Leo Innovation Lab wanted to develop an App giving users the opportunity to diagnose skin diseases based on a photo of skin. Being part of a healthcare company, it was imperative to Leo Innovation Lab that they properly estimated the uncertainty in the predictions in order to not miss diagnose the user, but rather know when they needed too retake the photo or consult with a doctor.
Accurately quantifying uncertainty in the diagnoses is imperative for predictions of images. It can be as simple as dealing with photos that don’t contain skin, finding out when a picture needs to be retaken from a different angle, or when a doctor needs to be consulted.
Using Alvíss AI built on Bayesian Deep Probabilistic Programming, models that were both accurate and uncertainty-aware were developed and tested for the identification of skin diseases.
Alvíss AI requires much less data and extracts more information per data observation, contributing to better uncertainty estimation. This is a major improvement in the quality of the actual suggested diagnostic results compared to results based only on maximum likelihood where overconfident predictions are abundant.