Csaba Pinter
Research Subjects
Research Interests
Csaba Pinter has a total of 106 co-authors. Here are the top co-authors by number of shared publications:
Gabor Fichtinger
Laboratory for Percutaneous Surgery, Queen’s University, Kingston, Canada
7 shared publications- Andras Lasso
Laboratory for Percutaneous Surgery, Queen's University, Kingston, Ontario, Canada
5 shared publications Andras Lasso
Laboratory for Percutaneous Surgery, School of Computing, 557 Goodwin Hall, Queen's University, K7L 2N8, Kingston, Ontario, Canada
4 shared publications- Gabor Fichtinger
Laboratory for Percutaneous Surgery, School of Computing, Queen’s University, Kingston, Canada
4 shared publications - Kevin M Alexander
Medical Physics Department, Kingston Health Sciences Centre Cancer Centre of Southeastern Ontario, 25 King Street West, Kingston, Ontario, K7L 5P9, CANADA
2 shared publications Lindsay S. Rogers
Division of Cardiology, Children’s Hospital of Philadelphia, PA. (T.R.W., Y.W., M.D.Q., L.S.R., M.A.J.)
2 shared publications
Latest Publications
- Open Access
- 2 April 2026
- Published by Elsevier in Computer Methods and Programs in Biomedicine
- Vol. 281, 109344
BoneMesh: An open-source 3D slicer framework for automated mesh generation and material mapping from CT to finite elements
- 2 April 2026
- Published by Elsevier in Computer Methods and Programs in Biomedicine
- Vol. 281, 109344
Open AccessBoneMesh: An open-source 3D slicer framework for automated mesh generation and material mapping from CT to finite elements
- Open Access
- 13 May 2025
- Published by Elsevier in Journal of Dentistry
- Vol. 159, 105813
Automatic deep learning segmentation of mandibular periodontal bone topography on cone-beam computed tomography images
- 13 May 2025
- Published by Elsevier in Journal of Dentistry
- Vol. 159, 105813
Open AccessAutomatic deep learning segmentation of mandibular periodontal bone topography on cone-beam computed tomography images
- Open Access
Automated Segmentation of Trunk Musculature with a Deep CNN Trained from Sparse Annotations in Radiation Therapy Patients with Metastatic Spine Disease
Purpose Given the high prevalence of vertebral fractures post-radiotherapy in patients with metastatic spine disease, accurate and rapid muscle segmentation could support efforts to quantify muscular changes due to disease or treatment and enable biomechanical modeling for assessments of vertebral loading to improve personalized evaluation of vertebral fracture risk. This study presents a deep-learning approach for segmenting the complete volume of the trunk muscles from clinical CT images trained using sparsely annotated data. Materials and Methods we extracted 2,009 axial CT images at the midpoint of each vertebral level (T4 to L4) from clinical CT of 148 cancer patients. The key extensor and flexor muscles (up to 8 muscles per side) were manually contoured and labeled per image in the thoracic and lumbar regions. We first trained a 2D nnU-Net deep-learning model on these labels to segment key extensor and flexor muscles. Using these sparse annotations per spine, we trained the model to segment each muscle’s entire 3D volume Results The proposed method achieved comparable performance to manual segmentations, as assessed by expert radiologists, with a mean Dice score above 0.769. Significantly, the model drastically reduced segmentation time, from 4.3-6.5 hours for manual segmentation of 14 single axial CT images to approximately 1 minute for segmenting the complete thoracic-abdominal 3D volume. Conclusion The approach demonstrates high potential for automating 3D muscle segmentation, significantly reducing the manual intervention required for generating musculoskeletal models, and could be instrumental in enhancing clinical decision-making and patient care in radiation oncology. Summary A deep learning 2D nnU-Net model, trained on a sparse set of 2D muscle annotations, successfully segmented the entire volume of 20 thoracolumbar muscles from cancer patients’ clinical CT data. The model showed a remarkable increase in segmentation efficacy and generalizability, achieving comparable performance to manual segmentations in delineating each muscle anatomy. Key Points ▪ A deep learning model (2D nnU-Net), developed using a sparse set of single axial CT-slice at each mid-per vertebral level, containing manual image annotation of 20 thoracic and lumbar muscles, achieved comparable performance to manual segmentations, as assessed by expert radiologists, with a mean Dice score above 0.769. ▪ The model drastically reduced segmentation time, from 4.3-6.5 hours for manual segmentation of 14 single axial CT images to approximately 1 minute for segmenting the complete thoracic-abdominal 3D volume. ▪ Radiologist assessment based on a Likert scale (0-5) for clinical acceptability of the muscle anatomical segmentation showed strong model performance for a representative sample of clinical CT data (a (mean(SD) of 4.66 (0.73)) and external data (4.66 (0.73).
Open AccessAutomated Segmentation of Trunk Musculature with a Deep CNN Trained from Sparse Annotations in Radiation Therapy Patients with Metastatic Spine Disease
Purpose Given the high prevalence of vertebral fractures post-radiotherapy in patients with metastatic spine disease, accurate and rapid muscle segmentation could support efforts to quantify muscular changes due to disease or treatment and enable biomechanical modeling for assessments of vertebral loading to improve personalized evaluation of vertebral fracture risk. This study presents a deep-learning approach for segmenting the complete volume of the trunk muscles from clinical CT images trained using sparsely annotated data. Materials and Methods we extracted 2,009 axial CT images at the midpoint of each vertebral level (T4 to L4) from clinical CT of 148 cancer patients. The key extensor and flexor muscles (up to 8 muscles per side) were manually contoured and labeled per image in the thoracic and lumbar regions. We first trained a 2D nnU-Net deep-learning model on these labels to segment key extensor and flexor muscles. Using these sparse annotations per spine, we trained the model to segment each muscle’s entire 3D volume Results The proposed method achieved comparable performance to manual segmentations, as assessed by expert radiologists, with a mean Dice score above 0.769. Significantly, the model drastically reduced segmentation time, from 4.3-6.5 hours for manual segmentation of 14 single axial CT images to approximately 1 minute for segmenting the complete thoracic-abdominal 3D volume. Conclusion The approach demonstrates high potential for automating 3D muscle segmentation, significantly reducing the manual intervention required for generating musculoskeletal models, and could be instrumental in enhancing clinical decision-making and patient care in radiation oncology. Summary A deep learning 2D nnU-Net model, trained on a sparse set of 2D muscle annotations, successfully segmented the entire volume of 20 thoracolumbar muscles from cancer patients’ clinical CT data. The model showed a remarkable increase in segmentation efficacy and generalizability, achieving comparable performance to manual segmentations in delineating each muscle anatomy. Key Points ▪ A deep learning model (2D nnU-Net), developed using a sparse set of single axial CT-slice at each mid-per vertebral level, containing manual image annotation of 20 thoracic and lumbar muscles, achieved comparable performance to manual segmentations, as assessed by expert radiologists, with a mean Dice score above 0.769. ▪ The model drastically reduced segmentation time, from 4.3-6.5 hours for manual segmentation of 14 single axial CT images to approximately 1 minute for segmenting the complete thoracic-abdominal 3D volume. ▪ Radiologist assessment based on a Likert scale (0-5) for clinical acceptability of the muscle anatomical segmentation showed strong model performance for a representative sample of clinical CT data (a (mean(SD) of 4.66 (0.73)) and external data (4.66 (0.73).
- 12 September 2022
- Published by Wolters Kluwer Health in Circulation: Cardiovascular Imaging
- Vol. 15 (12) , e014424
Truncal Valve Repair: 3-Dimensional Imaging and Modeling to Enhance Preoperative Surgical Planning
- 12 September 2022
- Published by Wolters Kluwer Health in Circulation: Cardiovascular Imaging
- Vol. 15 (12) , e014424
Truncal Valve Repair: 3-Dimensional Imaging and Modeling to Enhance Preoperative Surgical Planning
- Open Access
- 15 July 2021
- Published by Elsevier in Journal of the American Society of Echocardiography
- Vol. 34 (10) , 1111-1114
Simulation of Delivery of Clip-Based Therapies Within Multimodality Images to Facilitate Preprocedural Planning
- 15 July 2021
- Published by Elsevier in Journal of the American Society of Echocardiography
- Vol. 34 (10) , 1111-1114
Open AccessSimulation of Delivery of Clip-Based Therapies Within Multimodality Images to Facilitate Preprocedural Planning
- 24 June 2020
- Published by Springer Nature in Soft Computing
- Vol. 24 (24) , 18959-18970
Using fuzzy logics to determine optimal oversampling factor for voxelizing 3D surfaces in radiation therapy
- 24 June 2020
- Published by Springer Nature in Soft Computing
- Vol. 24 (24) , 18959-18970
Using fuzzy logics to determine optimal oversampling factor for voxelizing 3D surfaces in radiation therapy
- 26 March 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Medical Robotics and Bionics
- Vol. 2 (2) , 108-117
SlicerVR for Medical Intervention Training and Planning in Immersive Virtual Reality
Virtual reality (VR) provides immersive visualization that has proved to be useful in a variety of medical applications. Currently, however, no free open-source software platform exists that would provide comprehensive support for translational clinical researchers in prototyping experimental VR scenarios in training, planning or guiding medical interventions. By integrating VR functions in 3D Slicer, an established medical image analysis and visualization platform, SlicerVR enables virtual reality experience by a single click. It provides functions to navigate and manipulate the virtual scene, as well as various settings to abate the feeling of motion sickness. SlicerVR allows for shared collaborative VR experience both locally and remotely. We present illustrative scenarios created with SlicerVR in a wide spectrum of applications, including echocardiography, neurosurgery, spine surgery, brachytherapy, intervention training and personalized patient education. SlicerVR is freely available under BSD type license as an extension to 3D Slicer and it has been downloaded over 7,800 times at the time of writing this article.
- 26 March 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Medical Robotics and Bionics
- Vol. 2 (2) , 108-117
SlicerVR for Medical Intervention Training and Planning in Immersive Virtual Reality
Virtual reality (VR) provides immersive visualization that has proved to be useful in a variety of medical applications. Currently, however, no free open-source software platform exists that would provide comprehensive support for translational clinical researchers in prototyping experimental VR scenarios in training, planning or guiding medical interventions. By integrating VR functions in 3D Slicer, an established medical image analysis and visualization platform, SlicerVR enables virtual reality experience by a single click. It provides functions to navigate and manipulate the virtual scene, as well as various settings to abate the feeling of motion sickness. SlicerVR allows for shared collaborative VR experience both locally and remotely. We present illustrative scenarios created with SlicerVR in a wide spectrum of applications, including echocardiography, neurosurgery, spine surgery, brachytherapy, intervention training and personalized patient education. SlicerVR is freely available under BSD type license as an extension to 3D Slicer and it has been downloaded over 7,800 times at the time of writing this article.
- 1 April 2019
- Published by Elsevier in Computer Methods and Programs in Biomedicine
- Vol. 171, 19-26
Polymorph segmentation representation for medical image computing
- 1 April 2019
- Published by Elsevier in Computer Methods and Programs in Biomedicine
- Vol. 171, 19-26
Polymorph segmentation representation for medical image computing
- 1 July 2018
- Published by IOP Publishing in Biomedical Physics & Engineering Express
- Vol. 4 (4) , 045041
Streamlined open-source gel dosimetry analysis in 3D slicer
Three dimensional dosimetry is being used in an increasingly wide variety of clinical applications as more gel and radiochromic plastic dosimeters become available. However, accessible 3D dosimetry analysis tools have not kept pace. 3D dosimetry data analysis is time consuming and laborious, creating a barrier to entry for busy clinical environments. To help in the adoption of 3D dosimetry, we have produced a streamlined, open-source dosimetry analysis system by developing a custom extension in 3D Slicer, called the Gel Dosimetry Analysis slicelet, which enables rapid and accurate data analysis. To assist those interested in adopting 3D dosimetry in their clinic or those unfamiliar with what is involved in a 3D dosimeter experiment, we first present the workflow of a typical gel dosimetry experiment. This is followed by the results of experiments used to validate, step-wise, each component of our software. Overall, our software has made a full 3D gel dosimeter analysis roughly 20 times faster than previous analysis systems.
- 1 July 2018
- Published by IOP Publishing in Biomedical Physics & Engineering Express
- Vol. 4 (4) , 045041
Streamlined open-source gel dosimetry analysis in 3D slicer
Three dimensional dosimetry is being used in an increasingly wide variety of clinical applications as more gel and radiochromic plastic dosimeters become available. However, accessible 3D dosimetry analysis tools have not kept pace. 3D dosimetry data analysis is time consuming and laborious, creating a barrier to entry for busy clinical environments. To help in the adoption of 3D dosimetry, we have produced a streamlined, open-source dosimetry analysis system by developing a custom extension in 3D Slicer, called the Gel Dosimetry Analysis slicelet, which enables rapid and accurate data analysis. To assist those interested in adopting 3D dosimetry in their clinic or those unfamiliar with what is involved in a 3D dosimeter experiment, we first present the workflow of a typical gel dosimetry experiment. This is followed by the results of experiments used to validate, step-wise, each component of our software. Overall, our software has made a full 3D gel dosimeter analysis roughly 20 times faster than previous analysis systems.
- 3 March 2017
- Published by SPIE-Intl Soc Optical Eng
- p. 101351G-101351G-8
Open-source software for collision detection in external beam radiation therapy
- 3 March 2017
- Published by SPIE-Intl Soc Optical Eng
- p. 101351G-101351G-8
Open-source software for collision detection in external beam radiation therapy