IEEE International Symposium on Biomedical Imaging (ISBI) 2026
April 8 – 11, 2026 | London, UK
Kitware is excited to announce our participation in the IEEE International Symposium on Biomedical Imaging (ISBI) 2026, taking place in London, UK. ISBI brings together experts working on the theory, algorithms, and computational methods that power modern biomedical imaging—from microscopic analysis to whole-body systems. The conference creates a space where different imaging disciplines connect, exchange ideas, and push the field forward through collaboration.
Kitware’s Activities and Involvement
Segmentation from Partial Views via Patient-specific Shape Priors
Thursday, April 9 at 10:50 AM (BST)
Authors: Jared Vicory, Dženan Zukić, Balazs Vagvolgyi, Peter Kazanzides, Andinet Enquobahrie, Emad Boctor
We are working on a problem that requires estimating the full 3D shape of the prostate from ultrasound images that only capture part of the organ. Our approach combines a standard segmentation approach on the visible region with a shape model to estimate the missing parts of the boundary. This model has two components: a pre-trained model built from a large population of prostate shapes, and a patient-specific adjustment derived from a pre-operative MRI scan. By combining these sources of information, our method produces more accurate results than traditional segmentation approaches, especially when only a small portion of the prostate is visible.
This work was funded by the Advanced Research Projects Agency for Health (ARPA-H) under Award Number D24AC00359-00. The ARPA-H award provided 100% of total costs and total up to $20.9M. The content is solely the responsibility of the authors and does not necessarily represent the official views of the Advanced Research Projects Agency for Health.
Advances in Shape Modeling and 3D Reconstruction in Medical Imaging
Our work builds on a broader set of efforts around using anatomical shape as a meaningful source of information in medical imaging. Shape can capture subtle structural differences that traditional intensity-based approaches often miss, opening the door to earlier detection, better patient stratification, and more personalized care. Through tools like SlicerSALT, we’ve explored how analyzing shape across populations can help uncover high-dimensional biomarkers and provide deeper clinical insight.
We also draw on ongoing work in statistical shape modeling, where population-level models are used to understand how anatomy varies and to support applications like 3D reconstruction, predictive modeling, and patient-specific simulation. These approaches combine geometric representations with data-driven methods, making them a strong foundation for many precision medicine workflows.
More broadly, this work connects with efforts to reconstruct full anatomical context from partial or limited imaging. Techniques that infer 3D structure from inherently 2D modalities, such as X-ray or ultrasound, are helping improve the accuracy and usability of imaging data for surgical planning and intervention, while also reducing cost and complexity.
Partner with Kitware
We work closely with academic, clinical, and industry partners to solve complex challenges in medical imaging, including segmentation, registration, and multimodal data integration. Through open source development and collaborative R&D, we help translate advanced imaging methods into scalable, real-world clinical solutions.
If you’re interested in collaborating or exploring how Kitware can support your research or product development efforts, send us a message.