Harnessing machine learning to support evidence-based medicine: A pragmatic reconciliation framework

Abstract

Decision making is a central activity in all clinical professions. Clinical decisions bear wellbeing and economic risks and consequences for patients, families, employers, and national economies. Thus, clinicians should employ sound scientific knowledge to promote optimum decision outcomes. Evidence-based medicine organizes clinical decision-making activities with philosophical, ethical, and methodological foundations to ensure accessibility to the best scientific knowledge to inform clinical decision-making. But high-quality evidence could be lacking or methodically, ethically, or economically unfeasible, compelling clinicians to make risk-bearing decisions without an ideal evidence base. In uncertain clinical conditions, decisions' outcomes are probabilistic and the clinicians’ knowledge about outcomes is limited, which makes the clinical decision in significant need for aid. The modern analytical approaches that abstract knowledge from data such as machine learning provide an unprecedented opportunity to address some of the evidence-based medicine challenges. However, despite the plethora of literature that proposes the machine learning as a remedy to evidence-based medicine challenges, there is no serious literature-based framework that helps guide the reconciliation between evidence-based medicine and machine learning paradigms. This study proposes a pragmatic reconciliation framework that guides the paradigms reconciliation agenda. This paper is among the first to present a literature-based framework to guide the reconciliation agenda and is thought to pave the way towards the future research quest.