The SDMT test can be used to measure cognitive decline in patients with multiple sclerosis (MS) However, the usual test requires manual effort, which in MS patients may be compromised, and so interfere with the test.  Our collaborator, Dr. Anthony Feinstein, developed a computer-based test that did not require physical movement, but still needed a human operator to advance the test, described here. In this project, we developed a hands-free fully automated SDMT test that uses voice recognition to determine accuracy of the responses, and when to advance the screen.

In collaboration with several clinics, we are now confirming the validity of this method, as wel as establishing clinical norms for diagnosis.

Collaborators: Dr. Anthony Feinstein, Sunnybrook Hospital and the University of Toronto

• Viral Patel, Sunnybrook

Student (graduated, but still collaborating): Lingkai Shen

In a different collaboration, with Professor Moshe Eizenman, we sought to develop an eye tracking system that could work on a mobile device.  The long-term goal was to use mobile eye tracking for various cognitive, physiological and mental health screening and measurement, which Professor Eizenman continues. 

Collaborator: Professor Moshe Eizenman

Graduate Students: Dr. Braiden Brousseau and Soumil Chugh

Publications

• V. Patel, L. Shen, J.Rose, and A. Feinstein, "Taking the tester out of the SDMT: A proof of concept fully automated approach to assessing processing speed in people with MS" in Multiple Sclerosis Journal, 2019;25(11):1506-1513. 
• B. Brousseau, J. Rose and M. Eizenman, "Accurate Model-Based Point of Gaze Estimation on Mobile Devices," Vision 2018, Vol. 2 No. 3, p. 35.
• B. Brousseau, J.Rose, M. Eizenman, "Hybrid Eye-Tracking on a Smartphone with CNN Feature Extraction and an Infrared 3D Model," Sensors 2020, 20(2), 543;
• S. Chugh, B. Brousseau, J. Rose and M. Eizenman, "Detection and Correspondence Matching of Corneal Reflections for Eye Tracking Using Deep Learning," 2020 25th International Conference on Pattern Recognition (ICPR), 2021, pp. 2210-2217, doi: 10.1109/ICPR48806.2021.9412066.