Major project
Healthcare & Hospitals
Skin lesion classification for teledermatology
Classify skin lesion photos to help prioritise teledermatology cases.
The brief
Rural clinics send photos of skin lesions to distant dermatologists. Build a classifier that sorts lesions into common types and flags possible melanoma for urgent review, and test how it performs across skin tones. Good looks like: strong recall for melanoma, a fairness analysis across skin tones, and a referral demo.
What you will build
- Prepare images and handle class imbalance
- Train and compare CNN models
- Evaluate melanoma recall and per-class results
- Analyse performance across skin tones
- Build a referral demo
- Report safety and fairness limits
Free, open data to build it with
Check each dataset's licence before you publish your work. Your mentor gives your team its own version of the brief (data, scope or region), so every team's project is its own.