Explainable Artificial Intelligence in the Medical Domain: A Systematic Review

The applications of Artificial Intelligence (AI) and Machine Learning (ML) techniques in different medical fields is rapidly growing. AI holds great promise in terms of beneficial, accurate and effective preventive and curative interventions. At the same time, there is also concerns regarding potential risks, harm and trust issues arising from the opacity of some AI algorithms because of their un-explainability. Overall, how can the decisions from these AI-based systems be trusted if the decision-making logic cannot be properly explained? Explainable Artificial Intelligence (XAI) tries to shed light to these questions. We study the recent development on this topic within the medical domain. The objective of this study is to provide a systematic review of the methods and techniques of explainable AI within the medical domain as observed within the literature while identifying future research opportunities.