Feature analyses of pupil light reflex to chromatic stimuli in Alzheimer's patients

The possibility of detecting Alzheimer's Disease (AD) was examined using the pupil light reflex (PLRs). PLRs for chromatic stimuli were recorded for 14 AD patients and 21 healthy participants, and some features of the waveform shapes in their PLRs were extracted. Though there were no significant differences in the features between the two participant groups, most healthy participants were predicted to be so by using a random forest classification technique on the features. Classification performance of the data sets, which consisted of responses to chromatic stimuli, were studied and the possibility of detection was examined.