ENGINEERINGNET.BE – ‘As a dermatologist, I pay attention to things like pigment spots, wrinkles or the fact that someone has lived in a sunny country. The algorithm picks up more characteristics than just those classic risk factors for skin cancer,’ says dermatologist Dr. Marlies Wakkee.
The Erasmus MC team trained the AI model with more than 2,800 facial photos, each taken under the same conditions. The researchers know which people developed skin cancer after taking the photo.
The algorithm distilled 200 characteristics from the photos, on the basis of which the risk of developing skin cancer is predicted, on the face or elsewhere on the body. The AI prediction is better than a prediction based on skin research, questionnaires and genetic research.
The researchers used so-called explainable AI methods. This means that it can be determined on what the algorithm bases its predictions. That is not always the case with AI. It is often a kind of black box where it is not clear how the model arrives at a certain outcome.
‘We make visible the 200 characteristics on which the model bases its prediction. If we try to translate this back into what is known about the risk of skin cancer, we see that the algorithm extracts features from faces that we already know are associated with skin cancer. Consider wrinkles, for example,” Wakkee explains.
But the model also picks up features that are less clear to people. ‘Explainable is not the same as interpretable. We can explain what the algorithm does, but its meaning is not so clear,” said co-researcher Dr. Gennady Roshchupkin.
Wakkee and Roshchupkin see two possible roles for their AI skin cancer risk model: prevention and awareness. ‘If you know from a photo that you have a high risk of skin cancer, you will use the sun more consciously. We also hope that warned people will keep a closer eye on their skin.’
For the time being, the AI model cannot yet be used in the clinic, but the focus is on follow-up research. ‘For example, we want to test the algorithm on larger datasets with photos of people with more ages and ethnic backgrounds. By publicly sharing the model in the research community, fellow scientists can build on it. That can accelerate the entire community of AI,” Roshchupkin said.
