Selecting Interpretability Techniques for Healthcare Machine Learning models

Abstract: In healthcare there is a pursuit for employing interpretable algorithms to assist healthcare professionals in several decision scenarios. Following the Predictive, Descriptive and Relevant (PDR) framework, the definition of interpretable machine learning as a machine-learning model that explicitly and in a simple frame determines relationships either contained in data or learned by the model that are relevant for its functioning and the categorization of models by post-hoc, acquiring interpretability after training, or model-based, being intrinsically embedded in the algorithm design. We overview a selection of eight algorithms, both post-hoc and model-based, that can be used for such purposes. As the capabilities of machine learning models evolve, so does the urgency to understand the decision-making processes within them. This pursuit has led to an array of different related terms with various definitions depending on the authors and the context, such as explainability, transparency, intelligibility, etc. However, for the purpose of this article, we will adopt the definition of interpretability provided in the Predictive, Descriptive, and Relevant (PDR) framework. This framework defines interpretable machine learning as "the extraction of relevant knowledge from a machine-learning model concerning relationships either contained in data or learned by the model" (1). This definition encloses many processes that can be categorized as interpretable and that vary drastically depending on the type of problem that is trying to be solved.