Researchers at University of Waterloo, Canada are developing an AI based system to help detect melanoma – a deadly form of skin cancer – in its early stages.
The system analyses images of skin lesions and provides doctors with data on telltale bio-markers of melanoma, which is deadly if detected late, but highly treatable if caught early.
It deciphers levels of bio-marker substances in lesions, adding consistent, quantitative information to assessments currently based on appearance alone, researchers said.
In particular, changes in the concentration and distribution of eumelanin, a chemical that gives skin its color, and hemoglobin, a protein in red blood cells, are strong indicators of melanoma, they said.
The artificial intelligence (AI) system – trained using tens of thousands of skin images and their corresponding eumelanin and hemoglobin levels – could initially reduce the number of unnecessary biopsies, a significant health-care cost, researchers said.
It gives doctors objective information on lesion characteristics to help them rule out melanoma before taking more invasive action.
“This could be a very powerful tool for skin cancer clinical decision support. The more interpretable information there is, the better the decisions are,” said Alexander Wong, a professor of systems design engineering at Waterloo.
“There can be a huge lag time before doctors even figure out what is going on with the patient. Our goal is to shorten that process” Wong added.