# Tencent/MedicalNet

Many studies have shown that the performance on deep learning is significantly affected by volume of training data. The MedicalNet project provides a series of 3D-ResNet pre-trained models and relative code.

Repository: https://github.com/Tencent/MedicalNet
Canonical: https://ross.abutalabs.com/products/medicalnet
Language: Python
License: NOASSERTION
License Family: other
Last push: 2025-11-27T06:45:26+00:00

## Health v2 (maintenance only)
Score: 57/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 54, release rhythm 35, longevity 100
- inputs: {"age_days": 2604, "days_push": 279, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2255, forks 462 (observed 2026-08-28T04:06:31.239029+00:00)

## What it is
MedicalNet provides a series of 3D-ResNet pre-trained models trained on 23 diverse medical imaging datasets, with PyTorch transfer-learning code based on the Med3D paper. It enables fine-tuning strong 3D backbones for medical image segmentation and classification tasks.

## Use cases
- fine-tune 3D resnet on medical images
- pretrained 3D medical image segmentation backbone
- transfer learning for CT/MRI deep learning
- improve segmentation accuracy with pretrained weights
- medical image classification with limited data

## When to choose
- you need 3D medical imaging pretrained weights for transfer learning
- your labeled medical dataset is too small to train from scratch
- you work in PyTorch with 3D ResNet architectures

## When to avoid
- you need 2D pretrained models or non-ResNet architectures
- you require a permissively maintained, actively updated library
- you need out-of-the-box inference without writing training code

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, deep-learning, image-processing
- domain: healthcare, deep-learning, machine-learning, computer-vision
- platform: python
- tags: pretrained-models, 3d-resnet, medical-imaging, transfer-learning, segmentation, pytorch, gpu, linux

## Member repositories
- Tencent/MedicalNet (main) score 57

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:06:31.239029+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 2026-08-30T02:43:44.333070+00:00, confidence not recorded.
  - readme: https://github.com/Tencent/MedicalNet (fetched 2026-08-28T04:06:31.239029+00:00, sha 825187b7dcb8)
- Data as of 2026-08-30T08:39:29.467469+00:00.