Eye Movements Could Help in Diagnosis of Neurological Disorders
From: USC News - 08/30/2012
By: Katie Dunham 

University of Southern California (USC) researchers have developed a method
for detecting several neurological disorders by studying a person's eye
movements. Attention Deficit Hyperactivity Disorder (ADHD), Fetal Alcohol
Spectrum Disorder (FASD), and Parkinson's disease each involve ocular control
and attention dysfunctions, which can be identified through an evaluation of
how people move their eyes while they watch TV. As part of the study,
participants were asked to watch TV clips for 20 minutes while their eye
movements were recorded. Their eye-tracking data then was combined with
eye-tracking data from healthy volunteers, and a computational model of
visual attention extracted 224 quantitative features. The researchers then
used new machine-learning techniques to identify critical features that stood
out in ADHD, FASD, and Parkinson's disease patients. The technique was able
to identify older adults with Parkinson's disease with 89.6 percent accuracy,
and children with either ADHD or FASD with 77.3 percent accuracy. The
researches say their method provides an easily deployed, low-cost,
high-throughput screening tool. "For the first time, we can actually decode a
person’s neurological state from their everyday behavior, without having to
subject them to difficult or time-consuming tests," says USC professor
Laurent Itti. 

Read the entire article at:
http://news.usc.edu/#!/article/40778/eye-movements-could-help-in-diagnosis-of-neurological-disorders/

Links:
High-throughput classification of clinical populations from natural viewing eye movements 
http://www.springerlink.com/content/65105l60761745w1
http://www.springerlink.com/content/65105l60761745w1/fulltext.pdf

Laurent Itti
http://ilab.usc.edu/itti/