Figure 1 - uploaded by Yuntong Bai
Content may be subject to copyright.
An illustration of the difference between LASSO, group LASSO and sparse group LASSO. Assuming a view of data can be divided into 5 groups, LASSO selects features regardless of the group structure; group LASSO selects features on a group basis; sparse group LASSO selects features on a group basis while also selects individual feature within each selected group
Citations
... Given the flexibility of GNN to integrate multi-modality data, we will investigate BrainGNN on biomarker detection tasks using an integration of multi-paradigm fMRI data (i.e. Bai et al. (2020)). We will explore the connections between the Ra-GConv layers and the tensor decomposition-based clustering methods and the patterns of ROI selection and ROI clustering. ...
Understanding which brain regions are related to a specific neurological disorder or cognitive stimuli has been an important area of neuroimaging research. We propose BrainGNN, a graph neural network (GNN) framework to analyze functional magnetic resonance images (fMRI) and discover neurological biomarkers. Considering the special property of brain graphs, we design novel ROI-aware graph convolutional (Ra-GConv) layers that leverage the topological and functional information of fMRI. Motivated by the need for transparency in medical image analysis, our BrainGNN contains ROI-selection pooling layers (R-pool) that highlight salient ROIs (nodes in the graph), so that we can infer which ROIs are important for prediction. Furthermore, we propose regularization terms—unit loss, topK pooling (TPK) loss and group-level consistency (GLC) loss—on pooling results to encourage reasonable ROI-selection and provide flexibility to encourage either fully individual- or patterns that agree with group-level data. We apply the BrainGNN framework on two independent fMRI datasets: an Autism Spectrum Disorder (ASD) fMRI dataset and data from the Human Connectome Project (HCP) 900 Subject Release. We investigate different choices of the hyper-parameters and show that BrainGNN outperforms the alternative fMRI image analysis methods in terms of four different evaluation metrics. The obtained community clustering and salient ROI detection results show a high correspondence with the previous neuroimaging-derived evidence of biomarkers for ASD and specific task states decoded for HCP. We will make BrainGNN codes public available after acceptance.
