TY - GEN
T1 - Deep Transfer Learning for Whole-Brain FMRI Analyses
AU - Thomas, Armin W.
AU - Müller, Klaus Robert
AU - Samek, Wojciech
N1 - Funding Information:
FMRI Data. All analyzed fMRI data were provided in a preprocessed format by the Human Connectome Project (HCP), WU Minn Consortium (Principal Investigators: David VanEssen and Kamil Ugurbil; 1U54MH091657) funded by the 16 NIH Institutes and Centers that support the NIH Blueprint for Neuroscience Research; and by the McDonnell Center for Systems Neuroscience at Washington University. Whole-brain EPI acquisitions were acquired with a 32 channel head coil on a modified 3T Siemens Skyra with TR = 720 ms and TE = 33.1 ms (for further details on fMRI acquisition, see [14]).
Publisher Copyright:
© 2019, Springer Nature Switzerland AG.
PY - 2019
Y1 - 2019
N2 - The application of deep learning (DL) models to the decoding of cognitive states from whole-brain functional Magnetic Resonance Imaging (fMRI) data is often hindered by the small sample size and high dimensionality of these datasets. Especially, in clinical settings, where patient data are scarce. In this work, we demonstrate that transfer learning represents a solution to this problem. Particularly, we show that a DL model, which has been previously trained on a large openly available fMRI dataset of the Human Connectome Project, outperforms a model variant with the same architecture, but which is trained from scratch, when both are applied to the data of a new, unrelated fMRI task. The pre-trained DL model variant is able to correctly decode 67.51% of the cognitive states from a test dataset with 100 individuals, when fine-tuned on a dataset of the size of only three subjects.
AB - The application of deep learning (DL) models to the decoding of cognitive states from whole-brain functional Magnetic Resonance Imaging (fMRI) data is often hindered by the small sample size and high dimensionality of these datasets. Especially, in clinical settings, where patient data are scarce. In this work, we demonstrate that transfer learning represents a solution to this problem. Particularly, we show that a DL model, which has been previously trained on a large openly available fMRI dataset of the Human Connectome Project, outperforms a model variant with the same architecture, but which is trained from scratch, when both are applied to the data of a new, unrelated fMRI task. The pre-trained DL model variant is able to correctly decode 67.51% of the cognitive states from a test dataset with 100 individuals, when fine-tuned on a dataset of the size of only three subjects.
KW - Decoding
KW - Deep learning
KW - Transfer learning
KW - fMRI
UR - http://www.scopus.com/inward/record.url?scp=85075558317&partnerID=8YFLogxK
U2 - 10.1007/978-3-030-32695-1_7
DO - 10.1007/978-3-030-32695-1_7
M3 - Conference contribution
AN - SCOPUS:85075558317
SN - 9783030326944
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 59
EP - 67
BT - OR 2.0 Context-Aware Operating Theaters and Machine Learning in Clinical Neuroimaging - 2nd International Workshop, OR 2.0 2019, and 2nd International Workshop, MLCN 2019, Held in Conjunction with MICCAI 2019, Proceedings
A2 - Zhou, Luping
A2 - Sarikaya, Duygu
A2 - Kia, Seyed Mostafa
A2 - Speidel, Stefanie
A2 - Malpani, Anand
A2 - Hashimoto, Daniel
A2 - Habes, Mohamad
A2 - Löfstedt, Tommy
A2 - Ritter, Kerstin
A2 - Wang, Hongzhi
PB - Springer
T2 - 2nd International Workshop on Context-Aware Surgical Theaters, OR 2.0 2019, and the 2nd International Workshop on Machine Learning in Clinical Neuroimaging, MLCN 2019, held in conjunction with the 22nd International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2019
Y2 - 17 October 2019 through 17 October 2019
ER -