Identifying visual search patterns in eye gaze data; gaining insights into physician visual workflow.

Area of Science:

• Health Informatics
• Human-Computer Interaction
• Medical Education

Background:

• Health information technologies (HIT) are increasingly integrated into physician workflows.
• Understanding physician interaction with HIT, particularly information search behavior, is crucial.
• Traditional analysis of eye-tracking data uses summative metrics, potentially oversimplifying complex visual search.

Purpose of the Study:

• To develop and demonstrate an algorithmic approach for identifying distinct visual search patterns in physicians.
• To evaluate the efficacy of this new approach compared to conventional methods.
• To gain insights into physician information-seeking behaviors within simulated healthcare environments.

Main Methods:

• An algorithmic method was developed to identify and categorize visual search patterns.
• Physician eye-tracking data was collected using a simulated prototype emergency department patient tracking system.
• The proposed algorithmic patterns were compared against first-order transition metrics.

Main Results:

• The algorithm successfully identified common visual search patterns among physicians.
• Comparison with first-order transitions highlighted differences in pattern identification.
• Initial evaluation provided insights into the benefits and limitations of the algorithmic approach.

Conclusions:

• The proposed algorithmic approach offers a more nuanced understanding of physician visual search patterns in HIT.
• This method provides valuable data for optimizing the design and usability of health information technologies.
• Further research is warranted to explore the full potential and applications of this analytical technique.