Three-Stream Convolutional Neural Network for Depression Detection With Ocular Imaging
Depression is a prevalent and severe mental disorder that significantly affects both mind and body, leading to persistent feelings of sadness, despair, and impaired functionality. Diagnosis of depression primarily relies on clinical assessment and observation of symptoms. However, due to the lack of...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2023-01-01
|
| Series: | IEEE Transactions on Neural Systems and Rehabilitation Engineering |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10342842/ |
| _version_ | 1797322541651460096 |
|---|---|
| author | Minqiang Yang Ziru Weng Yuhong Zhang Yongfeng Tao Bin Hu |
| author_facet | Minqiang Yang Ziru Weng Yuhong Zhang Yongfeng Tao Bin Hu |
| author_sort | Minqiang Yang |
| collection | DOAJ |
| description | Depression is a prevalent and severe mental disorder that significantly affects both mind and body, leading to persistent feelings of sadness, despair, and impaired functionality. Diagnosis of depression primarily relies on clinical assessment and observation of symptoms. However, due to the lack of objective indicators, the experience and skills of doctor may lead to misdiagnosis. Current researches indicate that eye movement patterns and pupil dilation can serve as potential biomarkers for emotional and cognitive dysregulation in individuals with depression. However, most studies are based on manually extracted eye movement features, overlooking a significant portion of information available in ocular imaging. This paper proposes Three-Stream Convolutional Neural Network (TSCNN) for detecting depression, leveraging both spatio-temporal information of raw ocular imaging and paradigmatic semantic features. We suggest using optical flow with different sampling intervals to capture temporal features. In the third stream, we employ an encoder to learn semantic information from paradigm images and use it as prior knowledge. Finally, we utilize a fully connected network for classification, achieving an accuracy of 79.3% on our self-collected dataset. The proposed method may demonstrate significant clinical utility in the future. |
| first_indexed | 2024-03-08T05:15:48Z |
| format | Article |
| id | doaj.art-1b13b4184e7540bcb405006d9d883d54 |
| institution | Directory Open Access Journal |
| issn | 1558-0210 |
| language | English |
| last_indexed | 2024-03-08T05:15:48Z |
| publishDate | 2023-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Transactions on Neural Systems and Rehabilitation Engineering |
| spelling | doaj.art-1b13b4184e7540bcb405006d9d883d542024-02-07T00:00:06ZengIEEEIEEE Transactions on Neural Systems and Rehabilitation Engineering1558-02102023-01-01314921493010.1109/TNSRE.2023.333951810342842Three-Stream Convolutional Neural Network for Depression Detection With Ocular ImagingMinqiang Yang0https://orcid.org/0000-0002-7571-6439Ziru Weng1Yuhong Zhang2https://orcid.org/0000-0001-6180-4457Yongfeng Tao3Bin Hu4https://orcid.org/0000-0003-3514-5413School of information Science and Engineering, Lanzhou University, Lanzhou, ChinaSchool of information Science and Engineering, Lanzhou University, Lanzhou, ChinaSchool of information Science and Engineering, Lanzhou University, Lanzhou, ChinaSchool of information Science and Engineering, Lanzhou University, Lanzhou, ChinaGansu Provincial Key Laboratory of Wearable Computing, School of Information Science and Engineering, Lanzhou University, Lanzhou, ChinaDepression is a prevalent and severe mental disorder that significantly affects both mind and body, leading to persistent feelings of sadness, despair, and impaired functionality. Diagnosis of depression primarily relies on clinical assessment and observation of symptoms. However, due to the lack of objective indicators, the experience and skills of doctor may lead to misdiagnosis. Current researches indicate that eye movement patterns and pupil dilation can serve as potential biomarkers for emotional and cognitive dysregulation in individuals with depression. However, most studies are based on manually extracted eye movement features, overlooking a significant portion of information available in ocular imaging. This paper proposes Three-Stream Convolutional Neural Network (TSCNN) for detecting depression, leveraging both spatio-temporal information of raw ocular imaging and paradigmatic semantic features. We suggest using optical flow with different sampling intervals to capture temporal features. In the third stream, we employ an encoder to learn semantic information from paradigm images and use it as prior knowledge. Finally, we utilize a fully connected network for classification, achieving an accuracy of 79.3% on our self-collected dataset. The proposed method may demonstrate significant clinical utility in the future.https://ieeexplore.ieee.org/document/10342842/Depression detectioneye movementocular imagingthree-stream convolutional neural network |
| spellingShingle | Minqiang Yang Ziru Weng Yuhong Zhang Yongfeng Tao Bin Hu Three-Stream Convolutional Neural Network for Depression Detection With Ocular Imaging IEEE Transactions on Neural Systems and Rehabilitation Engineering Depression detection eye movement ocular imaging three-stream convolutional neural network |
| title | Three-Stream Convolutional Neural Network for Depression Detection With Ocular Imaging |
| title_full | Three-Stream Convolutional Neural Network for Depression Detection With Ocular Imaging |
| title_fullStr | Three-Stream Convolutional Neural Network for Depression Detection With Ocular Imaging |
| title_full_unstemmed | Three-Stream Convolutional Neural Network for Depression Detection With Ocular Imaging |
| title_short | Three-Stream Convolutional Neural Network for Depression Detection With Ocular Imaging |
| title_sort | three stream convolutional neural network for depression detection with ocular imaging |
| topic | Depression detection eye movement ocular imaging three-stream convolutional neural network |
| url | https://ieeexplore.ieee.org/document/10342842/ |
| work_keys_str_mv | AT minqiangyang threestreamconvolutionalneuralnetworkfordepressiondetectionwithocularimaging AT ziruweng threestreamconvolutionalneuralnetworkfordepressiondetectionwithocularimaging AT yuhongzhang threestreamconvolutionalneuralnetworkfordepressiondetectionwithocularimaging AT yongfengtao threestreamconvolutionalneuralnetworkfordepressiondetectionwithocularimaging AT binhu threestreamconvolutionalneuralnetworkfordepressiondetectionwithocularimaging |