Assessment of electrocardiogram visual interpretation strategy based on scanpath analysis

Area of Science:

• Biomedical Engineering
• Medical Informatics
• Cognitive Science

Background:

• Automated electrocardiogram (ECG) interpretation systems aim to mimic human expert reasoning.
• The precise visual interpretation strategies employed by human experts remain largely uncharacterized.
• Understanding these strategies is crucial for developing more sophisticated and accurate automated diagnostic tools.

Purpose of the Study:

• To investigate the visual interpretation process of ECGs by human experts.
• To identify objective markers of diagnostic expertise using eye-tracking data.
• To uncover the underlying principles of ECG interpretation strategies.

Main Methods:

• Utilized eye-tracking technology to capture eyeglobe trajectories of participants with varying expertise during ECG interpretation tasks.
• Analyzed scanpath data in relation to the visual features of ECG traces.
• Correlated oculomotoric patterns with diagnostic performance and expertise levels.

Main Results:

• Eye-tracking data revealed distinct patterns in how experts visually scan and process ECG information.
• Quantitative analysis provided insights into the distribution of diagnostic data on ECG traces.
• Identified specific eyetrack parameters that correlate with an individual's ECG interpretation skill.
• Scanpath data offers an objective assessment, unaffected by memory or verbalization biases.

Conclusions:

• Eye-tracking provides a novel and objective method for studying human ECG interpretation strategies.
• Oculomotoric habits and visual search patterns are valuable indicators of diagnostic expertise in ECG analysis.
• This research lays the groundwork for more effective automated ECG interpretation systems informed by human visual cognition.