# ImprintLab/Medical-SAM-Adapter

A lightweight adapter bridges SAM with medical imaging [MedIA]

Repository: https://github.com/ImprintLab/Medical-SAM-Adapter
Canonical: https://ross.abutalabs.com/products/medical-sam-adapter
Language: Python
License: GPL-3.0
License Family: copyleft
Topics: adapter, deep-learning, medical-imaging, segment-anything-model, segmentagtion
Last push: 2025-09-08T16:38:31+00:00

## Health v2 (maintenance only)
Score: 39/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 41, release rhythm 8, longevity 87
- inputs: {"age_days": 1226, "days_push": 359, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1322, forks 127 (observed 2026-08-28T04:04:22.046353+00:00)

## What it is
Medical SAM Adapter (MSA) is a Python framework that fine-tunes Meta's Segment Anything Model for medical image segmentation using lightweight adapter modules. It supports multiple SAM variants (SAM, MobileSAM, EfficientSAM) and LoRA-based adaptation, with pre-trained adapters available in a model zoo.

## Use cases
- segment medical images like fundus or CT scans with SAM
- fine-tune Segment Anything Model on medical imaging datasets
- adapt SAM to a new domain without full model retraining
- use pre-trained medical segmentation adapters without training
- compare SAM, MobileSAM, and EfficientSAM for medical segmentation
- train a lightweight LoRA adapter on top of SAM

## When to choose
- you need parameter-efficient fine-tuning of SAM for medical images
- you want pre-trained medical segmentation adapters instead of training from scratch
- you need a lightweight alternative to full SAM fine-tuning

## When to avoid
- you need general-purpose (non-medical) image segmentation
- you want a production-ready clinical application rather than a research framework
- you need segmentation models unrelated to the SAM architecture

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, deep-learning, image-processing, computer-vision, llm-training
- domain: deep-learning, computer-vision, healthcare, machine-learning
- platform: python
- tags: segment-anything, adapter-tuning, medical-image-segmentation, parameter-efficient-fine-tuning, lora, pytorch, research-code, medical-imaging, gpu

## Member repositories
- ImprintLab/Medical-SAM-Adapter (main) score 39

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:22.046353+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 2026-08-30T04:47:06.534773+00:00, confidence not recorded.
  - readme: https://github.com/ImprintLab/Medical-SAM-Adapter (fetched 2026-08-28T04:04:22.046353+00:00, sha 745160d4b13f)
- Data as of 2026-08-30T08:39:29.467469+00:00.