proofreading must be developed. We can learn from recent methods (Wang et al. 2016 ).

By registering the deformable phantoms with individual test subjects, we demonstrate the feasibility of using the deformable phantom to model individual torso anatomy based on body surface data or torso CT images. Compared with many existing organ segmentation methods, a benefit of deformable phantom registration is the estimation of all torso organs, including those that cannot be segmented from CT images, e.g. cardiac chambers, skeletal muscles, and renal vessels. This feature is important for simulation applications that do not require accurate organ segmentation but demand good estimation of organ volume and comprehensive modelling of all internal structures (Divoli et al. 2009 ). Moreover, the results of deformable phantom registration also provide close initialization for subsequent organ segmentation, which is not our current research scope but will be investigated in a future study.

The CT‐registration experiment gives quantitative results of each organ's registration accuracy. The heart seems to be the most accurate organ, with median values for Dice, RC vlm , and ASD of 0.92, 1.06, and 1.16 mm, respectively. The heart also has compact distributions of the three metrics, meaning that the registration of the heart is stably accurate for all test subjects. The high and stable accuracy of the heart is attributed to its spherical shape and stable position. Spherical shapes are easy to overlap, and a stable anatomical position ensures consistent registration accuracy. The kidneys also have high median Dice values (0.90 and 0.91 for the left and right kidneys, respectively), low median ASD values (1.08 and 1.05 mm for the left and right kidneys, respectively), and median RC vlm values close to 1 (1.01 and 0.99 for the left and right kidneys, respectively) thanks to the spherical shape of the kidneys. However, the Dice and ASD distributions of the kidneys are not quite compact. This is because the positions of kidneys are rather variable (see Fig.  3 B), making the registration accuracy less stable for different subjects. The spleen has a curved shape that is difficult to overlap completely; therefore, the median Dice (0.84) and ASD (1.27 mm) values were relatively lower. The anatomical position of the spleen is also quite variable (see Fig.  3 B), leading to non‐compact accuracy distributions. The lungs and liver are large organs that are easy to overlap, resulting in moderate median Dice (0.87 and 0.88, respectively) and RC vlm (0.90 and 0.91, respectively) values. However, due to their large sizes, the shape variations always occur on a large scale, resulting in relatively large ASDs (1.37 and 1.39 mm, respectively).

This study did not validate the registration accuracy of type II organs (e.g. the pancreas and gallbladder) because the radiologist could not generate reliable ground truth segmentation for these organs from the low‐dose CT images. We will keep collecting enough high‐contrast diagnostic CT images to validate these organs in future studies. However, for the small organs such as the pancreas and gallbladder, we do not expect as good a registration accuracy as for the large organs (e.g. the heart, liver, and lungs), as their shapes and locations vary noticeably even for the same person at different time points. Our objective is to obtain a reasonable estimation of their volumes.

A limitation of the current phantoms is that the SSM shape coefficients { a i } do not correspond to intuitive anthropometry parameters such as body height, weight, and breast size. It is not feasible manually to adjust the phantom shape according to desired anthropometrical values. To solve this problem, we tried to use the method of Brett et al. ( 2003 ), which recombines SSM modes of human body surfaces to generate deformation modes related to body weight and heights. However, this method was not effective because our model includes many more organs than a single body surface. We still need to develop a more effective method to correlate the phantom deformation with intuitive anthropometrical changes. As different medical applications require different anatomical parameters, it might be appropriate to correlate the phantom deformation with application‐specific parameters, such as bone length for orthopaedics research or lung volume for respiratory simulations.

Comparison with existing deformable human phantoms

As introduced at the beginning of this paper, there are several existing deformable human phantoms incorporating intra‐ and inter‐subject anatomical variations, including the 4D NCAT torso phantom, the 4D XCAT whole‐body phantom, and the RPI deformable adult phantoms. To date, all the existing deformable phantoms have been constructed for Caucasians; we have modelled the anatomical variations of Chinese subjects for the first time. A common feature of the DCHT‐M/F phantoms and the existing deformable phantoms is the use of BREP. The RPI and DCHT‐M/F phantoms both use triangular meshes, whereas the 4D XCAT phantom uses NURBS for large organs (e.g. the liver) and sub‐division (SD) meshes for fine‐detail structures (e.g. the cerebral cortex). Compared with the triangular meshes, NURBS is more efficient for modelling large organs, while the SD mesh is smoother for representing small‐scale structures. For future improvement of the DCHT‐M/F phantoms, we will consider learning from the XCAT phantoms to use the combination of NURBS and SD meshes.

Regarding the modelling of anatomical variations, the RPI deformable phantom adapts organ volumes according to reference anthropometry values from the publications of ICRP‐89 (ICRP, 2002 ) and NHANES (McDowell et al. 2005 ); it does not include realistic variation in organ shapes. The XCAT phantoms incorporate actual‐subject respiratory and cardiac motions from gated CT images, but these motions are only for local organs rather than the whole torso range. The XCAT phantom also allows the user to adapt anatomical parameters of body height, chest measurements, diaphragm position, heart size position, and orientation to simulate inter‐subject variations, but these variations are not learned from real patient data. Our DCHT‐M/F phantoms use the SSM approach to determine realistic inter‐subject anatomical variations, including but not restricted to the changes in body height, weight, respiratory status, internal organ geometry, torso posture, male muscle size, and female breast size. No existing deformable phantom has incorporated all these variations. The training set of the DCHT‐M/F phantoms is also larger than any existing deformable phantom or phantom library. The DCHT‐M/F phantoms do not include a beating heart and therefore cannot be used to simulate dynamic cardiac imaging like the XCAT phantoms.

In terms of the modelled body range, both the XCAT and RPI deformable phantoms are for the entire body, whereas the DCHT‐M/F phantoms are for the torso. Just like the extension from NCAT to XCAT, we also plan to extend the current torso phantoms to the whole‐body range. We may learn from the strategy used by Segars et al. ( 2013 ) for extending the XCAT phantom to a phantom library. They added pseudo limbs and a head to the patient CT data and morphed the XCAT phantoms to match each individual patient. To do so, anthropometry data for Chinese limbs and heads must be obtained first.

In this study, deformable phantoms of Chinese adult males and females were constructed based on 79 segmented CT images of normal subjects. The SSM approach was used to learn anatomically meaningful inter‐subject variation. The evaluation results demonstrate the capability of phantoms to model personalized anatomy. Future improvements of the phantom can be achieved by using more training subjects or more advanced shape modelling methods. In particular, as we are still collecting more PET/CT data, our future studies could explore sub‐population modelling for different ages or country areas, which will hopefully lead to more meaningful results. To achieve these goals, automated methods for image segmentation and template registration must be specifically developed for low‐dose CT images to tackle the heavy burden of data processing.

Acknowledgements

The authors sincerely thank Dr Dongmei Guo for her careful proofreading of the CT segmentation results. This study was supported by the youth programme of the National Natural Science Fund of China (No. 81401475), the general programme of the National Natural Science Fund of China (No. 61571076, 81171405, 61371187, 61671158 and 81671771), the National Science and Technology Major Project (pre‐approve No. SQ2018ZX100301), the general programme of Liaoning Science & Technology Project (No. 2015020040), the cultivating programme of the Major National Natural Science Fund of China (No. 91546123), the National Key Research and Development Program (No. 2016YFC0103101, 2016YFC0103102, 2016YFC0106402, 2016YFC0106403), the Science and Technology Star Project Fund of Dalian City (No. 2016RQ019), and the Basic Research Funding and Xinghai Scholar Cultivating Funding of Dalian University of Technology (No. DUT14RC(3)066 and DUT15LN02). We declare that we have no conflicts of interest.

• Ackerman MJ (1998) The visible human project . Proc IEEE 86 , 504–511. [ Google Scholar ]
• Ali‐Hamadi D, Liu T, Gilles B, et al. (2013) Anatomy transfer . ACM Trans Graph 32 , 1–8. [ Google Scholar ]
• Alziar I, Bonniaud GD (2009) Individual radiation therapy patient whole‐body phantoms for peripheral dose evaluations: method and specific software . Phys Med Biol 54 , 375–383. [ PubMed ] [ Google Scholar ]
• Besl PJ, McKay H (1992) A method for registration of 3‐D shapes . IEEE Trans Pattern Anal Mach Intell 14 , 239–256. [ Google Scholar ]
• Bijar A, Rohan PY, Perrier P, et al. (2016) Atlas‐based automatic generation of subject‐specific finite element tongue meshes . Ann Biomed Eng 44 , 16–34. [ PubMed ] [ Google Scholar ]
• Blum T, Kleeberger V, Bichlmeier C, et al. (2012) mirracle: an augmented reality magic mirror system for anatomy education . Proc IEEE Virt Real Ann PP , 115–116. [ Google Scholar ]
• Bookstein FL (1996) Landmark methods for forms without landmarks: localizing group differences in outline shape . MMBIA PP , 279–289. [ PubMed ] [ Google Scholar ]
• Boykov Y, Jolly M‐P (2000) Interactive organ segmentation using graph cuts . MICCAI 1935 , 276–286. [ Google Scholar ]
• Brett A, Brian C, Zoran P (2003) The space of human body shapes: reconstruction and parameterization from range scans . ACM Trans Graph 22 , 587–594. [ Google Scholar ]
• Broggio D, Beurrier J, Bremaud M, et al. (2011) Construction of an extended library of adult male 3D models: rationale and results . Phys Med Biol 56 , 7659–7692. [ PubMed ] [ Google Scholar ]
• Cassola VF, Lima VJ, Kramer R, et al. (2010) FASH and MASH: female and male adult human phantoms based on polygon mesh surfaces: I. Development of the anatomy . Phys Med Biol 55 , 133–162. [ PubMed ] [ Google Scholar ]
• Cassola VF, Milian FM, Kramer R, et al. (2011) Standing adult human phantoms based on 10th, 50th and 90th mass and height percentiles of male and female Caucasian populations . Phys Med Biol 56 , 3749–3772. [ PubMed ] [ Google Scholar ]
• Christ A, Kainz W, Hahn EG, et al. (2010) The Virtual Family – development of surface‐based anatomical models of two adults and two children for dosimetric simulations . Phys Med Biol 55 , N23–N38. [ PubMed ] [ Google Scholar ]
• Chui HL, Rangarajan A (2003) A new point matching algorithm for non‐rigid registration . J Comput Vis Image Underst 89 , 114–141. [ Google Scholar ]
• Conil E, Hadjem A, Lacroux F, et al. (2008) Variability analysis of SAR from 20 MHz to 2.4 GHz for different adult and child models using finite‐difference time‐domain . Phys Med Biol 53 , 1511–1525. [ PubMed ] [ Google Scholar ]
• Ding A, Mille MM, Liu T, et al. (2012) Extension of RPI‐adult male and female computational phantoms to obese patients and a Monte Carlo study of the effect on CT imaging dose . Phys Med Biol 57 , 2441–2459. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Divoli A, Chiavassa S, Ferrer L, et al. (2009) Effect of patient morphology on dosimetric calculations for internal irradiation as assessed by comparisons of Monte Carlo versus conventional methodologies . J Nucl Med 50 , 316–323. [ PubMed ] [ Google Scholar ]
• Farah J, Broggio D, Franck D (2010) Female workers and in vivo lung monitoring: a simple model for morphological dependence of counting efficiency curves . Phys Med Biol 55 , 7377–7395. [ PubMed ] [ Google Scholar ]
• Geyer AM, O'Reilly S, Lee C, et al. (2014) The UF/NCI family of hybrid computational phantoms representing the current US population of male and female children, adolescents, and adults – application to CT dosimetry . Phys Med Biol 59 , 5225–5242. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Gibbs SJ, Pujol A Jr, Chen TS, et al. (1984) Patient risk from interproximal radiography . Oral Surg Oral Med Oral Pathol 58 , 347–354. [ PubMed ] [ Google Scholar ]
• Gosselin MC, Neufeld E, Moser H, et al. (2014) Development of a new generation of high‐resolution anatomical models for medical device evaluation: the Virtual Population 3.0 . Phys Med Biol 59 , 5287–5303. [ PubMed ] [ Google Scholar ]
• Hegenbart L, Na YH, Zhang JY, et al. (2008) A Monte Carlo study of lung counting efficiency for female workers of different breast sizes using deformable phantoms . Phys Med Biol 53 , 5527–5538. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Heimann T, Meinzer HP (2009) Statistical shape models for 3D medical image segmentation: a review . Med Image Anal 13 , 543–563. [ PubMed ] [ Google Scholar ]
• Human‐zBuilder (2017) Human zBuilder. http://mattinian.com/ceco/ZbuilderZbrush.html (accessed October 2017).
• ICRP (1959) Report of Committee II on Permissible Dose for Internal Radiation . ICRP Publication 2.
• ICRP (2002) Basic Anatomical and Physiological Data for Use in Radiological Protection Reference Values . ICRP Publication 89. [ PubMed ]
• ICRP (2009) Adult Reference Computational Phantoms . ICRP Publication 110.
• Kawamura H, Tanaka G‐I, Shiraishi K, et al. (1998) Compilation of anatomical, physiological and metabolic characteristics for a Reference Asian Man‐IAEA‐TECDOC‐1005 . J Radiat Res 39 , 1–350. [ Google Scholar ]
• Kim CH, Jeong JH, Bolch WE, et al. (2011) A polygon‐surface reference Korean male phantom (PSRK‐Man) and its direct implementation in Geant4 Monte Carlo simulation . Phys Med Biol 56 , 3137–3161. [ PubMed ] [ Google Scholar ]
• Kramer R, Cassola VF, Khoury HJ, et al. (2010) FASH and MASH: female and male adult human phantoms based on polygon mesh surfaces: II. Dosimetric calculations . Phys Med Biol 55 , 163–189. [ PubMed ] [ Google Scholar ]
• Li X, Samei E, Segars WP, et al. (2008) Patient‐specific dose estimation for pediatric chest CT . Med Phys 35 , 5821–5828. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Li C, Kang Y, Wu T, et al. (2017) Numerical analysis for human perception of temperature rise on the fingertips during usage of a mobile device . Bioelectromagnetics 39 , 164–169. [ PubMed ] [ Google Scholar ]
• Lorensen WE, Cline HE (1987) Marching cubes: a high resolution 3D surface construction algorithm . Comput Graph 21 , 163–169. [ Google Scholar ]
• Lovejoy JC, Sainsbury A (2009) Sex differences in obesity and the regulation of energy homeostasis . Obes Rev 10 , 154–167. [ PubMed ] [ Google Scholar ]
• Marcatili S, Villoing D, Mauxion T, et al. (2015) Model‐based versus specific dosimetry in diagnostic context: comparison of three dosimetric approaches . Med Phys 42 , 1288–1296. [ PubMed ] [ Google Scholar ]
• McDowell MA, Fryar CD, Hirsch R, et al. (2005) Anthropometric reference data for children and adults: US population, 1999–2002 . Adv Data 361 , 1–5. [ PubMed ] [ Google Scholar ]
• Miller K (2016) Computational biomechanics for patient‐specific applications . Ann Biomed Eng 44 , 1–2. [ PubMed ] [ Google Scholar ]
• Mofrad FB, Zoroofi RA, Tehrani‐Fard AA, et al. (2010) Statistical construction of a Japanese male liver phantom for internal radionuclide dosimetry . Radiat Prot Dosimetry 141 , 140–148. [ PubMed ] [ Google Scholar ]
• Momennezhad M, Nasseri S, Zakavi SR, et al. (2016) A 3D Monte Carlo Method for estimation of patient‐specific internal organs absorbed dose for (99 m)Tc‐hynic‐Tyr(3)‐octreotide imaging . World J Nucl Med 15 , 114–123. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Na YH, Zhang B, Zhang J, et al. (2010) Deformable adult human phantoms for radiation protection dosimetry: anthropometric data representing size distributions of adult worker populations and software algorithms . Phys Med Biol 55 , 3789–3811. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• O'Sullivan AJ (2009) Does oestrogen allow women to store fat more efficiently? A biological advantage for fertility and gestation . Obes Rev 10 , 168–177. [ PubMed ] [ Google Scholar ]
• Park JS, Chung MS, Hwang SB, et al. (2006) Visible Korean Human: its techniques and applications . Clin Anat 19 , 216–224. [ PubMed ] [ Google Scholar ]
• Robinette KM, Blackwell S, Daanen H, et al. (2002) Civilian American and European Surface Anthropometry Resource (CAESAR) . Final Report , 1 .
• Sang HC, Jeong JH, Cho S, et al. (2008) Construction of a high‐quality voxel model VKH‐man using serially sectioned images from Visible Korean Human Project in Korea . J Nucl Sci Technol 45 , 179–182. [ Google Scholar ]
• Schiemann T, Freudenberg J, Pflesser B, et al. (2000) Exploring the visible human using the VOXEL‐MAN framework . Comput Med Imaging Graph 24 , 127–132. [ PubMed ] [ Google Scholar ]
• Segars WP, Tsui BM (2009) MCAT to XCAT: the evolution of 4‐D computerized phantoms for imaging research: computer models that take account of body movements promise to provide evaluation and improvement of medical imaging devices and technology . Proc IEEE Inst Electr Electron Eng 97 , 1954–1968. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Segars WP, Tsui BM, Lalush DS, et al. (2001) Development and application of the new dynamic Nurbs‐based Cardiac‐Torso (NCAT) phantom . J Nucl Med 42 , 23. [ Google Scholar ]
• Segars WP, Sturgeon G, Mendonca S, et al. (2010) 4D XCAT phantom for multimodality imaging research . Med Phys 37 , 4902–4915. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Segars WP, Bond J, Frush J, et al. (2013) Population of anatomically variable 4D XCAT adult phantoms for imaging research and optimization . Med Phys 40 , 043701–043711. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Stabin MG, Xu XG, Emmons MA, et al. (2012) RADAR reference adult, pediatric, and pregnant female phantom series for internal and external dosimetry . J Nucl Med 53 , 1807–1813. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Tong J (2016) Cancer screening of asymptomatic individuals using 18F‐FDG PET/CT in China: a retrospective study . Discov Med 22 , 181–188. [ PubMed ] [ Google Scholar ]
• Turbosquid (2017) Male and Female Anatomy Complete Pack (Textured). https://www.turbosquid.com/3d-models/male-female-anatomy-body-3d-max/602826 (last accessed October 2017).
• Wang H, Stout DB, Chatziioannou AF (2012) Estimation of mouse organ locations through registration of a statistical mouse atlas with micro‐CT images . IEEE Trans Med Imaging 31 , 88–102. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Wang H, Stout DB, Chatziioannou AF (2015) A deformable atlas of the laboratory mouse . Mol Imaging Biol 17 , 18–28. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Wang H, Udupa JK, Odhner D, et al. (2016) Automatic anatomy recognition in whole‐body PET/CT images . Med Phys 43 , 613–629. [ PubMed ] [ Google Scholar ]
• Williams G, Zankl M, Abmayr W, et al. (1986) The calculations of dose from external photon exposures using reference and realistic human phantoms and Monte Carlo methods . Phys Med Biol 31 , 449–452. [ PubMed ] [ Google Scholar ]
• Wittek A, Grosland NM, Joldes GR, et al. (2016) From finite element meshes to clouds of points: a review of methods for generation of computational biomechanics models for patient‐specific applications . Ann Biomed Eng 44 , 3–15. [ PubMed ] [ Google Scholar ]
• Wolf I, Vetter M, Wegner I, et al. (2005) The medical imaging interaction toolkit . Med Image Anal 9 , 594‐604. [ PubMed ] [ Google Scholar ]
• Wu T, Tan L, Shao Q, et al. (2011) Chinese adult anatomical models and the application in evaluation of RF exposures . Phys Med Biol 56 , 2075–2089. [ PubMed ] [ Google Scholar ]
• Xu XG (2014) An exponential growth of computational phantom research in radiation protection, imaging, and radiotherapy: a review of the fifty‐year history . Phys Med Biol 59 , R233–R302. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Xu XG, Chao TC, Bozkurt A (2000) VIP‐man: an image‐based whole‐body adult male model constructed from color photographs of the visible human project for multi‐particle Monte Carlo calculations . Health Phys 78 , 476–486. [ PubMed ] [ Google Scholar ]
• Zaidi H, Tsui BMW (2009) Review of computational anthropomorphic anatomical and physiological models . Proc IEEE 97 , 1938–1953. [ Google Scholar ]
• Zaidi H, Xu XG (2007) Computational anthropomorphic models of the human anatomy: the path to realistic Monte Carlo modeling in radiological sciences . Annu Rev Biomed Eng 9 , 471–500. [ PubMed ] [ Google Scholar ]
• Zhang SX, Heng PA, Liu ZJ, et al. (2004) The Chinese Visible Human (CVH) datasets incorporate technical and imaging advances on earlier digital humans . J Anat 204 , 165–173. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Zhang J, Na YH, Caracappa PF, et al. (2009) RPI‐AM and RPI‐AF, a pair of mesh‐based, size‐adjustable adult male and female computational phantoms using ICRP‐89 parameters and their calculations for organ doses from monoenergetic photon beams . Phys Med Biol 54 , 5885–5908. [ PMC free article ] [ PubMed ] [ Google Scholar ]