3D printing of radioactive wall-less PET phantoms improves threshold-based target delineation and quantification

Abstract Background Validation of threshold-based PET segmentation and PET quantification is typically performed with fillable phantoms. Theoretical considerations show that the inactive walls of the phantom cavities introduce a contrast dependence of the volume-reproducing threshold (VRT), potentia...

Full description

Saved in:
Bibliographic Details
Main Authors: Adrian Jun Zounek, Nico Maximilian Joerg, Felix Lindheimer, Artem Zatcepin, Giovanna Palumbo, Rosel Oos, Astrid Delker, Franz Josef Gildehaus, Andreas Bollenbacher, Guido Boening, Peter Bartenstein, Matthias Brendel, Nathalie Lisa Albert, Sibylle Ziegler, Lena Kaiser
Format: Article
Language:English
Published: SpringerOpen 2025-06-01
Series:EJNMMI Physics
Subjects:
Online Access:https://doi.org/10.1186/s40658-025-00768-x
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849725119699615744
author Adrian Jun Zounek
Nico Maximilian Joerg
Felix Lindheimer
Artem Zatcepin
Giovanna Palumbo
Rosel Oos
Astrid Delker
Franz Josef Gildehaus
Andreas Bollenbacher
Guido Boening
Peter Bartenstein
Matthias Brendel
Nathalie Lisa Albert
Sibylle Ziegler
Lena Kaiser
author_facet Adrian Jun Zounek
Nico Maximilian Joerg
Felix Lindheimer
Artem Zatcepin
Giovanna Palumbo
Rosel Oos
Astrid Delker
Franz Josef Gildehaus
Andreas Bollenbacher
Guido Boening
Peter Bartenstein
Matthias Brendel
Nathalie Lisa Albert
Sibylle Ziegler
Lena Kaiser
author_sort Adrian Jun Zounek
collection DOAJ
description Abstract Background Validation of threshold-based PET segmentation and PET quantification is typically performed with fillable phantoms. Theoretical considerations show that the inactive walls of the phantom cavities introduce a contrast dependence of the volume-reproducing threshold (VRT), potentially leading to segmentation errors and therefore miscalculations of target volumes. The goal of this study was to experimentally show the contrast independence of the VRT when using wall-less phantoms. Results Radioactive spheres were produced according to NEMA specifications (D = 10/13/17/22/28/37 mm) using a stereolithographic (SLA) 3D printer. For comparison, hollow spheres were filled with a similar activity concentration. Image data from both sphere types were acquired with five different signal-to-background ratios (SBR = 2/4/6/8/10) using a Siemens mCT 20 and a Biograph 64 TruePoint PET/CT system. Results from wall-less and fillable spheres were compared to evaluate contrast dependence and segmentation accuracy based on VRT and intensity profiles. Wall-less phantoms demonstrated consistent VRT values, with a coefficient of variation of 2% over all SBRs, indicating independence from contrast. Conversely, fillable phantoms exhibited significant VRT variability, with a coefficient of variation (CV) of 9% over all SBRs and up to 40% volume overestimation at low contrast. Additionally, activity distribution in the printed spheres was evaluated using PET-based statistical analysis and autoradiography. The PET intensity distribution in the printed material was highly uniform (CV = 4.2%), with a Kullback–Leibler divergence near zero and no statistically significant difference to the fillable spheres. Autoradiography revealed microscopic regions with elevated counts, showing a CV of 11.7%, which was effectively reduced to 2.4% after Gaussian filtering. Conclusions The theoretical predictions of a significant influence of inactive walls in low-contrast images and contrast-independent VRT in wall-less phantoms were successfully confirmed. SLA 3D printing of phantoms is a promising method for the reliable evaluation of PET quantification methods, particularly in low-contrast scenarios commonly encountered in clinical settings.
format Article
id doaj-art-6fceeda37ec14788b4e86aaa6fe37170
institution DOAJ
issn 2197-7364
language English
publishDate 2025-06-01
publisher SpringerOpen
record_format Article
series EJNMMI Physics
spelling doaj-art-6fceeda37ec14788b4e86aaa6fe371702025-08-20T03:10:32ZengSpringerOpenEJNMMI Physics2197-73642025-06-0112111210.1186/s40658-025-00768-x3D printing of radioactive wall-less PET phantoms improves threshold-based target delineation and quantificationAdrian Jun Zounek0Nico Maximilian Joerg1Felix Lindheimer2Artem Zatcepin3Giovanna Palumbo4Rosel Oos5Astrid Delker6Franz Josef Gildehaus7Andreas Bollenbacher8Guido Boening9Peter Bartenstein10Matthias Brendel11Nathalie Lisa Albert12Sibylle Ziegler13Lena Kaiser14Department of Nuclear Medicine, LMU University Hospital, LMU MunichDepartment of Nuclear Medicine, LMU University Hospital, LMU MunichDepartment of Nuclear Medicine, LMU University Hospital, LMU MunichDepartment of Nuclear Medicine, LMU University Hospital, LMU MunichDepartment of Nuclear Medicine, LMU University Hospital, LMU MunichDepartment of Nuclear Medicine, LMU University Hospital, LMU MunichDepartment of Nuclear Medicine, LMU University Hospital, LMU MunichDepartment of Nuclear Medicine, LMU University Hospital, LMU MunichDepartment of Nuclear Medicine, LMU University Hospital, LMU MunichDepartment of Nuclear Medicine, LMU University Hospital, LMU MunichDepartment of Nuclear Medicine, LMU University Hospital, LMU MunichDepartment of Nuclear Medicine, LMU University Hospital, LMU MunichDepartment of Nuclear Medicine, LMU University Hospital, LMU MunichDepartment of Nuclear Medicine, LMU University Hospital, LMU MunichDepartment of Nuclear Medicine, LMU University Hospital, LMU MunichAbstract Background Validation of threshold-based PET segmentation and PET quantification is typically performed with fillable phantoms. Theoretical considerations show that the inactive walls of the phantom cavities introduce a contrast dependence of the volume-reproducing threshold (VRT), potentially leading to segmentation errors and therefore miscalculations of target volumes. The goal of this study was to experimentally show the contrast independence of the VRT when using wall-less phantoms. Results Radioactive spheres were produced according to NEMA specifications (D = 10/13/17/22/28/37 mm) using a stereolithographic (SLA) 3D printer. For comparison, hollow spheres were filled with a similar activity concentration. Image data from both sphere types were acquired with five different signal-to-background ratios (SBR = 2/4/6/8/10) using a Siemens mCT 20 and a Biograph 64 TruePoint PET/CT system. Results from wall-less and fillable spheres were compared to evaluate contrast dependence and segmentation accuracy based on VRT and intensity profiles. Wall-less phantoms demonstrated consistent VRT values, with a coefficient of variation of 2% over all SBRs, indicating independence from contrast. Conversely, fillable phantoms exhibited significant VRT variability, with a coefficient of variation (CV) of 9% over all SBRs and up to 40% volume overestimation at low contrast. Additionally, activity distribution in the printed spheres was evaluated using PET-based statistical analysis and autoradiography. The PET intensity distribution in the printed material was highly uniform (CV = 4.2%), with a Kullback–Leibler divergence near zero and no statistically significant difference to the fillable spheres. Autoradiography revealed microscopic regions with elevated counts, showing a CV of 11.7%, which was effectively reduced to 2.4% after Gaussian filtering. Conclusions The theoretical predictions of a significant influence of inactive walls in low-contrast images and contrast-independent VRT in wall-less phantoms were successfully confirmed. SLA 3D printing of phantoms is a promising method for the reliable evaluation of PET quantification methods, particularly in low-contrast scenarios commonly encountered in clinical settings.https://doi.org/10.1186/s40658-025-00768-xSegmentationPhantoms3D printingPET
spellingShingle Adrian Jun Zounek
Nico Maximilian Joerg
Felix Lindheimer
Artem Zatcepin
Giovanna Palumbo
Rosel Oos
Astrid Delker
Franz Josef Gildehaus
Andreas Bollenbacher
Guido Boening
Peter Bartenstein
Matthias Brendel
Nathalie Lisa Albert
Sibylle Ziegler
Lena Kaiser
3D printing of radioactive wall-less PET phantoms improves threshold-based target delineation and quantification
EJNMMI Physics
Segmentation
Phantoms
3D printing
PET
title 3D printing of radioactive wall-less PET phantoms improves threshold-based target delineation and quantification
title_full 3D printing of radioactive wall-less PET phantoms improves threshold-based target delineation and quantification
title_fullStr 3D printing of radioactive wall-less PET phantoms improves threshold-based target delineation and quantification
title_full_unstemmed 3D printing of radioactive wall-less PET phantoms improves threshold-based target delineation and quantification
title_short 3D printing of radioactive wall-less PET phantoms improves threshold-based target delineation and quantification
title_sort 3d printing of radioactive wall less pet phantoms improves threshold based target delineation and quantification
topic Segmentation
Phantoms
3D printing
PET
url https://doi.org/10.1186/s40658-025-00768-x
work_keys_str_mv AT adrianjunzounek 3dprintingofradioactivewalllesspetphantomsimprovesthresholdbasedtargetdelineationandquantification
AT nicomaximilianjoerg 3dprintingofradioactivewalllesspetphantomsimprovesthresholdbasedtargetdelineationandquantification
AT felixlindheimer 3dprintingofradioactivewalllesspetphantomsimprovesthresholdbasedtargetdelineationandquantification
AT artemzatcepin 3dprintingofradioactivewalllesspetphantomsimprovesthresholdbasedtargetdelineationandquantification
AT giovannapalumbo 3dprintingofradioactivewalllesspetphantomsimprovesthresholdbasedtargetdelineationandquantification
AT roseloos 3dprintingofradioactivewalllesspetphantomsimprovesthresholdbasedtargetdelineationandquantification
AT astriddelker 3dprintingofradioactivewalllesspetphantomsimprovesthresholdbasedtargetdelineationandquantification
AT franzjosefgildehaus 3dprintingofradioactivewalllesspetphantomsimprovesthresholdbasedtargetdelineationandquantification
AT andreasbollenbacher 3dprintingofradioactivewalllesspetphantomsimprovesthresholdbasedtargetdelineationandquantification
AT guidoboening 3dprintingofradioactivewalllesspetphantomsimprovesthresholdbasedtargetdelineationandquantification
AT peterbartenstein 3dprintingofradioactivewalllesspetphantomsimprovesthresholdbasedtargetdelineationandquantification
AT matthiasbrendel 3dprintingofradioactivewalllesspetphantomsimprovesthresholdbasedtargetdelineationandquantification
AT nathalielisaalbert 3dprintingofradioactivewalllesspetphantomsimprovesthresholdbasedtargetdelineationandquantification
AT sibylleziegler 3dprintingofradioactivewalllesspetphantomsimprovesthresholdbasedtargetdelineationandquantification
AT lenakaiser 3dprintingofradioactivewalllesspetphantomsimprovesthresholdbasedtargetdelineationandquantification