Mapping Algorithmic Bias in AI-Powered ECG Interpretation Across the AI Lifecycle: A Scoping Review Protocol (Preprint)

BACKGROUND Artificial intelligence (AI)–powered analysis of electrocardiograms (ECGs) is reshaping cardiac diagnostics, offering faster and often more accurate detection of conditions such as arrhythmias and heart failure. However, growing evidence suggests that algorithmic bias, defined as performance disparities across patient subgroups, may undermine diagnostic equity. These biases can emerge at any stage of the AI lifecycle, including data collection, model development, evaluation, deployment, and clinical use. If unaddressed, they risk exacerbating health disparities, particularly in underrepresented populations and low-resource settings. Early identification and mitigation of such bias are essential to ensuring diagnostic equity. OBJECTIVE This scoping review aims to systematically map the current evidence on algorithmic bias in AI-enabled ECG interpretation. Specifically, we will (1) identify the types and sources of bias reported, (2) assess how performance varies across demographic or geographic subgroups, and (3) document any mitigation strategies applied. Our goal is to clarify how fairness is addressed in this growing field and highlight gaps that remain to be addressed to ensure equitable use of AI in cardiology. METHODS We will conduct a comprehensive literature search across five electronic databases (PubMed, EMBASE, Cochrane CENTRAL, CINAHL, and IEEE Xplore), as well as grey literature sources including preprint servers and clinical trial registries. Eligible studies will include original research (2015–2025) evaluating the performance of AI-based ECG models across different subgroups or reporting on bias mitigation strategies. Two reviewers will independently screen studies, extract data using a standardized form, and resolve disagreements through consensus. The review will follow the PRISMA-ScR reporting framework. RESULTS At the time of publication, study screening had not yet begun. Searches will commence in August 2025, with data extraction anticipated by September 2025. Results will be synthesized narratively and descriptively, and a PRISMA flow diagram will summarize the study selection process. CONCLUSIONS This review will be the first to comprehensively map the landscape of algorithmic bias in AI-powered ECG interpretation. By identifying patterns of inequity and evaluating proposed solutions, it will provide actionable insights for developers, clinicians, and policymakers aiming to promote fairness in AI-enabled cardiac care.