wasserth/TotalSegmentator

Health v2 · maintenance only

Adoption not part of the score

2952 stars · 473 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

TotalSegmentator is a Python command-line tool that robustly segments over 100 anatomical structures in CT and MR images using deep learning models built on nnUNet. It supports Nifti and DICOM inputs, runs on CPU or GPU, and offers subtasks, body statistics prediction, and a 3D Slicer extension.

Use cases

• segment organs in CT scans
• segment anatomical structures in MRI images
• convert DICOM slices to segmentation masks
• estimate body stats like height and weight from CT
• run vertebrae segmentation on spine images
• batch process medical images for research

When to choose

• you need automated multi-structure segmentation of CT or MR images
• you want a robust pretrained model that generalizes across scanners and protocols
• you need a CLI tool that works on CPU and GPU across Linux, Mac, and Windows

When to avoid

• you need segmentation of structures not covered by its tasks
• you require real-time or interactive segmentation rather than batch inference
• you cannot install Python 3.10+ and PyTorch dependencies

Facets

cli-tool · maturity active

machine-learning image-processing computer-vision cli healthcare machine-learning image-processing windows python cli medical-imaging segmentation ct mri nifti dicom nnunet anatomy linux macos gpu

Member repositories

For agents

markdown · JSON · MCP: product_card(name="wasserth/TotalSegmentator")