Explainable AI for Discovering Disease Biomarkers: A Survey

Artificial intelligence (AI) and machine learning (ML) have found its applications in biomedical research for a long time. As these emerging technologies proceed to evolve, their utilization in different fields also adapts, including the field of biomedicine. Explainable AI (XAI) is a hot topic nowadays, and its different aspects and use cases are being explored more and more. Biomedicine is one of those disciplines where these tools and methods find their employment. More specifically, XAI has been found very significant in uncovering genetic and molecular biomarkers across various diseases. Thus, it is facilitating enhanced diagnostic and therapeutic strategies for a broad spectrum of diseases—from cancers to neurological and infectious illnesses. In this paper, we are going to give an overview of some state-of-the-art use cases of diverse XAI techniques in the field of biomarker discovery. Ultimately, it is a survey that should serve as a beacon for researchers and practitioners to further explore this interdisciplinary domain.