Application of machine learning for the prediction of diabetic foot ulceration in individuals at high risk of foot ulceration

The presentation is going to be focused on the application of machine learning techniques for the prediction of diabetic foot ulceration. Traditionally, statistical regression techniques have been used to study predictors for diabetic foot ulceration. However, these methods may miss to capture complex nonlinear relationships and interdependence among predictors. To address this, we selected features from abroad range of potential predictors, including biomechanical, clinical and behavioral variables and we implemented machine learning for the prediction of diabetic foot ulceration, focusing on tree-ensemble models.