Programme 01

Equitable Dermatology

Skin diagnostics engineered to perform across the full Fitzpatrick spectrum, built for the places where no dermatologist is coming.

Why This Programme

Under 10%

Dermatology training image datasets adequately representing Fitzpatrick types V and VI.

1 in 1,000,000

Dermatologists per head of population in many low and middle income regions, against 1 in 30,000 in high income countries.

A model trained without the full spectrum of human skin does not fail loudly. It fails quietly, and it fails hardest for the people who already wait longest for care. For the billions who will never see a dermatologist, the quality of this software is not a convenience question. It is the whole of their access.

Three Parts

One Signal Path, Three Pieces of Work

The Sensor

A non invasive skin biosensor that measures directly rather than inferring from a photograph. It supplies the signal everything else reads.

In development

The Method

PCA feature extraction stratified across Fitzpatrick types I to VI, trained in a federated framework with differential privacy so institutions can contribute without moving patient data. Published open access.

The Research ›

The Product

SkinWise turns the measurement into something a person can act on this morning, and it is how the programme reaches people directly.

SkinWise ›

The sensor measures. The method reads. The product decides. Each part is only as good as the one before it, which is why we build all three.

Work With Us

Three kinds of people we want to hear from.