Franz Wegner
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Franz Wegner has a total of 175 co-authors. Here are the top co-authors by number of shared publications:
- Jörg Barkhausen
Institute of Radiology and Nuclear Medicine University of Luebeck Lübeck Germany
15 shared publications - Malte Maria Sieren
Institute of Radiology and Nuclear Medicine University of Luebeck Lübeck Germany
12 shared publications Thomas Friedrich
Fraunhofer Research Institution for Individualized Medical Technology and Engineering IMTE Lübeck Germany
10 shared publications- Roman Kloeckner
Institute of Interventional Radiology University of Luebeck Lübeck Germany
8 shared publications Thorsten M. Buzug
Fraunhofer Research Institution for Individualized Medical Technology and Engineering IMTE Lübeck Germany
8 shared publications- Mandy Ahlborg
Fraunhofer IMTE, Fraunhofer Research Institution for Individualized Medical Technology and Engineering, Lübeck, Germany
6 shared publications
Latest Publications
- Open Access
- 30 October 2026
- Published by Wiley in Nano Select
- Vol. 7 (11) , e70198
Influencing Factors on the Degradation of PLA‐SPION Markers for Medical Devices in Magnetic Imaging Modalities—An In Vitro Study
Polylactic acid (PLA), with its degradable characteristics, is a versatile polymer used for various application scenarios. Coating medical instruments with a composite of PLA and superparamagnetic iron‐oxide nanoparticles (SPIONs) allows for their visualization in Magnetic Particle Imaging and Magnetic Resonance Imaging, which could enable functional diagnostics in medicine. This study investigates the influencing factors on the degradation of PLA‐SPION markers. The influence of the SPION mass fraction, surrounding medium/pH, carrier material, and initial coating mass on the hydrolytic degradation is monitored at 37°C for up to 4 weeks. It is shown that the addition of SPIONs generally decreases the degradation rate compared to neat PLA. However, within the composites, a higher SPION mass fraction increases the degradation rate. The markers degrade relatively quickly at a physiological pH of 7.4, while a more acidic environment reduces the degradation rate. A lighter initial coating mass also facilitates degradation, while the carrier material seems to have a marginal effect on the degradation rate. These results allow for a better understanding of the influencing factors on the degradation of PLA‐SPION markers and support the feasibility of functional imaging with these markers.
- 30 October 2026
- Published by Wiley in Nano Select
- Vol. 7 (11) , e70198
Open AccessInfluencing Factors on the Degradation of PLA‐SPION Markers for Medical Devices in Magnetic Imaging Modalities—An In Vitro Study
Polylactic acid (PLA), with its degradable characteristics, is a versatile polymer used for various application scenarios. Coating medical instruments with a composite of PLA and superparamagnetic iron‐oxide nanoparticles (SPIONs) allows for their visualization in Magnetic Particle Imaging and Magnetic Resonance Imaging, which could enable functional diagnostics in medicine. This study investigates the influencing factors on the degradation of PLA‐SPION markers. The influence of the SPION mass fraction, surrounding medium/pH, carrier material, and initial coating mass on the hydrolytic degradation is monitored at 37°C for up to 4 weeks. It is shown that the addition of SPIONs generally decreases the degradation rate compared to neat PLA. However, within the composites, a higher SPION mass fraction increases the degradation rate. The markers degrade relatively quickly at a physiological pH of 7.4, while a more acidic environment reduces the degradation rate. A lighter initial coating mass also facilitates degradation, while the carrier material seems to have a marginal effect on the degradation rate. These results allow for a better understanding of the influencing factors on the degradation of PLA‐SPION markers and support the feasibility of functional imaging with these markers.
- Open Access
- 27 August 2026
- Published by Springer Nature in CVIR Endovascular
- Vol. 9 (1) , 108
Introduction and validation of OSCAR—optimal stent choice algorithm
Purpose: Standardization and international guidelines for stent size selection are lacking. In this study, we introduce and validate an artificial intelligence (AI)- and image processing-supported, modular software algorithm trained on a multicentric vascular segmentation dataset that identifies stenoses, performs segmentations of stenotic vessel segments and suggests the optimal stent size for implantation. Material and methods: This retrospective multicenter study included 149 patients who underwent stent implantation for symptomatic stenoses of the common and external iliac arteries between August 2017 and July 2024. Peri-interventional angiography datasets were evaluated by four board-certified interventional radiologists. For AI-training, all relevant stenoses were annotated and segmented to reflect intended stent sizing. The segmentation criteria were consensus-defined, and all readers completed a prior training session to ensure consistency. The modular algorithm comprises components for stenosis detection, segmentation and stent parameter prediction. Following pre-training on a publicly available coronary artery dataset, the model was fine-tuned on the study-specific iliac artery dataset using leave-one-out cross-validation. Results: OSCAR detected stenoses in 84.6% of cases. The model achieved a high recall (0.89 ± 0.21), meaning that most expert-annotated stenoses were correctly identified, while a moderate precision (0.65 ± 0.28) indicated some false-positive detections. Segmentation accuracy was good (DSC 0.77 ± 0.11). Stent diameter and length predictions demonstrated mean absolute percentage errors of 0.13 ± 0.18 and 0.33 ± 0.31, respectively, comparable to expert variability. Conclusions: This proof-of-concept study demonstrates the potential of AI-assisted stent selection in vascular interventions. Furthermore, the option of a closed-loop framework promotes sustainability, reproducibility and cost-effectiveness in stent implantation procedures.
- 27 August 2026
- Published by Springer Nature in CVIR Endovascular
- Vol. 9 (1) , 108
Open AccessIntroduction and validation of OSCAR—optimal stent choice algorithm
Purpose: Standardization and international guidelines for stent size selection are lacking. In this study, we introduce and validate an artificial intelligence (AI)- and image processing-supported, modular software algorithm trained on a multicentric vascular segmentation dataset that identifies stenoses, performs segmentations of stenotic vessel segments and suggests the optimal stent size for implantation. Material and methods: This retrospective multicenter study included 149 patients who underwent stent implantation for symptomatic stenoses of the common and external iliac arteries between August 2017 and July 2024. Peri-interventional angiography datasets were evaluated by four board-certified interventional radiologists. For AI-training, all relevant stenoses were annotated and segmented to reflect intended stent sizing. The segmentation criteria were consensus-defined, and all readers completed a prior training session to ensure consistency. The modular algorithm comprises components for stenosis detection, segmentation and stent parameter prediction. Following pre-training on a publicly available coronary artery dataset, the model was fine-tuned on the study-specific iliac artery dataset using leave-one-out cross-validation. Results: OSCAR detected stenoses in 84.6% of cases. The model achieved a high recall (0.89 ± 0.21), meaning that most expert-annotated stenoses were correctly identified, while a moderate precision (0.65 ± 0.28) indicated some false-positive detections. Segmentation accuracy was good (DSC 0.77 ± 0.11). Stent diameter and length predictions demonstrated mean absolute percentage errors of 0.13 ± 0.18 and 0.33 ± 0.31, respectively, comparable to expert variability. Conclusions: This proof-of-concept study demonstrates the potential of AI-assisted stent selection in vascular interventions. Furthermore, the option of a closed-loop framework promotes sustainability, reproducibility and cost-effectiveness in stent implantation procedures.
- Open Access
- 13 August 2026
- Published by Springer Nature in European Radiology Experimental
- Vol. 10 (1) , 114
3-T MRI in vitro characterisation of a SPION-polymer balloon catheter: a proof-of-concept study
Objective: Magnetic resonance imaging (MRI)-guided endovascular interventions represent a radiation-free alternative to x-ray fluoroscopy but remain limited by the lack of compatible and visible instruments. This study aimed to characterise a balloon catheter with polymer-embedded superparamagnetic iron oxide nanoparticles (SPIONs) as a passive MR-visible marker. Materials and methods: The SPION balloons consisted of a polymer with integrated iron oxide particles. A custom-built phantom study was conducted on a clinical 3-T MRI scanner using a balanced steady-state free precession sequence. Parameters influencing the balloons’ imaging properties were evaluated: SPION content (5 weight percent (wt%), 20 wt% and 30 wt%), flip angle (10° to 60°), the balloons’ spatial orientation relative to the main magnetic field B 0 , the inflation state and phase-encoding direction across different spatial orientations. The SPION-induced susceptibility artefacts were quantified from segmentations independently performed by two board-certified radiologists. Results: All SPION-polymer balloons were discernible at 3-T MRI across all tested SPION concentrations. No linear correlation was observed between SPION concentration and artefact dimension, with the intermediate concentration of 20 wt% yielding the smallest artefact size. The balloons’ artefact dimensions increased with higher angulations relative to B 0 . Balloons were visualised in the inflated and deflated configurations. The flip angle had a limited impact on artefact dimensions, while higher flip angles resulted in increased SNR. The influence of phase-encoding direction was generally minor with a single exception (90° coronal plane, AP to RL). Conclusion: The SPION-polymer balloons were effectively visualised at 3-T MRI irrespective of orientation and inflation. This in vitro study outlines a proof-of-concept supporting future clinical use of polymer-integrated SPIONs for passive device visualisation in interventional MRI. Relevance statement: SPION-polymer balloons can be visualised in 3-T MRI, providing a foundation for future MRI-visible interventional instrument designs. Key Points: Visualisation properties of a SPION-polymer balloon catheter were systematically assessed at 3-T MRI to overcome the shortage of MRI-visible instruments. SPION-polymer balloons were effectively visualised across different spatial orientations in the inflated and deflated state. Polymer-integrated SPIONs demonstrate potential as passive instrument markers for MRI-guided interventions.
- 13 August 2026
- Published by Springer Nature in European Radiology Experimental
- Vol. 10 (1) , 114
Open Access3-T MRI in vitro characterisation of a SPION-polymer balloon catheter: a proof-of-concept study
Objective: Magnetic resonance imaging (MRI)-guided endovascular interventions represent a radiation-free alternative to x-ray fluoroscopy but remain limited by the lack of compatible and visible instruments. This study aimed to characterise a balloon catheter with polymer-embedded superparamagnetic iron oxide nanoparticles (SPIONs) as a passive MR-visible marker. Materials and methods: The SPION balloons consisted of a polymer with integrated iron oxide particles. A custom-built phantom study was conducted on a clinical 3-T MRI scanner using a balanced steady-state free precession sequence. Parameters influencing the balloons’ imaging properties were evaluated: SPION content (5 weight percent (wt%), 20 wt% and 30 wt%), flip angle (10° to 60°), the balloons’ spatial orientation relative to the main magnetic field B 0 , the inflation state and phase-encoding direction across different spatial orientations. The SPION-induced susceptibility artefacts were quantified from segmentations independently performed by two board-certified radiologists. Results: All SPION-polymer balloons were discernible at 3-T MRI across all tested SPION concentrations. No linear correlation was observed between SPION concentration and artefact dimension, with the intermediate concentration of 20 wt% yielding the smallest artefact size. The balloons’ artefact dimensions increased with higher angulations relative to B 0 . Balloons were visualised in the inflated and deflated configurations. The flip angle had a limited impact on artefact dimensions, while higher flip angles resulted in increased SNR. The influence of phase-encoding direction was generally minor with a single exception (90° coronal plane, AP to RL). Conclusion: The SPION-polymer balloons were effectively visualised at 3-T MRI irrespective of orientation and inflation. This in vitro study outlines a proof-of-concept supporting future clinical use of polymer-integrated SPIONs for passive device visualisation in interventional MRI. Relevance statement: SPION-polymer balloons can be visualised in 3-T MRI, providing a foundation for future MRI-visible interventional instrument designs. Key Points: Visualisation properties of a SPION-polymer balloon catheter were systematically assessed at 3-T MRI to overcome the shortage of MRI-visible instruments. SPION-polymer balloons were effectively visualised across different spatial orientations in the inflated and deflated state. Polymer-integrated SPIONs demonstrate potential as passive instrument markers for MRI-guided interventions.
- Open Access
- 6 June 2026
- Published by Springer Nature in CVIR Endovascular
- Vol. 9 (1)
Differences between manufacturer-specified and measured effective inner diameters of vascular introducer sheaths: a micro-CT analysis
Purpose: To quantify deviations between the inner diameters specified by manufacturers and the effective diameter of vascular introducer sheaths. Materials and methods: A total of 18 different introducer sheaths (one device per type) with nominal inner diameters (nID) of 5 French (Fr), 6 Fr, and 7 Fr and lengths of 10 cm, 25 cm, and 45 cm, manufactured by four different vendors (Terumo, Biotronik, Merit, and Cook Medical) were evaluated. Non-destructive testing was performed using a commercial Micro-CT system (Comet Yxlon) at three distinct anatomical locations: the hub, middle, and tip/aperture. Cylindrical and circular least-squares fits were applied to calculate the effective inner diameters (eID). The eIDs were compared with the inner diameters specified by the manufacturer (mID) and the nominal inner diameter (nID) in Fr, which both are regularly mentioned for each product. Results: All non-braided sheaths revealed eIDs at the hubs, middles, and tips exceeding the mIDs (up to + 8.5%). All eIDs at the sheath aperture were consistently smaller than the mIDs and, with one exception, smaller than the nID. For braided sheaths, most eIDs at the hub were up to 3.4% larger than specified. However, at the tips and the apertures all braided sheaths were smaller than the mID (up to − 10.99%). In comparison with the nID, most braided sheaths were larger. Conclusion: Relevant deviations exist between the effective and manufacturer-specified inner diameters of vascular sheaths, especially at the tips. These discrepancies may necessitate the selection of larger sheath sizes to accommodate specific interventional devices, particularly when aperture constraints are critical.
- 6 June 2026
- Published by Springer Nature in CVIR Endovascular
- Vol. 9 (1)
Open AccessDifferences between manufacturer-specified and measured effective inner diameters of vascular introducer sheaths: a micro-CT analysis
Purpose: To quantify deviations between the inner diameters specified by manufacturers and the effective diameter of vascular introducer sheaths. Materials and methods: A total of 18 different introducer sheaths (one device per type) with nominal inner diameters (nID) of 5 French (Fr), 6 Fr, and 7 Fr and lengths of 10 cm, 25 cm, and 45 cm, manufactured by four different vendors (Terumo, Biotronik, Merit, and Cook Medical) were evaluated. Non-destructive testing was performed using a commercial Micro-CT system (Comet Yxlon) at three distinct anatomical locations: the hub, middle, and tip/aperture. Cylindrical and circular least-squares fits were applied to calculate the effective inner diameters (eID). The eIDs were compared with the inner diameters specified by the manufacturer (mID) and the nominal inner diameter (nID) in Fr, which both are regularly mentioned for each product. Results: All non-braided sheaths revealed eIDs at the hubs, middles, and tips exceeding the mIDs (up to + 8.5%). All eIDs at the sheath aperture were consistently smaller than the mIDs and, with one exception, smaller than the nID. For braided sheaths, most eIDs at the hub were up to 3.4% larger than specified. However, at the tips and the apertures all braided sheaths were smaller than the mID (up to − 10.99%). In comparison with the nID, most braided sheaths were larger. Conclusion: Relevant deviations exist between the effective and manufacturer-specified inner diameters of vascular sheaths, especially at the tips. These discrepancies may necessitate the selection of larger sheath sizes to accommodate specific interventional devices, particularly when aperture constraints are critical.
- Open Access
- 1 February 2026
- Published by Taylor & Francis in International Journal of Nanomedicine
- Vol. ume 21, 1-12
Continuous Coagulation Monitoring in Human Blood Samples via Magnetic Particle Spectroscopy
Introduction: Magnetic Particle Imaging (MPI) is a radiation-free imaging modality based on the nonlinear magnetic response of iron oxide nanoparticles, providing high sensitivity and real-time, quantitative, background-free imaging. With the clinical approval of Resotran as an MPI-suitable tracer and the development of first human-scale scanners, clinical applications are within reach. Magnetic Particle Spectroscopy (MPS), the non-imaging counterpart of MPI, enables sensitive analytics by exploiting the signal response of magnetic nanoparticles. In this pilot study, we prove the potential of MPS to continuously monitor blood coagulation in real time. Methods: Blood samples from five volunteers were mixed with the commercial magnetic resonance imaging contrast agent Resotran. The dynamics of the particle signal were assessed in a custom-built MPS-system for a duration of 45 minutes under various conditions, including the presence of anticoagulants (EDTA, Heparin, Citrate) and mechanical stress. The signal amplitude of the fifth harmonic of the MPS was analyzed. To exclude potential thermal effects, the temperature inside the MPS was monitored by using a fiber optic thermometer during the measurements. Results: All Resotran-containing blood samples showed a signal decrease over time. Samples with anticoagulants exhibited no relevant signal decrease (EDTA, Citrate) or a smaller decrease (Heparin) compared to samples without anticoagulants. Additionally, mechanical stress induced a signal decay in all samples, further indicating the link between the observed MPS signal decay and blood coagulation. Conclusion: This study shows that continuous monitoring of human blood coagulation via MPS is feasible, making bedside coagulation monitoring in clinical settings a concrete perspective.
- 1 February 2026
- Published by Taylor & Francis in International Journal of Nanomedicine
- Vol. ume 21, 1-12
Open AccessContinuous Coagulation Monitoring in Human Blood Samples via Magnetic Particle Spectroscopy
Introduction: Magnetic Particle Imaging (MPI) is a radiation-free imaging modality based on the nonlinear magnetic response of iron oxide nanoparticles, providing high sensitivity and real-time, quantitative, background-free imaging. With the clinical approval of Resotran as an MPI-suitable tracer and the development of first human-scale scanners, clinical applications are within reach. Magnetic Particle Spectroscopy (MPS), the non-imaging counterpart of MPI, enables sensitive analytics by exploiting the signal response of magnetic nanoparticles. In this pilot study, we prove the potential of MPS to continuously monitor blood coagulation in real time. Methods: Blood samples from five volunteers were mixed with the commercial magnetic resonance imaging contrast agent Resotran. The dynamics of the particle signal were assessed in a custom-built MPS-system for a duration of 45 minutes under various conditions, including the presence of anticoagulants (EDTA, Heparin, Citrate) and mechanical stress. The signal amplitude of the fifth harmonic of the MPS was analyzed. To exclude potential thermal effects, the temperature inside the MPS was monitored by using a fiber optic thermometer during the measurements. Results: All Resotran-containing blood samples showed a signal decrease over time. Samples with anticoagulants exhibited no relevant signal decrease (EDTA, Citrate) or a smaller decrease (Heparin) compared to samples without anticoagulants. Additionally, mechanical stress induced a signal decay in all samples, further indicating the link between the observed MPS signal decay and blood coagulation. Conclusion: This study shows that continuous monitoring of human blood coagulation via MPS is feasible, making bedside coagulation monitoring in clinical settings a concrete perspective.
- Open Access
- 1 February 2026
- Published by Taylor & Francis in Medical Devices: Evidence and Research
- Vol. ume 19, 1-8
Simple-to-Fabricate and Water-Stable Instrument Markers for Preclinical Magnetic Particle Imaging and Magnetic Resonance Imaging
Magnetic Particle Imaging (MPI) is a tracer-based imaging modality with high spatial and temporal resolution, very promising for cardiovascular imaging and peri-interventional guidance. However, most interventional instruments are invisible in MPI due to a lack of signal generation. Existing instrument marking approaches do not address water durability, which is essential for maintaining temporal consistency during experiments. This study aims to develop a simple and water-stable marking technique with smooth surfaces for interventional devices for preclinical MPI and MRI research. Commercial superparamagnetic iron-oxide nanoparticles (SPIONs) were mixed with a transparent varnish and applied to nitinol stents via dip-coating. An impregnation solution was used to seal the markers. The coating was optically evaluated by microscopy. The water-stability of the markers was assessed after 24 hours in water by Magnetic Particle Spectroscopy (MPS) of the surrounding fluid. The imaging performance of the markers was tested in preclinical MPI and MRI systems. Microscopy showed homogeneous SPION distribution and smooth marker surfaces after the sealing. MPS revealed no detectable signal of the water which surrounded the sealed markers, indicating their water-stability. The MPI-scans demonstrated sufficient visualization of individual markers at the stent ends (SNR=11.5). In MRI, susceptibility artefacts of the marked stents were 4.6 times larger compared to unmarked references. The presented technique allows for easy fabrication of instrument markers with commercially available components for preclinical experiments with the magnetic imaging modalities MPI and MRI.
- 1 February 2026
- Published by Taylor & Francis in Medical Devices: Evidence and Research
- Vol. ume 19, 1-8
Open AccessSimple-to-Fabricate and Water-Stable Instrument Markers for Preclinical Magnetic Particle Imaging and Magnetic Resonance Imaging
Magnetic Particle Imaging (MPI) is a tracer-based imaging modality with high spatial and temporal resolution, very promising for cardiovascular imaging and peri-interventional guidance. However, most interventional instruments are invisible in MPI due to a lack of signal generation. Existing instrument marking approaches do not address water durability, which is essential for maintaining temporal consistency during experiments. This study aims to develop a simple and water-stable marking technique with smooth surfaces for interventional devices for preclinical MPI and MRI research. Commercial superparamagnetic iron-oxide nanoparticles (SPIONs) were mixed with a transparent varnish and applied to nitinol stents via dip-coating. An impregnation solution was used to seal the markers. The coating was optically evaluated by microscopy. The water-stability of the markers was assessed after 24 hours in water by Magnetic Particle Spectroscopy (MPS) of the surrounding fluid. The imaging performance of the markers was tested in preclinical MPI and MRI systems. Microscopy showed homogeneous SPION distribution and smooth marker surfaces after the sealing. MPS revealed no detectable signal of the water which surrounded the sealed markers, indicating their water-stability. The MPI-scans demonstrated sufficient visualization of individual markers at the stent ends (SNR=11.5). In MRI, susceptibility artefacts of the marked stents were 4.6 times larger compared to unmarked references. The presented technique allows for easy fabrication of instrument markers with commercially available components for preclinical experiments with the magnetic imaging modalities MPI and MRI.
- Open Access
- 1 December 2025
- Published by Springer Nature in Communications Engineering
- Vol. 4 (1)
Towards clinical magnetic particle imaging: safety measurements of medical implants in a human cadaver model
Magnetic Particle Imaging (MPI) is a preclinical imaging modality with potential for future clinical usage. The radiation-free guidance of endovascular interventions with MPI is especially promising. Here, we present a safety study on the heating of metallic medical implants during MPI measurements under realistic conditions in an extracorporeally-perfused cadaver model. The measurements were conducted by fiberoptic thermometers and showed no detectable heating of the tested endovascular devices in the cadaver model. A temperature increase of no more than 0.11 K was observed on the surface of the investigated proximal femoral nail. The in vitro testing of orthopedic prostheses (knee and hip) revealed a slight heating effect of 0.45 K. The dependence of heating on the applied excitation frequency was measured. Overall, the tested repertoire of implants did not heat by a clinically-relevant amount in a human-sized MPI-scanner under realistic conditions, indicating their safe usage in future clinical applications.
- 1 December 2025
- Published by Springer Nature in Communications Engineering
- Vol. 4 (1)
Open AccessTowards clinical magnetic particle imaging: safety measurements of medical implants in a human cadaver model
Magnetic Particle Imaging (MPI) is a preclinical imaging modality with potential for future clinical usage. The radiation-free guidance of endovascular interventions with MPI is especially promising. Here, we present a safety study on the heating of metallic medical implants during MPI measurements under realistic conditions in an extracorporeally-perfused cadaver model. The measurements were conducted by fiberoptic thermometers and showed no detectable heating of the tested endovascular devices in the cadaver model. A temperature increase of no more than 0.11 K was observed on the surface of the investigated proximal femoral nail. The in vitro testing of orthopedic prostheses (knee and hip) revealed a slight heating effect of 0.45 K. The dependence of heating on the applied excitation frequency was measured. Overall, the tested repertoire of implants did not heat by a clinically-relevant amount in a human-sized MPI-scanner under realistic conditions, indicating their safe usage in future clinical applications.
- Open Access
- 4 August 2025
- Published by Springer Nature in European Radiology
- Vol. 36 (2) , 1330-1340
Bake your phantom—low-cost recipes for dough-based, tissue-mimicking CT phantoms
Objective: Phantoms are essential for minimizing radiation-intensive experiments on humans and enhancing patient safety through model-based teaching and experimentation. While many phantoms are commercially available, their widespread use is limited by manufacturing complexity and high costs. This study aims to demonstrate a dough-based method to create customizable, realistic CT phantoms using affordable, readily available ingredients. Materials and methods: For this study, various doughs composed of flour, salt, water, and oil were created, scanned, and evaluated to assess their suitability for CT applications. Additionally, the effects of storage conditions, preservation, and temperature variations were analyzed. As an example, a liver was segmented from a 3D CT scan, scaled to 1:2, and a negative mold was 3D printed. The mold was subsequently filled with the most suitable dough composition to replicate the organ’s anatomy and density. Results: The evaluation of the scanned ingredients and doughs demonstrated that Hounsfield unit (HU) values ranging from below −200 HU to above 1200 HU can be achieved, enabling the simulation of various human tissue densities. Based on the analysis, simple recipes are proposed to replicate radiodensities of different anatomical structures. Additionally, the results from the liver phantom confirm the feasibility of mimicking liver tissue and morphology. Conclusions: CT phantoms with specific radiodensities, mimicking human tissues, can be created using simple dough recipes, as demonstrated with our liver CT phantom. Refrigeration or freezing extends usability for a longer time, but temperature effects must be considered to ensure accurate HU values in CT scans. Key Points: Question CT phantoms are costly and complex, limiting accessibility for customized imaging studies. Can common ingredients from the kitchen be used to create tissue-mimicking phantoms ? Findings The dough-based phantoms simulated radiodensities from −200 to 1200+ HU, enabling the simulation of various human tissue densities . Clinical relevance Affordable, anatomically accurate CT phantoms can be made from kitchen ingredients for training and research, as demonstrated with our liver CT phantom . Graphical
- 4 August 2025
- Published by Springer Nature in European Radiology
- Vol. 36 (2) , 1330-1340
Open AccessBake your phantom—low-cost recipes for dough-based, tissue-mimicking CT phantoms
Objective: Phantoms are essential for minimizing radiation-intensive experiments on humans and enhancing patient safety through model-based teaching and experimentation. While many phantoms are commercially available, their widespread use is limited by manufacturing complexity and high costs. This study aims to demonstrate a dough-based method to create customizable, realistic CT phantoms using affordable, readily available ingredients. Materials and methods: For this study, various doughs composed of flour, salt, water, and oil were created, scanned, and evaluated to assess their suitability for CT applications. Additionally, the effects of storage conditions, preservation, and temperature variations were analyzed. As an example, a liver was segmented from a 3D CT scan, scaled to 1:2, and a negative mold was 3D printed. The mold was subsequently filled with the most suitable dough composition to replicate the organ’s anatomy and density. Results: The evaluation of the scanned ingredients and doughs demonstrated that Hounsfield unit (HU) values ranging from below −200 HU to above 1200 HU can be achieved, enabling the simulation of various human tissue densities. Based on the analysis, simple recipes are proposed to replicate radiodensities of different anatomical structures. Additionally, the results from the liver phantom confirm the feasibility of mimicking liver tissue and morphology. Conclusions: CT phantoms with specific radiodensities, mimicking human tissues, can be created using simple dough recipes, as demonstrated with our liver CT phantom. Refrigeration or freezing extends usability for a longer time, but temperature effects must be considered to ensure accurate HU values in CT scans. Key Points: Question CT phantoms are costly and complex, limiting accessibility for customized imaging studies. Can common ingredients from the kitchen be used to create tissue-mimicking phantoms ? Findings The dough-based phantoms simulated radiodensities from −200 to 1200+ HU, enabling the simulation of various human tissue densities . Clinical relevance Affordable, anatomically accurate CT phantoms can be made from kitchen ingredients for training and research, as demonstrated with our liver CT phantom . Graphical
- Open Access
- 21 July 2025
- Published by Springer Nature in Scientific Reports
- Vol. 15 (1) , 1-10
AI-based body composition analysis of CT data has the potential to predict disease course in patients with multiple myeloma
The aim of this study was to evaluate the benefit of a volumetric AI-based body composition analysis (BCA) algorithm in multiple myeloma (MM). Therefore, a retrospective monocentric cohort of 91 MM patients was analyzed. The BCA algorithm, powered by a convolutional neural network, quantified tissue compartments and bone density based on routine CT scans. Correlations between BCA data and demographic/clinical parameters were investigated. BCA-endotypes were identified and survival rates were compared between BCA-derived patient clusters. Patients with high-risk cytogenetics exhibited elevated cardiac marker index values. Across Revised-International Staging System (R-ISS) categories, BCA parameters did not show significant differences. However, both subcutaneous and total adipose tissue volumes were significantly lower in patients with progressive disease or death during follow-up compared to patients without progression. Cluster analysis revealed two distinct BCA-endotypes, with one group displaying significantly better survival. Furthermore, a combined model composed of clinical parameters and BCA data demonstrated a higher predictive capability for disease progression compared to models based solely on high-risk cytogenetics or R-ISS. These findings underscore the potential of BCA to improve patient stratification and refining prognostic models in MM.
- 21 July 2025
- Published by Springer Nature in Scientific Reports
- Vol. 15 (1) , 1-10
Open AccessAI-based body composition analysis of CT data has the potential to predict disease course in patients with multiple myeloma
The aim of this study was to evaluate the benefit of a volumetric AI-based body composition analysis (BCA) algorithm in multiple myeloma (MM). Therefore, a retrospective monocentric cohort of 91 MM patients was analyzed. The BCA algorithm, powered by a convolutional neural network, quantified tissue compartments and bone density based on routine CT scans. Correlations between BCA data and demographic/clinical parameters were investigated. BCA-endotypes were identified and survival rates were compared between BCA-derived patient clusters. Patients with high-risk cytogenetics exhibited elevated cardiac marker index values. Across Revised-International Staging System (R-ISS) categories, BCA parameters did not show significant differences. However, both subcutaneous and total adipose tissue volumes were significantly lower in patients with progressive disease or death during follow-up compared to patients without progression. Cluster analysis revealed two distinct BCA-endotypes, with one group displaying significantly better survival. Furthermore, a combined model composed of clinical parameters and BCA data demonstrated a higher predictive capability for disease progression compared to models based solely on high-risk cytogenetics or R-ISS. These findings underscore the potential of BCA to improve patient stratification and refining prognostic models in MM.
- Open Access
- 19 March 2025
- Published by MDPI AG in Diagnostics
- Vol. 15 (6) , 777
Impact of Radiologist Experience on AI Annotation Quality in Chest Radiographs: A Comparative Analysis
Background/Objectives: In the burgeoning field of medical imaging and Artificial Intelligence (AI), high-quality annotations for training AI-models are crucial. However, there are still only a few large datasets, as segmentation is time-consuming, experts have limited time. This study investigates how the experience of radiologists affects the quality of annotations. Methods: We randomly collected 53 anonymized chest radiographs. Fifteen readers with varying levels of expertise annotated the anatomical structures of different complexity, pneumonic opacities and central venous catheters (CVC) as examples of pathologies and foreign material. The readers were divided into three groups of five. The groups consisted of medical students (MS), junior professionals (JP) with less than five years of working experience and senior professionals (SP) with more than five years of experience. Each annotation was compared to a gold standard consisting of a consensus annotation of three senior board-certified radiologists. We calculated the Dice coefficient (DSC) and Hausdorff distance (HD) to evaluate annotation quality. Inter- and intrareader variability and time dependencies were investigated using Intraclass Correlation Coefficient (ICC) and Ordinary Least Squares (OLS). Results: Senior professionals generally showed better performance, while medical students had higher variability in their annotations. Significant differences were noted, especially for complex structures (DSC Pneumonic Opacities as mean [standard deviation]: MS: 0.516 [0.246]; SP: 0.631 [0.211]). However, it should be noted that overall deviation and intraclass variance was higher for these structures even for seniors, highlighting the inherent limitations of conventional radiography. Experience showed a positive relationship with annotation quality for VCS and lung but was not a significant factor for other structures. Conclusions: Experience level significantly impacts annotation quality. Senior radiologists provided higher-quality annotations for complex structures, while less experienced readers could still annotate simpler structures with satisfying accuracy. We suggest a mixed-expertise approach, enabling the highly experienced to utilize their knowledge most effectively. With the increase in numbers of examinations, radiology will rely on AI support tools in the future. Therefore, economizing the process of data acquisition and AI-training; for example, by integrating less experienced radiologists, will help to meet the coming challenges.
- 19 March 2025
- Published by MDPI AG in Diagnostics
- Vol. 15 (6) , 777
Open AccessImpact of Radiologist Experience on AI Annotation Quality in Chest Radiographs: A Comparative Analysis
Background/Objectives: In the burgeoning field of medical imaging and Artificial Intelligence (AI), high-quality annotations for training AI-models are crucial. However, there are still only a few large datasets, as segmentation is time-consuming, experts have limited time. This study investigates how the experience of radiologists affects the quality of annotations. Methods: We randomly collected 53 anonymized chest radiographs. Fifteen readers with varying levels of expertise annotated the anatomical structures of different complexity, pneumonic opacities and central venous catheters (CVC) as examples of pathologies and foreign material. The readers were divided into three groups of five. The groups consisted of medical students (MS), junior professionals (JP) with less than five years of working experience and senior professionals (SP) with more than five years of experience. Each annotation was compared to a gold standard consisting of a consensus annotation of three senior board-certified radiologists. We calculated the Dice coefficient (DSC) and Hausdorff distance (HD) to evaluate annotation quality. Inter- and intrareader variability and time dependencies were investigated using Intraclass Correlation Coefficient (ICC) and Ordinary Least Squares (OLS). Results: Senior professionals generally showed better performance, while medical students had higher variability in their annotations. Significant differences were noted, especially for complex structures (DSC Pneumonic Opacities as mean [standard deviation]: MS: 0.516 [0.246]; SP: 0.631 [0.211]). However, it should be noted that overall deviation and intraclass variance was higher for these structures even for seniors, highlighting the inherent limitations of conventional radiography. Experience showed a positive relationship with annotation quality for VCS and lung but was not a significant factor for other structures. Conclusions: Experience level significantly impacts annotation quality. Senior radiologists provided higher-quality annotations for complex structures, while less experienced readers could still annotate simpler structures with satisfying accuracy. We suggest a mixed-expertise approach, enabling the highly experienced to utilize their knowledge most effectively. With the increase in numbers of examinations, radiology will rely on AI support tools in the future. Therefore, economizing the process of data acquisition and AI-training; for example, by integrating less experienced radiologists, will help to meet the coming challenges.