The researchers at the University of
Southern California have revealed that certain neurological conditions can be
easily detected by monitoring a patient's eye movement.
Neurological conditions such as Attention Deficit Hyperactivity
Disorder, Fetal Alcohol Spectrum Disorder and Parkinson's disease involve
impairment in ocular control and attention dysfunctions. It is therefore possible
to identify these diseases by carefully watching how patients move their eyes
while watching television programs.
According to the researchers, natural
attention and the eye movements of the patients are akin to a drop of saliva.
It contains a biometric signature of a person and can indicate the state of
The conventional methods of diagnosis for
these diseases include clinical evaluation, neuroimaging and structured
behavioural tasks. These diagnostic methods are costly, involve lot of labor
and are heavily dependant on the patient's ability to follow instructions.
On the other hand, this new method is a
cost-effective screening tool that can be easily put to use in the young and
the old alike. It has been designed by Po-He Tseng, a doctoral student, and
Professor Laurent Itti of the Department of Computer Science at the USC Viterbi
School of Engineering and their collaborators at Queen's University, Canada.
During the study, participants were told
to "watch and enjoy" television clips for a period of 20 minutes. During this
time, their eye movements were recorded.
Eye-tracking data was combined with
normative eye-tracking data, along with a computational model of visual
attention, to extract 224 quantitative features that allowed the usage of new
techniques to differentiate key features that separated patients from control
Eye movement data from about 108
subjects were retrieved which allowed the researchers to identify older adults
suffering from Parkinson's with 89.6 percent accuracy and those with ADHD or
FASD with 77.3 percent accuracy.
It is, indeed, for the first time that
the neurological state of a person has been evaluated using eye movements. This
is a welcome development as it is possible to diagnose the problem subjecting them
to time-consuming and difficult tests.
This research work has been published in the Journal of Neurology.