TY - GEN
T1 - Sample Augmentation for Classification of Schizophrenia Patients and Healthy Controls Using ICA of fMRI Data and Convolutional Neural Networks
AU - Niu, Yan Wei
AU - Lin, Qiu Hua
AU - Qiu, Yue
AU - Kuang, Li Dan
AU - Calhoun, Vince D.
N1 - Funding Information:
This work was supported by the National Natural Science Foundation of China under Grants 61871067, 61379012, 61901061, NSF grants 1539067, 0840895, 1539067 and 0715022, NIH grants R01MH104680, R01MH107354, R01EB005846 and 5P20GM103472.
Publisher Copyright:
© 2019 IEEE.
PY - 2019/12
Y1 - 2019/12
N2 - Convolutional neural networks (CNN) have exhibited great success in image classification. The application of CNN to classification of patients with brain disorders and healthy controls is also promising using functional magnetic resonance imaging (fMRI) data. However, the shortage of the number of subjects is a challenge for training CNN. Spatial maps separated from the fMRI data by independent component analysis (ICA) can provide a solution to this problem within an ICA-CNN framework. As such, we propose three strategies for both prior to and post ICA sample augmentation in the ICA-CNN framework. More precisely, we propose to increase the number of samples by performing spatial smoothing and band-pass filtering on the observed fMRI data before ICA, and spatial smoothing on the spatial maps after ICA. We evaluate the proposed methods using 82 resting-state fMRI datasets including 42 Schizophrenia patients and 40 healthy controls. The spatial map of the default mode network is used for classification, and each data augmentation is constrained to have the same numbers of samples for a fair comparison. The results show a 2%15% increase in an average accuracy compared to the existing multiple-model-order method when adopting each of the proposed sample augmentation strategies. The spatial smoothing on the spatial maps is the most accurate among the three proposed methods. When using a combination of the proposed spatial smoothing on the spatial maps with the multiple-model-order method, the average accuracy increases above 90%.
AB - Convolutional neural networks (CNN) have exhibited great success in image classification. The application of CNN to classification of patients with brain disorders and healthy controls is also promising using functional magnetic resonance imaging (fMRI) data. However, the shortage of the number of subjects is a challenge for training CNN. Spatial maps separated from the fMRI data by independent component analysis (ICA) can provide a solution to this problem within an ICA-CNN framework. As such, we propose three strategies for both prior to and post ICA sample augmentation in the ICA-CNN framework. More precisely, we propose to increase the number of samples by performing spatial smoothing and band-pass filtering on the observed fMRI data before ICA, and spatial smoothing on the spatial maps after ICA. We evaluate the proposed methods using 82 resting-state fMRI datasets including 42 Schizophrenia patients and 40 healthy controls. The spatial map of the default mode network is used for classification, and each data augmentation is constrained to have the same numbers of samples for a fair comparison. The results show a 2%15% increase in an average accuracy compared to the existing multiple-model-order method when adopting each of the proposed sample augmentation strategies. The spatial smoothing on the spatial maps is the most accurate among the three proposed methods. When using a combination of the proposed spatial smoothing on the spatial maps with the multiple-model-order method, the average accuracy increases above 90%.
KW - Convolutional neural network
KW - ICA
KW - classification
KW - fMRI
KW - sample argumentation
KW - spatial maps
UR - http://www.scopus.com/inward/record.url?scp=85082241720&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85082241720&partnerID=8YFLogxK
U2 - 10.1109/ICICIP47338.2019.9012169
DO - 10.1109/ICICIP47338.2019.9012169
M3 - Conference contribution
AN - SCOPUS:85082241720
T3 - 10th International Conference on Intelligent Control and Information Processing, ICICIP 2019
SP - 297
EP - 302
BT - 10th International Conference on Intelligent Control and Information Processing, ICICIP 2019
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 10th International Conference on Intelligent Control and Information Processing, ICICIP 2019
Y2 - 14 December 2019 through 19 December 2019
ER -