Applying computer vision techniques to depression symptomatology through eye blink patterns in university students

ABSTRACT

Objective

The increasing rate of depression among university students is a cause of great concern worldwide. With the recent growth in computer vision technology, eye movement features are proving beneficial in the assessment of depression owing to their non-invasiveness. Our objective was to determine the presence of depression through emotional elicitation by studying blink patterns in a student population.

Method

The blink data of 50 university students (26 males, 24 females) from different regions of the country within the age group of 21–26 years were collected using an experimental setup.

Results

Statistical tests on blink data revealed that blink rate changes with changes in emotion from joy to sadness or vice versa, irrespective of one’s depression status. Another explanation is that the blink rates of the healthy and depressed groups differed significantly during sad emotional states. The tests also indicated a half-closed eye state as a possible symptom of depression.

Conclusions

These findings suggest that the prevalence of mental health problems can be detected at an early stage by designing simple non-invasive procedures, such as blink pattern analysis driven by technology and expert advice, which will help prevent the further development of such illnesses.

KEY POINTS

What is already known about this topic:

• (1) Computer vision techniques can detect the face and eye, and eye aspect ratio (EAR) is calculated to determine eye blinks.

• (2) Blink rates vary with different emotions.

• (3) The Dass-21 scale can measure depression, anxiety and stress.

What this topic adds:

• (1) Sufficient literature citing the importance of biomarkers in assessing depression is included.

• (2) Justification of emotional blink characteristics during sad and joy elicited emotions.

• (3) The significance of the half-closed eye state in the assessment of depression was demonstrated using statistical analysis.

KEYWORDS:

• Depression
• face detection
• eye detection
• blink detection
• EAR
• statistical analysis

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The data that support the findings of this study are available upon reasonable request.

Additional information

Funding

The author(s) reported there is no funding associated with the work featured in this article.