Autism Spectrum Disorder Detection by Hybrid Convolutional Recurrent Neural Networks from Structural and Resting State Functional MRI Images
This study aims to increase the accuracy of autism spectrum disorder (ASD) diagnosis based on cognitive and behavioral phenotypes through multiple neuroimaging modalities. We apply machine learning (ML) algorithms to classify ASD patients and healthy control (HC) participants using structural magnet...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Hindawi Limited
2023-01-01
|
| Series: | Autism Research and Treatment |
| Online Access: | http://dx.doi.org/10.1155/2023/4136087 |
| _version_ | 1797374415494709248 |
|---|---|
| author | Emel Koc Habil Kalkan Semih Bilgen |
| author_facet | Emel Koc Habil Kalkan Semih Bilgen |
| author_sort | Emel Koc |
| collection | DOAJ |
| description | This study aims to increase the accuracy of autism spectrum disorder (ASD) diagnosis based on cognitive and behavioral phenotypes through multiple neuroimaging modalities. We apply machine learning (ML) algorithms to classify ASD patients and healthy control (HC) participants using structural magnetic resonance imaging (s-MRI) together with resting state functional MRI (rs-f-MRI and f-MRI) data from the large multisite data repository ABIDE (autism brain imaging data exchange) and identify important brain connectivity features. The 2D f-MRI images were converted into 3D s-MRI images, and datasets were preprocessed using the Montreal Neurological Institute (MNI) atlas. The data were then denoised to remove any confounding factors. We show, by using three fusion strategies such as early fusion, late fusion, and cross fusion, that, in this implementation, hybrid convolutional recurrent neural networks achieve better performance in comparison to either convolutional neural networks (CNNs) or recurrent neural networks (RNNs). The proposed model classifies subjects as autistic or not according to how functional and anatomical connectivity metrics provide an overall diagnosis based on the autism diagnostic observation schedule (ADOS) standard. Our hybrid network achieved an accuracy of 96% by fusing s-MRI and f-MRI together, which outperforms the methods used in previous studies. |
| first_indexed | 2024-03-08T19:05:31Z |
| format | Article |
| id | doaj.art-27a24909a5f340fb8ab8c7fc41e1ee38 |
| institution | Directory Open Access Journal |
| issn | 2090-1933 |
| language | English |
| last_indexed | 2024-03-08T19:05:31Z |
| publishDate | 2023-01-01 |
| publisher | Hindawi Limited |
| record_format | Article |
| series | Autism Research and Treatment |
| spelling | doaj.art-27a24909a5f340fb8ab8c7fc41e1ee382023-12-28T00:00:15ZengHindawi LimitedAutism Research and Treatment2090-19332023-01-01202310.1155/2023/4136087Autism Spectrum Disorder Detection by Hybrid Convolutional Recurrent Neural Networks from Structural and Resting State Functional MRI ImagesEmel Koc0Habil Kalkan1Semih Bilgen2Istanbul Okan UniversityGebze Technical UniversityIstanbul Okan UniversityThis study aims to increase the accuracy of autism spectrum disorder (ASD) diagnosis based on cognitive and behavioral phenotypes through multiple neuroimaging modalities. We apply machine learning (ML) algorithms to classify ASD patients and healthy control (HC) participants using structural magnetic resonance imaging (s-MRI) together with resting state functional MRI (rs-f-MRI and f-MRI) data from the large multisite data repository ABIDE (autism brain imaging data exchange) and identify important brain connectivity features. The 2D f-MRI images were converted into 3D s-MRI images, and datasets were preprocessed using the Montreal Neurological Institute (MNI) atlas. The data were then denoised to remove any confounding factors. We show, by using three fusion strategies such as early fusion, late fusion, and cross fusion, that, in this implementation, hybrid convolutional recurrent neural networks achieve better performance in comparison to either convolutional neural networks (CNNs) or recurrent neural networks (RNNs). The proposed model classifies subjects as autistic or not according to how functional and anatomical connectivity metrics provide an overall diagnosis based on the autism diagnostic observation schedule (ADOS) standard. Our hybrid network achieved an accuracy of 96% by fusing s-MRI and f-MRI together, which outperforms the methods used in previous studies.http://dx.doi.org/10.1155/2023/4136087 |
| spellingShingle | Emel Koc Habil Kalkan Semih Bilgen Autism Spectrum Disorder Detection by Hybrid Convolutional Recurrent Neural Networks from Structural and Resting State Functional MRI Images Autism Research and Treatment |
| title | Autism Spectrum Disorder Detection by Hybrid Convolutional Recurrent Neural Networks from Structural and Resting State Functional MRI Images |
| title_full | Autism Spectrum Disorder Detection by Hybrid Convolutional Recurrent Neural Networks from Structural and Resting State Functional MRI Images |
| title_fullStr | Autism Spectrum Disorder Detection by Hybrid Convolutional Recurrent Neural Networks from Structural and Resting State Functional MRI Images |
| title_full_unstemmed | Autism Spectrum Disorder Detection by Hybrid Convolutional Recurrent Neural Networks from Structural and Resting State Functional MRI Images |
| title_short | Autism Spectrum Disorder Detection by Hybrid Convolutional Recurrent Neural Networks from Structural and Resting State Functional MRI Images |
| title_sort | autism spectrum disorder detection by hybrid convolutional recurrent neural networks from structural and resting state functional mri images |
| url | http://dx.doi.org/10.1155/2023/4136087 |
| work_keys_str_mv | AT emelkoc autismspectrumdisorderdetectionbyhybridconvolutionalrecurrentneuralnetworksfromstructuralandrestingstatefunctionalmriimages AT habilkalkan autismspectrumdisorderdetectionbyhybridconvolutionalrecurrentneuralnetworksfromstructuralandrestingstatefunctionalmriimages AT semihbilgen autismspectrumdisorderdetectionbyhybridconvolutionalrecurrentneuralnetworksfromstructuralandrestingstatefunctionalmriimages |