LUMINA: Laplacian-Unifying Mechanism for Interpretable Neurodevelopmental Analysis via Quad-Stream GCN
Minkyung Cha, Jooyoung Bae, Jaewon Jung, Ping Shu Ho, Ka Chun Cheung, Namjoon Kim · 2026
Read LUMINA: Laplacian-Unifying Mechanism for Interpretable Neurodevelopmental Analysis via Quad-Stream GCN by Minkyung Cha, Jooyoung Bae, Jaewon Jung, Ping Shu Ho, Ka Chun Cheung, Namjoon Kim (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2603-13329.txt.
Description
Minkyung Cha and co-authors propose LUMINA, a graph model for fMRI-based diagnosis of brain disorders in youth. Graph Convolutional Networks treat brain regions as linked nodes, which fits how the brain works as a whole. Yet their smoothing can blur the contrast that marks some disorders. LUMINA uses a Quad-Stream GCN with a bipolar RELU and dual-spectrum Laplacian filtering. That design keeps more of the varied dynamics in each fMRI scan. Cross-validation on ADHD200 and ABIDE data beat prior models on ADHD and ASD accuracy.
Catalog record
| CREATOR | Minkyung Cha, Jooyoung Bae, Jaewon Jung, Ping Shu Ho, Ka Chun Cheung, Namjoon Kim |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 17348 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2603.13329