# ZJU4HealthCare/HealthGPT

【ICML 2025 Spotlight】 Official Repo for Paper ‘’HealthGPT : A Medical Large Vision-Language Model for Unifying Comprehension and Generation via Heterogeneous Knowledge Adaptation‘’

Repository: https://github.com/ZJU4HealthCare/HealthGPT
Canonical: https://ross.abutalabs.com/products/zju4healthcare-healthgpt
Language: Python
License: Apache-2.0
License Family: permissive
Last push: 2026-07-31T23:27:29+00:00

## Health v2 (maintenance only)
Score: 63/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 95, release rhythm 35, longevity 40
- inputs: {"age_days": 562, "days_push": 33, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1654, forks 243 (observed 2026-08-28T04:05:17.123160+00:00)

## What it is
HealthGPT is a medical multimodal large language model family unifying medical image comprehension and generation via heterogeneous knowledge adaptation, with HealthGPT-Pro extending understanding to text, 2D images, and 3D volumes. It is the official research code release for an ICML 2025 Spotlight paper, with pretrained models on Hugging Face.

## Use cases
- build a medical vision-language model
- analyze medical images with an LLM
- unified medical image understanding and generation
- run inference on 3D medical volumes
- research multimodal medical AI
- fine-tune a medical MLLM on clinical data

## When to choose
- you need state-of-the-art medical multimodal understanding across text, 2D images, and 3D volumes
- you are doing research on unified medical comprehension and generation
- you want pretrained medical MLLM checkpoints to build on

## When to avoid
- you need a production-ready clinical system with regulatory compliance
- you lack GPU resources for large multimodal models
- your task is general-purpose (non-medical) vision-language processing

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, deep-learning, llm-training, image-processing, nlp
- domain: artificial-intelligence, large-language-models, healthcare, computer-vision, deep-learning
- platform: python
- tags: medical-imaging, multimodal-llm, vision-language-model, medical-ai, research-code, icml-2025, gpu, linux

## Member repositories
- ZJU4HealthCare/HealthGPT (main) score 63

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:17.123160+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 2026-08-30T03:44:54.895598+00:00, confidence not recorded.
  - readme: https://github.com/ZJU4HealthCare/HealthGPT (fetched 2026-08-28T04:05:17.123160+00:00, sha a34abf673805)
- Data as of 2026-08-30T08:39:29.467469+00:00.