U-shaped convolutional transformer GAN with multi-resolution consistency loss for restoring brain functional time-series and dementia diagnosis
IntroductionThe blood oxygen level-dependent (BOLD) signal derived from functional neuroimaging is commonly used in brain network analysis and dementia diagnosis. Missing the BOLD signal may lead to bad performance and misinterpretation of findings when analyzing neurological disease. Few studies ha...
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2024-04-01
|
| Series: | Frontiers in Computational Neuroscience |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fncom.2024.1387004/full |
| _version_ | 1797203077539823616 |
|---|---|
| author | Qiankun Zuo Qiankun Zuo Qiankun Zuo Ruiheng Li Ruiheng Li Binghua Shi Binghua Shi Jin Hong Yanfei Zhu Xuhang Chen Yixian Wu Jia Guo Jia Guo Jia Guo |
| author_facet | Qiankun Zuo Qiankun Zuo Qiankun Zuo Ruiheng Li Ruiheng Li Binghua Shi Binghua Shi Jin Hong Yanfei Zhu Xuhang Chen Yixian Wu Jia Guo Jia Guo Jia Guo |
| author_sort | Qiankun Zuo |
| collection | DOAJ |
| description | IntroductionThe blood oxygen level-dependent (BOLD) signal derived from functional neuroimaging is commonly used in brain network analysis and dementia diagnosis. Missing the BOLD signal may lead to bad performance and misinterpretation of findings when analyzing neurological disease. Few studies have focused on the restoration of brain functional time-series data.MethodsIn this paper, a novel U-shaped convolutional transformer GAN (UCT-GAN) model is proposed to restore the missing brain functional time-series data. The proposed model leverages the power of generative adversarial networks (GANs) while incorporating a U-shaped architecture to effectively capture hierarchical features in the restoration process. Besides, the multi-level temporal-correlated attention and the convolutional sampling in the transformer-based generator are devised to capture the global and local temporal features for the missing time series and associate their long-range relationship with the other brain regions. Furthermore, by introducing multi-resolution consistency loss, the proposed model can promote the learning of diverse temporal patterns and maintain consistency across different temporal resolutions, thus effectively restoring complex brain functional dynamics.ResultsWe theoretically tested our model on the public Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset, and our experiments demonstrate that the proposed model outperforms existing methods in terms of both quantitative metrics and qualitative assessments. The model's ability to preserve the underlying topological structure of the brain functional networks during restoration is a particularly notable achievement.ConclusionOverall, the proposed model offers a promising solution for restoring brain functional time-series and contributes to the advancement of neuroscience research by providing enhanced tools for disease analysis and interpretation. |
| first_indexed | 2024-04-24T08:13:36Z |
| format | Article |
| id | doaj.art-2141883fa9a24d888b58122f98fad50b |
| institution | Directory Open Access Journal |
| issn | 1662-5188 |
| language | English |
| last_indexed | 2024-04-24T08:13:36Z |
| publishDate | 2024-04-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Computational Neuroscience |
| spelling | doaj.art-2141883fa9a24d888b58122f98fad50b2024-04-17T04:36:32ZengFrontiers Media S.A.Frontiers in Computational Neuroscience1662-51882024-04-011810.3389/fncom.2024.13870041387004U-shaped convolutional transformer GAN with multi-resolution consistency loss for restoring brain functional time-series and dementia diagnosisQiankun Zuo0Qiankun Zuo1Qiankun Zuo2Ruiheng Li3Ruiheng Li4Binghua Shi5Binghua Shi6Jin Hong7Yanfei Zhu8Xuhang Chen9Yixian Wu10Jia Guo11Jia Guo12Jia Guo13Hubei Key Laboratory of Digital Finance Innovation, Hubei University of Economics, Wuhan, Hubei, ChinaSchool of Information Engineering, Hubei University of Economics, Wuhan, Hubei, ChinaHubei Internet Finance Information Engineering Technology Research Center, Hubei University of Economics, Wuhan, Hubei, ChinaHubei Key Laboratory of Digital Finance Innovation, Hubei University of Economics, Wuhan, Hubei, ChinaSchool of Information Engineering, Hubei University of Economics, Wuhan, Hubei, ChinaHubei Key Laboratory of Digital Finance Innovation, Hubei University of Economics, Wuhan, Hubei, ChinaSchool of Information Engineering, Hubei University of Economics, Wuhan, Hubei, ChinaMedical Research Institute, Guangdong Provincial People's Hospital (Guangdong Academy of Medical Sciences), Southern Medical University, Guangzhou, ChinaSchool of Foreign Languages, Sun Yat-sen University, Guangzhou, ChinaFaculty of Science and Technology, University of Macau, Taipa, Macao SAR, ChinaSchool of Mechanical Engineering, Beijing Institute of Petrochemical Technology, Beijing, ChinaHubei Key Laboratory of Digital Finance Innovation, Hubei University of Economics, Wuhan, Hubei, ChinaSchool of Information Engineering, Hubei University of Economics, Wuhan, Hubei, ChinaHubei Internet Finance Information Engineering Technology Research Center, Hubei University of Economics, Wuhan, Hubei, ChinaIntroductionThe blood oxygen level-dependent (BOLD) signal derived from functional neuroimaging is commonly used in brain network analysis and dementia diagnosis. Missing the BOLD signal may lead to bad performance and misinterpretation of findings when analyzing neurological disease. Few studies have focused on the restoration of brain functional time-series data.MethodsIn this paper, a novel U-shaped convolutional transformer GAN (UCT-GAN) model is proposed to restore the missing brain functional time-series data. The proposed model leverages the power of generative adversarial networks (GANs) while incorporating a U-shaped architecture to effectively capture hierarchical features in the restoration process. Besides, the multi-level temporal-correlated attention and the convolutional sampling in the transformer-based generator are devised to capture the global and local temporal features for the missing time series and associate their long-range relationship with the other brain regions. Furthermore, by introducing multi-resolution consistency loss, the proposed model can promote the learning of diverse temporal patterns and maintain consistency across different temporal resolutions, thus effectively restoring complex brain functional dynamics.ResultsWe theoretically tested our model on the public Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset, and our experiments demonstrate that the proposed model outperforms existing methods in terms of both quantitative metrics and qualitative assessments. The model's ability to preserve the underlying topological structure of the brain functional networks during restoration is a particularly notable achievement.ConclusionOverall, the proposed model offers a promising solution for restoring brain functional time-series and contributes to the advancement of neuroscience research by providing enhanced tools for disease analysis and interpretation.https://www.frontiersin.org/articles/10.3389/fncom.2024.1387004/fullhierarchical topological transformermulti-level temporal-correlated attentioncentral connectivity perceptiontime-series restorationmulti-head attentionbrain neurological disease |
| spellingShingle | Qiankun Zuo Qiankun Zuo Qiankun Zuo Ruiheng Li Ruiheng Li Binghua Shi Binghua Shi Jin Hong Yanfei Zhu Xuhang Chen Yixian Wu Jia Guo Jia Guo Jia Guo U-shaped convolutional transformer GAN with multi-resolution consistency loss for restoring brain functional time-series and dementia diagnosis Frontiers in Computational Neuroscience hierarchical topological transformer multi-level temporal-correlated attention central connectivity perception time-series restoration multi-head attention brain neurological disease |
| title | U-shaped convolutional transformer GAN with multi-resolution consistency loss for restoring brain functional time-series and dementia diagnosis |
| title_full | U-shaped convolutional transformer GAN with multi-resolution consistency loss for restoring brain functional time-series and dementia diagnosis |
| title_fullStr | U-shaped convolutional transformer GAN with multi-resolution consistency loss for restoring brain functional time-series and dementia diagnosis |
| title_full_unstemmed | U-shaped convolutional transformer GAN with multi-resolution consistency loss for restoring brain functional time-series and dementia diagnosis |
| title_short | U-shaped convolutional transformer GAN with multi-resolution consistency loss for restoring brain functional time-series and dementia diagnosis |
| title_sort | u shaped convolutional transformer gan with multi resolution consistency loss for restoring brain functional time series and dementia diagnosis |
| topic | hierarchical topological transformer multi-level temporal-correlated attention central connectivity perception time-series restoration multi-head attention brain neurological disease |
| url | https://www.frontiersin.org/articles/10.3389/fncom.2024.1387004/full |
| work_keys_str_mv | AT qiankunzuo ushapedconvolutionaltransformerganwithmultiresolutionconsistencylossforrestoringbrainfunctionaltimeseriesanddementiadiagnosis AT qiankunzuo ushapedconvolutionaltransformerganwithmultiresolutionconsistencylossforrestoringbrainfunctionaltimeseriesanddementiadiagnosis AT qiankunzuo ushapedconvolutionaltransformerganwithmultiresolutionconsistencylossforrestoringbrainfunctionaltimeseriesanddementiadiagnosis AT ruihengli ushapedconvolutionaltransformerganwithmultiresolutionconsistencylossforrestoringbrainfunctionaltimeseriesanddementiadiagnosis AT ruihengli ushapedconvolutionaltransformerganwithmultiresolutionconsistencylossforrestoringbrainfunctionaltimeseriesanddementiadiagnosis AT binghuashi ushapedconvolutionaltransformerganwithmultiresolutionconsistencylossforrestoringbrainfunctionaltimeseriesanddementiadiagnosis AT binghuashi ushapedconvolutionaltransformerganwithmultiresolutionconsistencylossforrestoringbrainfunctionaltimeseriesanddementiadiagnosis AT jinhong ushapedconvolutionaltransformerganwithmultiresolutionconsistencylossforrestoringbrainfunctionaltimeseriesanddementiadiagnosis AT yanfeizhu ushapedconvolutionaltransformerganwithmultiresolutionconsistencylossforrestoringbrainfunctionaltimeseriesanddementiadiagnosis AT xuhangchen ushapedconvolutionaltransformerganwithmultiresolutionconsistencylossforrestoringbrainfunctionaltimeseriesanddementiadiagnosis AT yixianwu ushapedconvolutionaltransformerganwithmultiresolutionconsistencylossforrestoringbrainfunctionaltimeseriesanddementiadiagnosis AT jiaguo ushapedconvolutionaltransformerganwithmultiresolutionconsistencylossforrestoringbrainfunctionaltimeseriesanddementiadiagnosis AT jiaguo ushapedconvolutionaltransformerganwithmultiresolutionconsistencylossforrestoringbrainfunctionaltimeseriesanddementiadiagnosis AT jiaguo ushapedconvolutionaltransformerganwithmultiresolutionconsistencylossforrestoringbrainfunctionaltimeseriesanddementiadiagnosis |