I-Corps: Developing an eye tracking algorithm that will assess the changes in depression, anxiety and other neurological insults

Stanford University, Stanford CA

Investigators

Abstract

The broader impact/commercial potential of this I-Corps project is the development of a comprehensive communication-detection solution using eye tracking for patients with communication loss due to neurological insults like stroke. Communication loss also can be accompanied by loss of motor control, followed by decrease in social interaction and depression. Such patients are at high risk for another stroke or other neurological insults. There is no single solution available to the speech pathologists who use paper-pencil tests as standard practice for diagnosis and rehabilitation. There is also no monitoring or mental health help provided at follow-up despite the fact that these communication disorders in older adults are a chronic condition and require any solution to be continuously adapted to their needs. The intended primary customers are speech pathologists who can remotely access the content and upload and monitor progress continuously. The goal of the proposed technology is to use eye tracking to provide a communication platform between caregiver and patient, increasing social interaction and reducing depression. In addition, the proposed eye-tracking algorithm may provide mental and physical progress to the provider as well as detect the next stroke. Future iterations may reduce the $4 billion healthcare costs per year in this population. This I-Corps project is based on the development of an eye tracking algorithm to assess the changes in depression, anxiety, and other neurological insults. The proposed technology utilizes existing eye tracking technology available in the Apple® AR kit to optimize the capture and delivery of standardized stimuli (pictures and words) digitally to another device, and use the metrics for the detection algorithm. This proposed method may be optimized for two environmental scenarios: acute in-patient settings and outpatient/in-home settings. A prototype is under development to test this method for detection and tracking of user gaze vectors for object recognition for clinical stimuli generated individually for each patient. Previously, eye tracking technology has been tested for detection of traumatic brain injury. The goal is to fuse gaze information with all other available sources of data (stimuli presented, information from previous visits including baseline results, self-report and guided assessments, and expert clinical feedback). Fused features will then be used to recognize a required “event/state of interest" defined as any event or condition that is used for prediction in the decision system. Pilot and patient data using this technology may provide a comprehensive method for speech pathologists that allows communication, monitoring and detection. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.