Medical Data for Machine Learning
A curated list of publicly available medical datasets for machine learning, covering imaging, EHRs, literature, and speech.
What is Medical Data for Machine Learning?
Medical Data for Machine Learning is a curated GitHub repository that aggregates publicly available medical datasets for AI and ML research. It provides organized links to datasets across domains like medical imaging, electronic health records, biomedical literature, and speech data, saving researchers time in data discovery.
Machine learning researchers, data scientists, and healthcare AI developers who need access to diverse, real-world medical data for training and evaluating models.
It centralizes scattered medical data resources into a single, well-structured list, reducing the overhead of finding and vetting datasets. The repository is community-maintained and focuses on datasets that are actually usable for ML tasks.
Overview
This repository is a comprehensive, community-maintained collection of medical datasets suitable for machine learning research. It serves as a centralized resource to help researchers and developers discover and access diverse medical data without scouring multiple sources.
Key Features
- Medical Imaging Datasets — Extensive list of MRI, CT, X-ray, and histology image collections for various conditions like cancer, Alzheimer's, and cardiac diseases.
- EHR and Clinical Data — Includes critical care databases (e.g., MIMIC-III), lab test embeddings, and public health datasets from sources like CDC and Medicare.
- Biomedical Literature Corpora — Datasets derived from PubMed articles, including RCT abstracts and question-answering sets for evidence-based medicine.
- Challenge and Competition Data — Links to data from Grand Challenges, Kaggle competitions, and MICCAI/ISBI events for benchmarking algorithms.
- Medical Speech Data — Acoustic and articulatory speech recordings from individuals with dysarthria, autism, or cancer treatment effects.
Philosophy
The project aims to democratize access to medical data for the machine learning community by aggregating and organizing disparate public resources, thereby accelerating research in healthcare AI.
Found a gem we're missing?
Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.