kakumarabhishek/Corrected-Skin-Image-Datasets

Data quality analysis of DermaMNIST (MedMNIST), HAM10000, and Fitzpatrick17k datasets

21
/ 100
Experimental

This project helps medical researchers and AI model developers ensure the quality of dermatological image datasets used for training diagnostic AI. It analyzes widely-used skin image collections like DermaMNIST, HAM10000, and Fitzpatrick17k to identify and correct issues like duplicates or mislabeled images. The output is a set of cleaned and enhanced metadata files, ready for building more reliable AI models.

No commits in the last 6 months.

Use this if you are a medical researcher or AI developer working with skin image datasets and need to verify or improve their quality before training AI models for dermatological diagnosis.

Not ideal if you are looking for new, raw dermatological image data or a tool for real-time image analysis.

dermatology medical imaging AI model training dataset curation diagnostic AI
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 5 / 25
Maturity 16 / 25
Community 0 / 25

How are scores calculated?

Stars

9

Forks

Language

Jupyter Notebook

License

Apache-2.0

Last pushed

Feb 13, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/kakumarabhishek/Corrected-Skin-Image-Datasets"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.