Eye movement behavior identification for Alzheimer's disease diagnosis
We develop a deep-learning approach to differentiate between the eye movement behavior of people with neurodegenerative diseases during reading compared to healthy control subjects. The subjects with and without Alzheimer’s disease read well-defined and previously validated sentences including high-...
Main Author: | |
---|---|
Format: | Article |
Language: | English |
Published: |
IMR (Innovative Medical Research) Press Limited
2018-11-01
|
Series: | Journal of Integrative Neuroscience |
Subjects: | |
Online Access: | https://jin.imrpress.com/fileup/1757-448X/PDF/1546069972927-1824017746.pdf |
id |
doaj-669db48d3f9e448da7ae76e3e5b480ee |
---|---|
record_format |
Article |
spelling |
doaj-669db48d3f9e448da7ae76e3e5b480ee2020-11-25T02:30:47ZengIMR (Innovative Medical Research) Press LimitedJournal of Integrative Neuroscience1757-448X2018-11-0117434935410.31083/j.jin.2018.04.0416Eye movement behavior identification for Alzheimer's disease diagnosisJuan Biondi, Gerardo Fernandez, Silvia Castro, Osvaldo Agamennoni01 Laboratorio de Desarrollo en Neurociencia Cognitiva, Instituto de Investigaciones en Ingeniería Eléctrica (IIIE), Departamento de Ingeniería Eléctrica y de Computadoras (DIEC), Universidad Nacional del Sur (UNS) - CONICET, San Andrés 800 – Bahía Blanca, 8000, Buenos Aires, Argentina;2 Laboratorio de Visualización y Computación Gráfica (VyGLab), Departamento de Ciencias e Ingeniería de la Computación (DCIC), Universidad Nacional del Sur (UNS), San Andrés 800 – Bahía Blanca, 8000, Buenos Aires, Argentina;3 Comisión de Investigaciones Científicas de la Provincia de Buenos Aires (CIC), San Andrés 800 – Bahía Blanca, 8000, Buenos Aires, ArgentinaWe develop a deep-learning approach to differentiate between the eye movement behavior of people with neurodegenerative diseases during reading compared to healthy control subjects. The subjects with and without Alzheimer’s disease read well-defined and previously validated sentences including high- and low-predictable sentences, and proverbs. From these eye-tracking data trial-wise information is derived consisting of descriptors that capture the reading behavior of the subjects. With this information a set of denoising sparse-autoencoders are trained and a deep neural network is built using the trained autoencoders and a softmax classifier that identifies subjects with Alzheimer’s disease with 89.78% accuracy. The results are very encouraging and show that such models promise to be helpful for understanding the dynamics of eye movement behavior and its relation with underlying neuropsychological processes.https://jin.imrpress.com/fileup/1757-448X/PDF/1546069972927-1824017746.pdf|eye-tracking|deep-learning|alzheimer’s disease|neurodegenerative diseases|eye movement behavior|neuropsychological processes |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Juan Biondi, Gerardo Fernandez, Silvia Castro, Osvaldo Agamennoni |
spellingShingle |
Juan Biondi, Gerardo Fernandez, Silvia Castro, Osvaldo Agamennoni Eye movement behavior identification for Alzheimer's disease diagnosis Journal of Integrative Neuroscience |eye-tracking|deep-learning|alzheimer’s disease|neurodegenerative diseases|eye movement behavior|neuropsychological processes |
author_facet |
Juan Biondi, Gerardo Fernandez, Silvia Castro, Osvaldo Agamennoni |
author_sort |
Juan Biondi, Gerardo Fernandez, Silvia Castro, Osvaldo Agamennoni |
title |
Eye movement behavior identification for Alzheimer's disease diagnosis |
title_short |
Eye movement behavior identification for Alzheimer's disease diagnosis |
title_full |
Eye movement behavior identification for Alzheimer's disease diagnosis |
title_fullStr |
Eye movement behavior identification for Alzheimer's disease diagnosis |
title_full_unstemmed |
Eye movement behavior identification for Alzheimer's disease diagnosis |
title_sort |
eye movement behavior identification for alzheimer's disease diagnosis |
publisher |
IMR (Innovative Medical Research) Press Limited |
series |
Journal of Integrative Neuroscience |
issn |
1757-448X |
publishDate |
2018-11-01 |
description |
We develop a deep-learning approach to differentiate between the eye movement behavior of people with neurodegenerative diseases during reading compared to healthy control subjects. The subjects with and without Alzheimer’s disease read well-defined and previously validated sentences including high- and low-predictable sentences, and proverbs. From these eye-tracking data trial-wise information is derived consisting of descriptors that capture the reading behavior of the subjects. With this information a set of denoising sparse-autoencoders are trained and a deep neural network is built using the trained autoencoders and a softmax classifier that identifies subjects with Alzheimer’s disease with 89.78% accuracy. The results are very encouraging and show that such models promise to be helpful for understanding the dynamics of eye movement behavior and its relation with underlying neuropsychological processes. |
topic |
|eye-tracking|deep-learning|alzheimer’s disease|neurodegenerative diseases|eye movement behavior|neuropsychological processes |
url |
https://jin.imrpress.com/fileup/1757-448X/PDF/1546069972927-1824017746.pdf |
work_keys_str_mv |
AT juanbiondigerardofernandezsilviacastroosvaldoagamennoni eyemovementbehavioridentificationforalzheimersdiseasediagnosis |
_version_ |
1724827908133879808 |