MIRAGE: Knowledge Graph-Guided Cross-Cohort MRI Synthesis for Alzheimer’s Disease Prediction

Guanchen Wu, Zhe Huang, Yuzhang Xie, Runze Yan, Akul Chopra, Deqiang Qiu, Xiao Hu, Fei Wang, Carl Yang🖂 · 2026

Read MIRAGE: Knowledge Graph-Guided Cross-Cohort MRI Synthesis for Alzheimer’s Disease Prediction by Guanchen Wu, Zhe Huang, Yuzhang Xie, Runze Yan, Akul Chopra, Deqiang Qiu, Xiao Hu, Fei Wang, Carl Yang🖂 (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-2603-02434.txt.

Description

Guanchen Wu, Zhe Huang, Yuzhang Xie, and colleagues predict Alzheimer's disease when brain scans are missing in this 2026 paper. Diagnosis increasingly mixes MRI scans with electronic health records. Scans are costly, though, and many patient groups lack them entirely. Building fake 3D scans from sparse records is risky and hard.

Their framework, MIRAGE, never builds a fake scan at test time. A biomedical knowledge graph first maps varied record fields into one shared space. Graph attention spreads that knowledge from scanned groups to unscanned ones. A frozen 3D decoder acts as a structural check during training, forcing the record code to hold plausible anatomy. Only that compact surrogate code is used for diagnosis. Accuracy rose notably over single-source baselines in scan-free groups.

Catalog record

CREATORGuanchen Wu, Zhe Huang, Yuzhang Xie, Runze Yan, Akul Chopra, Deqiang Qiu, Xiao Hu, Fei Wang, Carl Yang🖂
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE23353
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2603.02434

Browse the full library · About the Tower of Babel