Software
MedVIC maintains two flagship platforms and 15+ additional open-source tools. All code lives on GitHub / MedVIC-Lab.
ShapeWorks
ShapeWorks — Open-Source Anatomy Modeling Platform
ShapeWorks is a free, open-source platform for automated construction of statistical shape models of anatomical structures. Developed by the MedVIC lab and the Scientific Computing and Imaging Institute, it provides the complete analysis pipeline: image preprocessing and quality control, correspondence optimization, shape representation, statistical analysis, and visualization — all in a unified, modular framework.
Applications: cardiology · orthopedics · craniofacial surgery · neuroscience · spine · biomechanics
CranioRate
CranioRate — AI Severity Scoring for Craniosynostosis
CranioRate converts a patient's 3D CT scan into an objective, population-grounded severity score for craniosynostosis — replacing subjective, qualitative clinical assessment with reproducible, data-driven measurement. Severity is not a number: it is a shape. CranioRate quantifies deviation from normative anatomy across a continuous spectrum aligned with expert clinical judgment.
Deployed across 7 institutions, it is becoming a shared measurement standard across sites, surgeons, and studies — enabling consistent assessment regardless of where the patient is seen.
Supported subtypes: sagittal · metopic · unicoronal
Additional Open-Source Tools
Image Registration
| Tool | Description | Venue | Links |
|---|---|---|---|
| EfficientMorph | Parameter-efficient transformer for 3D image registration | WACV 2025 | GitHub |
| LEDA | Log-Euclidean Diffeomorphism Autoencoder for statistical analysis of deformations | IPMI 2025 | GitHub |
| MORPH-LER | Population-aware log-Euclidean regularization for deep registration | MIDL 2025 | GitHub |
Shape Modeling from Images
| Tool | Description | Venue | Links |
|---|---|---|---|
| VIB-DeepSSM / BVIB-DeepSSM | Probabilistic anatomical shape models from images | MICCAI 2022, 2023 | GitHub |
| Image2SSM | Statistical shape models from images using radial basis functions | MICCAI 2023 | GitHub |
| SCorP | Statistics-informed dense correspondence from unsegmented images | MIUA 2024 | GitHub |
| MASSM | Multi-anatomy shape modeling directly from images | ShapeMI 2024 | GitHub |
| Uncertain-DeepSSM | Probabilistic shape models with uncertainty quantification | ShapeMI 2020 | — |
Shape Modeling from Surfaces
| Tool | Description | Venue | Links |
|---|---|---|---|
| Point2SSM / Point2SSM++ | Shape models from point clouds | ICLR 2024 / MedIA 2026 | GitHub |
| Mesh2SSM | Shape models from surface meshes | MICCAI 2023 | GitHub |
| SPI-CorrNet | Probabilistic 3D correspondence from sparse unsegmented images | MICCAI-MLMI 2024 | GitHub |
Generative Modeling
| Tool | Description | Venue | Links |
|---|---|---|---|
| ShapeOdds | Variational Bayesian learning of generative shape models | CVPR 2017 | GitHub |
| MorphoFlow | Normalizing flows for morphological shape modeling | Under review 2026 | — |
| dpVAE | Decoupled priors for variational autoencoders | ACCV 2020 | GitHub |
Computational Pathology
| Tool | Description | Venue | Links |
|---|---|---|---|
| StainDiffuser | Virtual staining via dual-task score-based diffusion | arXiv 2024 | GitHub |
| Structural CycleGAN | Structure-preserving virtual immunohistochemistry | MICCAI-MLMI 2023 | GitHub |
| ImplicitStainer | Data-efficient virtual staining using local implicit functions | arXiv 2025 | GitHub |
| Histopathology Pretraining Study | Domain-specific pretraining analysis | MICCAI-MILLanD 2023 | GitHub |
All tools are open-source and available on GitHub / MedVIC-Lab. We welcome issues, pull requests, and collaboration inquiries.