Estimation of sleepiness using pupillary response and its frequency components

Area of Science:

• Ophthalmology
• Neuroscience
• Sleep Science

Background:

• Pupillary response is a potential indicator of sleepiness.
• The reliability of pupillary response as a sleepiness index requires further validation.
• Blinks can influence pupillary signals, potentially confounding sleepiness assessments.

Purpose of the Study:

• To examine the influence of blinks on pupillary unrest index (PUI) and power spectrum density (PSD) during sleepiness testing.
• To develop and assess pupillary indices derived from PSD for classifying subjective sleepiness.
• To elucidate the relationship between blinks, pupillary instability, and subjective sleepiness.

Main Methods:

• Analysis of pupillary changes during a sleepiness test, focusing on blink influence.
• Extraction of four frequency components from power spectrum density (PSD) of compensated pupillary changes.
• Application of discriminant analysis for two-class classification of subjective sleepiness using derived pupillary indices.

Main Results:

• Differences in the contribution ratios of the four factor components were observed between distinct subjective sleepiness states.
• The study identified specific pupillary signal characteristics associated with different levels of sleepiness.
• A causal path diagram illustrated the interplay between blinks, pupillary indices, and subjective sleepiness.

Conclusions:

• Blinks significantly influence pupillary measures used in sleepiness assessment.
• Derived pupillary indices show potential for differentiating subjective sleepiness levels.
• Further research is needed to refine pupillary-based sleepiness detection methods, accounting for blink artifacts.