A Deep Learning Approach to Predict Autism Spectrum Disorder Using Multisite Resting-State fMRI
Autism spectrum disorder (ASD) is a complex and degenerative neuro-developmental disorder. Most of the existing methods utilize functional magnetic resonance imaging (fMRI) to detect ASD with a very limited dataset which provides high accuracy but results in poor generalization. To overcome this lim...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-04-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/11/8/3636 |
| _version_ | 1827694831472738304 |
|---|---|
| author | Faria Zarin Subah Kaushik Deb Pranab Kumar Dhar Takeshi Koshiba |
| author_facet | Faria Zarin Subah Kaushik Deb Pranab Kumar Dhar Takeshi Koshiba |
| author_sort | Faria Zarin Subah |
| collection | DOAJ |
| description | Autism spectrum disorder (ASD) is a complex and degenerative neuro-developmental disorder. Most of the existing methods utilize functional magnetic resonance imaging (fMRI) to detect ASD with a very limited dataset which provides high accuracy but results in poor generalization. To overcome this limitation and to enhance the performance of the automated autism diagnosis model, in this paper, we propose an ASD detection model using functional connectivity features of resting-state fMRI data. Our proposed model utilizes two commonly used brain atlases, Craddock 200 (CC200) and Automated Anatomical Labelling (AAL), and two rarely used atlases Bootstrap Analysis of Stable Clusters (BASC) and Power. A deep neural network (DNN) classifier is used to perform the classification task. Simulation results indicate that the proposed model outperforms state-of-the-art methods in terms of accuracy. The mean accuracy of the proposed model was 88%, whereas the mean accuracy of the state-of-the-art methods ranged from 67% to 85%. The sensitivity, F1-score, and area under receiver operating characteristic curve (AUC) score of the proposed model were 90%, 87%, and 96%, respectively. Comparative analysis on various scoring strategies show the superiority of BASC atlas over other aforementioned atlases in classifying ASD and control. |
| first_indexed | 2024-03-10T12:13:15Z |
| format | Article |
| id | doaj.art-336d64cecb5f4ceb9ee20e3955254438 |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-03-10T12:13:15Z |
| publishDate | 2021-04-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-336d64cecb5f4ceb9ee20e39552544382023-11-21T16:03:05ZengMDPI AGApplied Sciences2076-34172021-04-01118363610.3390/app11083636A Deep Learning Approach to Predict Autism Spectrum Disorder Using Multisite Resting-State fMRIFaria Zarin Subah0Kaushik Deb1Pranab Kumar Dhar2Takeshi Koshiba3Department of Computer Science and Engineering, Chittagong University of Engineering & Technology, Chattogram 4349, BangladeshDepartment of Computer Science and Engineering, Chittagong University of Engineering & Technology, Chattogram 4349, BangladeshDepartment of Computer Science and Engineering, Chittagong University of Engineering & Technology, Chattogram 4349, BangladeshFaculty of Education and Integrated Arts and Sciences, Waseda University, 1-6-1 Nishiwaseda, Shinjuku-ku, Tokyo 169-8050, JapanAutism spectrum disorder (ASD) is a complex and degenerative neuro-developmental disorder. Most of the existing methods utilize functional magnetic resonance imaging (fMRI) to detect ASD with a very limited dataset which provides high accuracy but results in poor generalization. To overcome this limitation and to enhance the performance of the automated autism diagnosis model, in this paper, we propose an ASD detection model using functional connectivity features of resting-state fMRI data. Our proposed model utilizes two commonly used brain atlases, Craddock 200 (CC200) and Automated Anatomical Labelling (AAL), and two rarely used atlases Bootstrap Analysis of Stable Clusters (BASC) and Power. A deep neural network (DNN) classifier is used to perform the classification task. Simulation results indicate that the proposed model outperforms state-of-the-art methods in terms of accuracy. The mean accuracy of the proposed model was 88%, whereas the mean accuracy of the state-of-the-art methods ranged from 67% to 85%. The sensitivity, F1-score, and area under receiver operating characteristic curve (AUC) score of the proposed model were 90%, 87%, and 96%, respectively. Comparative analysis on various scoring strategies show the superiority of BASC atlas over other aforementioned atlases in classifying ASD and control.https://www.mdpi.com/2076-3417/11/8/3636autism spectrum disorderresting-state fMRIpredefined brain atlasABIDEfunctional connectivityconnectivity matrix |
| spellingShingle | Faria Zarin Subah Kaushik Deb Pranab Kumar Dhar Takeshi Koshiba A Deep Learning Approach to Predict Autism Spectrum Disorder Using Multisite Resting-State fMRI Applied Sciences autism spectrum disorder resting-state fMRI predefined brain atlas ABIDE functional connectivity connectivity matrix |
| title | A Deep Learning Approach to Predict Autism Spectrum Disorder Using Multisite Resting-State fMRI |
| title_full | A Deep Learning Approach to Predict Autism Spectrum Disorder Using Multisite Resting-State fMRI |
| title_fullStr | A Deep Learning Approach to Predict Autism Spectrum Disorder Using Multisite Resting-State fMRI |
| title_full_unstemmed | A Deep Learning Approach to Predict Autism Spectrum Disorder Using Multisite Resting-State fMRI |
| title_short | A Deep Learning Approach to Predict Autism Spectrum Disorder Using Multisite Resting-State fMRI |
| title_sort | deep learning approach to predict autism spectrum disorder using multisite resting state fmri |
| topic | autism spectrum disorder resting-state fMRI predefined brain atlas ABIDE functional connectivity connectivity matrix |
| url | https://www.mdpi.com/2076-3417/11/8/3636 |
| work_keys_str_mv | AT fariazarinsubah adeeplearningapproachtopredictautismspectrumdisorderusingmultisiterestingstatefmri AT kaushikdeb adeeplearningapproachtopredictautismspectrumdisorderusingmultisiterestingstatefmri AT pranabkumardhar adeeplearningapproachtopredictautismspectrumdisorderusingmultisiterestingstatefmri AT takeshikoshiba adeeplearningapproachtopredictautismspectrumdisorderusingmultisiterestingstatefmri AT fariazarinsubah deeplearningapproachtopredictautismspectrumdisorderusingmultisiterestingstatefmri AT kaushikdeb deeplearningapproachtopredictautismspectrumdisorderusingmultisiterestingstatefmri AT pranabkumardhar deeplearningapproachtopredictautismspectrumdisorderusingmultisiterestingstatefmri AT takeshikoshiba deeplearningapproachtopredictautismspectrumdisorderusingmultisiterestingstatefmri |