# MONAI

AI Toolkit for Healthcare Imaging

Repository: https://github.com/Project-MONAI/MONAI
Canonical: https://ross.abutalabs.com/products/monai
Homepage: https://project-monai.github.io/
Language: Python
License: Apache-2.0
License Family: permissive
Topics: healthcare-imaging, deep-learning, medical-image-computing, medical-image-processing, pytorch, python3, monai
Last push: 2026-08-26T13:54:22+00:00
Link (homepage): https://project-monai.github.io/

## Health v2 (maintenance only)
Score: 87/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 63, longevity 100
- inputs: {"age_days": 2518, "days_push": 7, "days_rel": 84, "gap_med": 130.0, "n_releases_24m": 5}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 8634, forks 1605 (observed 2026-08-28T04:10:23.946925+00:00)

## What it is
MONAI is a PyTorch-based open-source framework for deep learning in healthcare imaging, providing domain-specific transforms, 3D architectures, losses, and metrics. It is part of the PyTorch Ecosystem and includes companion projects for active-learning annotation (MONAI Label) and clinical deployment (MONAI Deploy).

## Use cases
- train a 3D segmentation model on CT or MRI volumes
- preprocess DICOM and NIfTI medical images for deep learning
- evaluate segmentation with Dice and Hausdorff metrics
- deploy a trained model as a clinical inference pipeline
- use pre-trained models like UNETR or SwinUNETR from the Model Zoo
- run multi-GPU distributed training on medical imaging data
- generate synthetic CT images with MAISI

## When to choose
- you are doing deep learning research or production work on medical imaging
- you need 3D spatial transforms and domain-specific architectures on top of PyTorch
- you want reproducible, portable model packaging via MONAI Bundles
- you need clinically validated segmentation metrics and losses

## When to avoid
- your project is general-purpose computer vision on natural images
- you need a non-PyTorch framework like TensorFlow or JAX
- you only need lightweight image viewing or DICOM parsing without deep learning

## Facets
- artifact type: framework
- maturity: stable
- function: deep-learning, machine-learning, image-processing, data-science
- domain: healthcare, deep-learning, machine-learning, image-processing, artificial-intelligence
- platform: python, cross-platform
- tags: medical-imaging, pytorch, segmentation, dicom, nifti, healthcare-imaging, model-zoo, gpu, docker

## Member repositories
- Project-MONAI/MONAI (main) score 87
- Project-MONAI/tutorials (examples) score 76

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:10:23.946925+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T17:25:47.153849+00:00, confidence not recorded.
  - readme: https://github.com/Project-MONAI/MONAI (fetched 2026-08-28T04:10:23.946925+00:00, sha 6eb2799943e6)
  - homepage: https://project-monai.github.io/ (fetched 2026-08-29T08:26:06.535552+00:00, sha 3a12c6d3df1d)
  - registry_pypi: https://pypi.org/pypi/monai/json (fetched 2026-08-29T08:26:06.538178+00:00, sha ce1ecb82e17b)
- Data as of 2026-08-30T08:39:29.467469+00:00.