Eye Movement Analysis For Depression Detection

Sharifa Alghowinem, Roland GOECKE, Michael WAGNER, Gordon Parker, Michael Breakspear

Research output: A Conference proceeding or a Chapter in BookConference contribution

34 Citations (Scopus)
2 Downloads (Pure)

Abstract

Depression is a common and disabling mental health disorder, which impacts not only on the sufferer but also on their families, friends and the economy overall. Despite its high prevalence, current diagnosis relies almost exclusively on patient self-report and clinical opinion, leading to a number of subjective biases. Our aim is to develop an objective affective sensing system that supports clinicians in their diagnosis and monitoring of clinical depression. In this paper, we analyse the performance of eye movement features extracted from face videos using Active Appearance Models for a binary classification task (depressed vs. non-depressed). We find that eye movement low-level features gave 70% accuracy using a hybrid classifier of Gaussian Mixture Models and Support Vector Machines, and 75% accuracy when using statistical measures with SVM classifiers over the entire interview. We also investigate differences while expressing positive and negative emotions, as well as the classification performance in gender-dependent versus gender- independent modes. Interestingly, even though the blinking rate was not significantly different between depressed and healthy controls, we find that the average distance between the eyelids (‘eye opening’) was significantly smaller and the average duration of blinks significantly longer in depressed subjects, which might be an indication of fatigue or eye contact avoidance.
Original languageEnglish
Title of host publication2013 IEEE International Conference on Image Processing ICIP2013
EditorsBrian Lovell, David Suter
Place of PublicationMelbourne, Australia
PublisherIEEE, Institute of Electrical and Electronics Engineers
Pages4220-4224
Number of pages5
ISBN (Print)9781479923410
DOIs
Publication statusPublished - 2013
Event2013 IEEE International Conference on Image Processing ICIP2013 - Melbourne, Melbourne, Australia
Duration: 5 Sep 201318 Sep 2013

Conference

Conference2013 IEEE International Conference on Image Processing ICIP2013
CountryAustralia
CityMelbourne
Period5/09/1318/09/13

Fingerprint

Eye movements
Classifiers
Support vector machines
Health
Fatigue of materials
Monitoring

Cite this

Alghowinem, S., GOECKE, R., WAGNER, M., Parker, G., & Breakspear, M. (2013). Eye Movement Analysis For Depression Detection. In B. Lovell, & D. Suter (Eds.), 2013 IEEE International Conference on Image Processing ICIP2013 (pp. 4220-4224). Melbourne, Australia: IEEE, Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/ICIP.2013.6738869
Alghowinem, Sharifa ; GOECKE, Roland ; WAGNER, Michael ; Parker, Gordon ; Breakspear, Michael. / Eye Movement Analysis For Depression Detection. 2013 IEEE International Conference on Image Processing ICIP2013. editor / Brian Lovell ; David Suter. Melbourne, Australia : IEEE, Institute of Electrical and Electronics Engineers, 2013. pp. 4220-4224
@inproceedings{de330b699b844daf92146611596b5990,
title = "Eye Movement Analysis For Depression Detection",
abstract = "Depression is a common and disabling mental health disorder, which impacts not only on the sufferer but also on their families, friends and the economy overall. Despite its high prevalence, current diagnosis relies almost exclusively on patient self-report and clinical opinion, leading to a number of subjective biases. Our aim is to develop an objective affective sensing system that supports clinicians in their diagnosis and monitoring of clinical depression. In this paper, we analyse the performance of eye movement features extracted from face videos using Active Appearance Models for a binary classification task (depressed vs. non-depressed). We find that eye movement low-level features gave 70{\%} accuracy using a hybrid classifier of Gaussian Mixture Models and Support Vector Machines, and 75{\%} accuracy when using statistical measures with SVM classifiers over the entire interview. We also investigate differences while expressing positive and negative emotions, as well as the classification performance in gender-dependent versus gender- independent modes. Interestingly, even though the blinking rate was not significantly different between depressed and healthy controls, we find that the average distance between the eyelids (‘eye opening’) was significantly smaller and the average duration of blinks significantly longer in depressed subjects, which might be an indication of fatigue or eye contact avoidance.",
keywords = "Eye Movement, Active Appearance Model, Affective Sensing, Depression analysis",
author = "Sharifa Alghowinem and Roland GOECKE and Michael WAGNER and Gordon Parker and Michael Breakspear",
year = "2013",
doi = "10.1109/ICIP.2013.6738869",
language = "English",
isbn = "9781479923410",
pages = "4220--4224",
editor = "Brian Lovell and David Suter",
booktitle = "2013 IEEE International Conference on Image Processing ICIP2013",
publisher = "IEEE, Institute of Electrical and Electronics Engineers",
address = "United States",

}

Alghowinem, S, GOECKE, R, WAGNER, M, Parker, G & Breakspear, M 2013, Eye Movement Analysis For Depression Detection. in B Lovell & D Suter (eds), 2013 IEEE International Conference on Image Processing ICIP2013. IEEE, Institute of Electrical and Electronics Engineers, Melbourne, Australia, pp. 4220-4224, 2013 IEEE International Conference on Image Processing ICIP2013, Melbourne, Australia, 5/09/13. https://doi.org/10.1109/ICIP.2013.6738869

Eye Movement Analysis For Depression Detection. / Alghowinem, Sharifa; GOECKE, Roland; WAGNER, Michael; Parker, Gordon; Breakspear, Michael.

2013 IEEE International Conference on Image Processing ICIP2013. ed. / Brian Lovell; David Suter. Melbourne, Australia : IEEE, Institute of Electrical and Electronics Engineers, 2013. p. 4220-4224.

Research output: A Conference proceeding or a Chapter in BookConference contribution

TY - GEN

T1 - Eye Movement Analysis For Depression Detection

AU - Alghowinem, Sharifa

AU - GOECKE, Roland

AU - WAGNER, Michael

AU - Parker, Gordon

AU - Breakspear, Michael

PY - 2013

Y1 - 2013

N2 - Depression is a common and disabling mental health disorder, which impacts not only on the sufferer but also on their families, friends and the economy overall. Despite its high prevalence, current diagnosis relies almost exclusively on patient self-report and clinical opinion, leading to a number of subjective biases. Our aim is to develop an objective affective sensing system that supports clinicians in their diagnosis and monitoring of clinical depression. In this paper, we analyse the performance of eye movement features extracted from face videos using Active Appearance Models for a binary classification task (depressed vs. non-depressed). We find that eye movement low-level features gave 70% accuracy using a hybrid classifier of Gaussian Mixture Models and Support Vector Machines, and 75% accuracy when using statistical measures with SVM classifiers over the entire interview. We also investigate differences while expressing positive and negative emotions, as well as the classification performance in gender-dependent versus gender- independent modes. Interestingly, even though the blinking rate was not significantly different between depressed and healthy controls, we find that the average distance between the eyelids (‘eye opening’) was significantly smaller and the average duration of blinks significantly longer in depressed subjects, which might be an indication of fatigue or eye contact avoidance.

AB - Depression is a common and disabling mental health disorder, which impacts not only on the sufferer but also on their families, friends and the economy overall. Despite its high prevalence, current diagnosis relies almost exclusively on patient self-report and clinical opinion, leading to a number of subjective biases. Our aim is to develop an objective affective sensing system that supports clinicians in their diagnosis and monitoring of clinical depression. In this paper, we analyse the performance of eye movement features extracted from face videos using Active Appearance Models for a binary classification task (depressed vs. non-depressed). We find that eye movement low-level features gave 70% accuracy using a hybrid classifier of Gaussian Mixture Models and Support Vector Machines, and 75% accuracy when using statistical measures with SVM classifiers over the entire interview. We also investigate differences while expressing positive and negative emotions, as well as the classification performance in gender-dependent versus gender- independent modes. Interestingly, even though the blinking rate was not significantly different between depressed and healthy controls, we find that the average distance between the eyelids (‘eye opening’) was significantly smaller and the average duration of blinks significantly longer in depressed subjects, which might be an indication of fatigue or eye contact avoidance.

KW - Eye Movement

KW - Active Appearance Model

KW - Affective Sensing

KW - Depression analysis

U2 - 10.1109/ICIP.2013.6738869

DO - 10.1109/ICIP.2013.6738869

M3 - Conference contribution

SN - 9781479923410

SP - 4220

EP - 4224

BT - 2013 IEEE International Conference on Image Processing ICIP2013

A2 - Lovell, Brian

A2 - Suter, David

PB - IEEE, Institute of Electrical and Electronics Engineers

CY - Melbourne, Australia

ER -

Alghowinem S, GOECKE R, WAGNER M, Parker G, Breakspear M. Eye Movement Analysis For Depression Detection. In Lovell B, Suter D, editors, 2013 IEEE International Conference on Image Processing ICIP2013. Melbourne, Australia: IEEE, Institute of Electrical and Electronics Engineers. 2013. p. 4220-4224 https://doi.org/10.1109/ICIP.2013.6738869