# A systematic review of eye-tracking technology in electrocardiogram interpretation research

## TL;DR

Eye-tracking helps assess how experts and novices interpret ECGs, revealing differences in visual behavior and diagnostic performance.

## Contribution

This systematic review identifies gaze metrics as potential indicators of ECG interpretation expertise and educational potential.

## Key findings

- Experts show faster, more targeted fixation on critical ECG leads compared to novices.
- Eye-tracking metrics like fixation duration and scan-path efficiency correlate with diagnostic accuracy.
- Educational tools using gaze modeling show promise but require further validation.

## Abstract

Eye-tracking technology provides an objective method for analyzing visual behavior during diagnostic tasks such as electrocardiogram (ECG) interpretation. As a high-stakes and visually complex clinical skill, ECG interpretation benefits from technologies that differentiate experts from novices via gaze patterns. We systematically reviewed eye-tracking in ECG research to examine links between gaze metrics and diagnostic performance and to appraise educational applications.

Following PRISMA (PROSPERO CRD420251052940), we searched PubMed, Embase, Web of Science, Scopus, ERIC, IEEE Xplore, and PubPsych databases to September 2025. Eligible studies used eye-tracking during ECG interpretation and reported visual behavior or diagnostic outcomes. Two independent reviewers performed the study selection, data extraction, and quality assessment using the Jadad and Newcastle–Ottawa scales. Given heterogeneity, we conducted a narrative synthesis.

Nineteen studies (573 participants) were included, mainly involving students or early-career clinicians. Common metrics included fixation duration, counts, time to first fixation, and scan-path efficiency. Experts showed faster, more targeted fixation on critical leads (V1, V2, and II), shorter interpretation times, and higher accuracy. Novices exhibited scattered visual behavior and delayed attention to diagnostically relevant regions. Eye movement data were informative of underlying cognitive processes, including the selection, organization, and integration of ECG features. Additional factors, such as presentation format, individual experience, metacognitive strategies, and emotional responses, may moderate gaze behavior, although these were underexplored. Educational interventions leveraging expert gaze modeling or structured checklists showed mixed but promising outcomes.

Eye-tracking differentiates expertise in ECG interpretation and may inform diagnostic reasoning, but prospective validation is needed. Gaze metrics are candidate indicators of performance and a foundation for feedback-based educational tools. Standardized protocols, larger multi-site studies, and inclusion of diverse learners are required to realize the potential of eye-tracking in ECG training and assessment.

The online version contains supplementary material available at 10.1186/s40001-025-03635-8.