Evaluation of Eye-tracking Methods for a Decision Support Application
AbstractEye-tracking is used widely to investigate visual and cognitive processes in the context of electronic medical record systems. We investigated a novel application of eye tracking to collect training data for machine learning-based clinical decision support. Specifically, we recorded the information-seeking behavior of physicians while they used electronic medical records in the context of a specific clinical task. Using data captured by a low-cost eye tracking device, we evaluated the performance of several methods for processing gaze points that were recorded using the device. Our results support the use of a low-cost eye tracking device and relatively simple methods for processing gaze points to record the information-seeking behavior of physicians. The eye-tracking methods and scripts that we developed offer a first step in developing novel uses for eye-tracking for clinical decision support.Lay summaryIn the context of electronic medical record systems, eye-tracking is used extensively to explore visual and cognitive processes. We investigated a novel application of eye tracking to collect training data for machine learning-based clinical decision support. We developed and evaluated several methods for processing gaze points that were recorded using a low-cost eye tracking device. The eye-tracking methods and scripts that we developed offer a first step in developing novel uses for eye-tracking for clinical decision support.