Eye tracking based dyslexia detection using a holistic approach

Abstract A new detection method for cognitive impairments is presented utilizing an eye tracking signals in a text reading test. This research enhances published articles that extract combination of various features. It does so by processing entire eye-tracking records either in time or frequency wh...

Full description

Bibliographic Details
Main Authors: Boris Nerušil, Jaroslav Polec, Juraj Škunda, Juraj Kačur
Format: Article
Language:English
Published: Nature Publishing Group 2021-08-01
Series:Scientific Reports
Online Access:https://doi.org/10.1038/s41598-021-95275-1
id doaj-49cfd8d82d194c6ca096f0c9afd68769
record_format Article
spelling doaj-49cfd8d82d194c6ca096f0c9afd687692021-08-08T11:24:33ZengNature Publishing GroupScientific Reports2045-23222021-08-0111111010.1038/s41598-021-95275-1Eye tracking based dyslexia detection using a holistic approachBoris Nerušil0Jaroslav Polec1Juraj Škunda2Juraj Kačur3Institute of Multimedia ICT, Slovak University of Technology in BratislavaInstitute of Multimedia ICT, Slovak University of Technology in BratislavaInstitute of Multimedia ICT, Slovak University of Technology in BratislavaInstitute of Multimedia ICT, Slovak University of Technology in BratislavaAbstract A new detection method for cognitive impairments is presented utilizing an eye tracking signals in a text reading test. This research enhances published articles that extract combination of various features. It does so by processing entire eye-tracking records either in time or frequency whereas applying only basic signal pre-processing. Such signals were classified as a whole by Convolutional Neural Networks (CNN) that hierarchically extract substantial features scatter either in time or frequency and nonlinearly binds them using machine learning to minimize a detection error. In the experiments we used a 100 fold cross validation and a dataset containing signals of 185 subjects (88 subjects with low risk and 97 subjects with high risk of dyslexia). In a series of experiments it was found that magnitude spectrum based representation of time interpolated eye-tracking signals recorded the best results, i.e. an average accuracy of 96.6% was reached in comparison to 95.6% that is the best published result on the same database. These findings suggest that a holistic approach involving small but complex enough CNNs applied to properly pre-process and expressed signals provides even better results than a combination of meticulously selected well-known features.https://doi.org/10.1038/s41598-021-95275-1
collection DOAJ
language English
format Article
sources DOAJ
author Boris Nerušil
Jaroslav Polec
Juraj Škunda
Juraj Kačur
spellingShingle Boris Nerušil
Jaroslav Polec
Juraj Škunda
Juraj Kačur
Eye tracking based dyslexia detection using a holistic approach
Scientific Reports
author_facet Boris Nerušil
Jaroslav Polec
Juraj Škunda
Juraj Kačur
author_sort Boris Nerušil
title Eye tracking based dyslexia detection using a holistic approach
title_short Eye tracking based dyslexia detection using a holistic approach
title_full Eye tracking based dyslexia detection using a holistic approach
title_fullStr Eye tracking based dyslexia detection using a holistic approach
title_full_unstemmed Eye tracking based dyslexia detection using a holistic approach
title_sort eye tracking based dyslexia detection using a holistic approach
publisher Nature Publishing Group
series Scientific Reports
issn 2045-2322
publishDate 2021-08-01
description Abstract A new detection method for cognitive impairments is presented utilizing an eye tracking signals in a text reading test. This research enhances published articles that extract combination of various features. It does so by processing entire eye-tracking records either in time or frequency whereas applying only basic signal pre-processing. Such signals were classified as a whole by Convolutional Neural Networks (CNN) that hierarchically extract substantial features scatter either in time or frequency and nonlinearly binds them using machine learning to minimize a detection error. In the experiments we used a 100 fold cross validation and a dataset containing signals of 185 subjects (88 subjects with low risk and 97 subjects with high risk of dyslexia). In a series of experiments it was found that magnitude spectrum based representation of time interpolated eye-tracking signals recorded the best results, i.e. an average accuracy of 96.6% was reached in comparison to 95.6% that is the best published result on the same database. These findings suggest that a holistic approach involving small but complex enough CNNs applied to properly pre-process and expressed signals provides even better results than a combination of meticulously selected well-known features.
url https://doi.org/10.1038/s41598-021-95275-1
work_keys_str_mv AT borisnerusil eyetrackingbaseddyslexiadetectionusingaholisticapproach
AT jaroslavpolec eyetrackingbaseddyslexiadetectionusingaholisticapproach
AT jurajskunda eyetrackingbaseddyslexiadetectionusingaholisticapproach
AT jurajkacur eyetrackingbaseddyslexiadetectionusingaholisticapproach
_version_ 1721215940749688832