Continuous Dictionary of Nodes Model and Bilinear-Diffusion Representation Learning for Brain Disease Analysis

Brain Sciences, Journal Year: 2024, Volume and Issue: 14(8), P. 810 - 810

Published: Aug. 13, 2024

Brain networks based on functional magnetic resonance imaging (fMRI) provide a crucial perspective for diagnosing brain diseases. Representation learning has recently attracted tremendous attention due to its strong representation capability, which can be naturally applied disease analysis. However, traditional only considers direct and local node interactions in original networks, posing challenges constructing higher-order represent indirect extensive interactions. To address this problem, we propose the Continuous Dictionary of Nodes model Bilinear-Diffusion (CDON-BD) network The CDON is innovatively used learn network, with encoder weights directly regarded as latent features. fully integrate features, further utilize Bilinear Pooling construct networks. Diffusion Module designed capture Compared state-of-the-art methods, CDON-BD demonstrates competitive classification performance two real datasets. Moreover, representations learned by our method reveal regions relevant diseases, contributing better understanding pathology

Language: Английский

Multi-modal neuroimaging technique: Innovations and applications

Brain Science Advances, Journal Year: 2023, Volume and Issue: 9(2), P. 53 - 55

Published: June 1, 2023