Predicting individual task contrasts from resting‐state functional connectivity using a surface‐based convolutional network
Task-based and resting-state represent the two most common experimental paradigms of functional neuroimaging. While resting-state offers a flexible and scalable approach for characterizing brain function, task-based techniques provide superior localization. In this paper, we build on recent deep lea...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2022-03-01
|
| Series: | NeuroImage |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S1053811921011204 |
| _version_ | 1818330295366057984 |
|---|---|
| author | Gia H. Ngo Meenakshi Khosla Keith Jamison Amy Kuceyeski Mert R. Sabuncu |
| author_facet | Gia H. Ngo Meenakshi Khosla Keith Jamison Amy Kuceyeski Mert R. Sabuncu |
| author_sort | Gia H. Ngo |
| collection | DOAJ |
| description | Task-based and resting-state represent the two most common experimental paradigms of functional neuroimaging. While resting-state offers a flexible and scalable approach for characterizing brain function, task-based techniques provide superior localization. In this paper, we build on recent deep learning methods to create a model that predicts task-based contrast maps from resting-state fMRI scans. Specifically, we propose BrainSurfCNN, a surface-based fully-convolutional neural network model that works with a representation of the brain’s cortical sheet. BrainSurfCNN achieves exceptional predictive accuracy on independent test data from the Human Connectome Project, which is on par with the repeat reliability of the measured subject-level contrast maps. Conversely, our analyses reveal that a previously published benchmark is no better than group-average contrast maps. Finally, we demonstrate that BrainSurfCNN can generalize remarkably well to novel domains with limited training data. |
| first_indexed | 2024-12-13T13:01:41Z |
| format | Article |
| id | doaj.art-2a546fa5d9594964a4835a4e721b6f92 |
| institution | Directory Open Access Journal |
| issn | 1095-9572 |
| language | English |
| last_indexed | 2024-12-13T13:01:41Z |
| publishDate | 2022-03-01 |
| publisher | Elsevier |
| record_format | Article |
| series | NeuroImage |
| spelling | doaj.art-2a546fa5d9594964a4835a4e721b6f922022-12-21T23:44:59ZengElsevierNeuroImage1095-95722022-03-01248118849Predicting individual task contrasts from resting‐state functional connectivity using a surface‐based convolutional networkGia H. Ngo0Meenakshi Khosla1Keith Jamison2Amy Kuceyeski3Mert R. Sabuncu4School of Electrical & Computer Engineering, Cornell University and Cornell Tech, United StatesSchool of Electrical & Computer Engineering, Cornell University and Cornell Tech, United StatesRadiology, Weill Cornell Medicine, United StatesRadiology, Weill Cornell Medicine, United StatesCorresponding author at: School of Electrical & Computer Engineering, Cornell University and Cornell Tech, United States.; School of Electrical & Computer Engineering, Cornell University and Cornell Tech, United States; Radiology, Weill Cornell Medicine, United StatesTask-based and resting-state represent the two most common experimental paradigms of functional neuroimaging. While resting-state offers a flexible and scalable approach for characterizing brain function, task-based techniques provide superior localization. In this paper, we build on recent deep learning methods to create a model that predicts task-based contrast maps from resting-state fMRI scans. Specifically, we propose BrainSurfCNN, a surface-based fully-convolutional neural network model that works with a representation of the brain’s cortical sheet. BrainSurfCNN achieves exceptional predictive accuracy on independent test data from the Human Connectome Project, which is on par with the repeat reliability of the measured subject-level contrast maps. Conversely, our analyses reveal that a previously published benchmark is no better than group-average contrast maps. Finally, we demonstrate that BrainSurfCNN can generalize remarkably well to novel domains with limited training data.http://www.sciencedirect.com/science/article/pii/S1053811921011204Surface-based convolutional neural networkTask-evoked contrastsResting-state functional connectivity |
| spellingShingle | Gia H. Ngo Meenakshi Khosla Keith Jamison Amy Kuceyeski Mert R. Sabuncu Predicting individual task contrasts from resting‐state functional connectivity using a surface‐based convolutional network NeuroImage Surface-based convolutional neural network Task-evoked contrasts Resting-state functional connectivity |
| title | Predicting individual task contrasts from resting‐state functional connectivity using a surface‐based convolutional network |
| title_full | Predicting individual task contrasts from resting‐state functional connectivity using a surface‐based convolutional network |
| title_fullStr | Predicting individual task contrasts from resting‐state functional connectivity using a surface‐based convolutional network |
| title_full_unstemmed | Predicting individual task contrasts from resting‐state functional connectivity using a surface‐based convolutional network |
| title_short | Predicting individual task contrasts from resting‐state functional connectivity using a surface‐based convolutional network |
| title_sort | predicting individual task contrasts from resting state functional connectivity using a surface based convolutional network |
| topic | Surface-based convolutional neural network Task-evoked contrasts Resting-state functional connectivity |
| url | http://www.sciencedirect.com/science/article/pii/S1053811921011204 |
| work_keys_str_mv | AT giahngo predictingindividualtaskcontrastsfromrestingstatefunctionalconnectivityusingasurfacebasedconvolutionalnetwork AT meenakshikhosla predictingindividualtaskcontrastsfromrestingstatefunctionalconnectivityusingasurfacebasedconvolutionalnetwork AT keithjamison predictingindividualtaskcontrastsfromrestingstatefunctionalconnectivityusingasurfacebasedconvolutionalnetwork AT amykuceyeski predictingindividualtaskcontrastsfromrestingstatefunctionalconnectivityusingasurfacebasedconvolutionalnetwork AT mertrsabuncu predictingindividualtaskcontrastsfromrestingstatefunctionalconnectivityusingasurfacebasedconvolutionalnetwork |