KEYWORDS: Brain, Brain diseases, Education and training, Machine learning, Design and modelling, Data modeling, Matrices, Brain mapping, Transform theory
Recently, machine learning attracts more attention on autism spectrum disorder (ASD) identification based on resting-state functional magnetic resonance imaging (rs-fMRI). Most studies on ASD identification use rs-fMRI data from multiple imaging sites to increase sample size, but they suffer from the data heterogeneity among sites. Besides, most ASD identification studies use features simply extracted from brain connections, ignoring the topological structure of brain. To address these issues, we propose a brain graph synthesis model based on generative adversarial network (GAN), which transforms data from the source domain to the target domain, solving the data heterogeneity among sites. Specifically, the generator and discriminator are designed for graph structured data and propose topological losses to construct the cycle consistency loss for maintaining the original topological structure of the reconstructed brain graph. We carry out experiments on six tasks using the open database Autism Brain Imaging Data Exchange (ABIDE) for ASD identification. Experimental findings demonstrate that our model can solve the problem of data heterogeneity effectively and achieve satisfying performance on multi-site ASD identification.
KEYWORDS: Functional magnetic resonance imaging, Data centers, Data conversion, Data modeling, Information technology, Feature selection, Brain imaging, Diseases and disorders
Recently resting-state functional magnetic resonance imaging (R-fMRI) has been applied as a powerful tool to explore potential biomarkers of autism spectrum disorder (ASD). However, in clinical data, the number of ASD patients is significantly less than that of typical development (TD) subjects, which causes the production of imbalanced data. When the imbalanced data are used to predict ASD, the prediction results are not satisfactory. To improve the ASD prediction performance of imbalanced data, this paper adopts the clustering oversampling method to enhance the representation for minority class (ASD), expecting to obtain the balanced data distribution. For the imbalanced data after feature selection, the clustering algorithm is used to form a few clusters in the ASD group and in the TD group, respectively, and then new samples for each cluster are generated by synthetic minority oversampling technique (SMOTE) to make the imbalanced data convert into the balanced data. Finally, we construct the linear support vector machine (SVM) classification model for ASD prediction. The prediction accuracy of multi-center imbalanced R-fMRI data increased from 59.70% to 66.62% using hierarchical clustering oversampling. The results of experiment show that the clustering oversampling method can effectively improve the prediction performance of imbalanced R-fMRI data.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.