A significant body of evidence suggests that eye movement behavior is a powerful window into the brain and cognitive function, and that eye movement behavior can be leveraged for performing neuropsychological assessments. This is because a number of developmental, aging, and injury-related diseases and syndromes have been proved to cause measurable changes in eye movement behavior.
However, the methods supporting these findings all suffer from significant limitations. 
Technology Overview
The proposed method addresses these issues directly by an approach similar to that used by cardiologists to evaluate heart health based on heartbeat recordings.
EMBED works by presenting patients with engaging dynamic stimuli movies or television shows. 
Their eye movement behavior is recorded during the content screening. 
The collected gaze data is then aligned with a map of values computed from each frame of the video stimulus that describes the joint level of bottom-up (pictorial or image feature level content, color, edges) and top-down (object-contextual appropriateness) “salience” or “interest” of each pixel. 
The resultant salience map aligned time series can then be analyzed using (Digital Fluorescein Angiography) DFA.
- Easily identifies shifts in cognitive flexibility relative to baseline as potential indicators of injury or disease state change
- Increased diagnostic sensitivity relative to existing methods by capturing disease-related effects on eye movement behavior
- Requires no receptive or productive language capacity or gross motor skill from the patient to be deployed
- Use of unconstrained natural viewing behavior prevents possible malingering by patients
- Easily calibrated to detect anomalies relative to observed individual differences in baseline performance
- Detection of concussion or traumatic brain injury 
- Dementia/Alzheimer’s status monitoring 
- License
- Partnering
- Research Collaboration
Patent Information:
For Information, Contact:
Mark Saulich
Associate Director of Commercialization
Northeastern University
Dylan Rose
Peter Bex