Neuroimaging studies have a growing interest in learning the association between the individual brain connectivity networks and their clinical characteristics. It is also of great interest to identify the sub-brain networks as biomarkers to predict the clinical symptoms, such as disease status, potentially providing insight on neuropathology. This motivates the need for developing a new type of regression model where the response variable is scalar, and predictors are networks that are typically represented as adjacent matrices or weighted adjacent matrices to which we refer as scalar-on-network regression. In this work we develop a new boosting method for model fitting with subnetwork markers selection. Our approach, as opposed to group lasso or other existing regularization methods, is, essentially, a gradient descent algorithm leveraging known network structure. We demonstrate the utility of our methods via simulation studies and analysis of the resting-state fMRI data in a cognitive developmental cohort study.
This work was partially supported by grants NIH R01 DA048993 (Kang and Morris), NIH R01MH105561 (Kang), NIH R01GM124061 (Kang) and NSF IIS2123777 (Kang).
The authors would like to thank the Editor, Professor Nicoleta Serban, the Associate Editor, and reviewers for their helpful comments and constructive suggestions which led to a much-improved manuscript.
"Scalar on network regression via boosting." Ann. Appl. Stat. 16 (4) 2755 - 2773, December 2022. https://doi.org/10.1214/22-AOAS1612