Abstract

Tau neurofibrillary pathology in Alzheimer’s disease (AD) follows stereotypical spatial patterns that are associated with white-matter connectivity. However, human brain structure varies substantially across individuals, thus understanding how these inter-individual differences in diffusion MRI (dMRI)-derived structural connectivity (SC) account for the corresponding heterogeneity in tau distribution remains unclear. We introduce SC-TauPath, a pairwise attribution framework that maps SC differences directly to tau pattern heterogeneity across subjects. For each pair of participants, SC-TauPath learns a mapping from pairwise SC differences (∆SC) to pairwise tau PET differences (∆τ ). The learned operator is then analytically decomposed to provide multi-scale attribution, identifying influential connections (edges), coordinated connection patterns (pathways), and key regions (hubs) most strongly associated with tau heterogeneity. Applied to 234 ADNI participants with dMRI-derived SC and 18 F-Flortaucipir PET, SC-TauPath substantially outperforms the baseline by achieving higher cross-validated Pearson correlation in predicting inter-subject tau pattern differences. More importantly, our SC-TauPath captures SC-tau relationships not only across diagnostic groups, thereby revealing connectivity patterns associated with disease progression, but also within the same diagnostic stage, demonstrating that individual differences in SC contribute to tau heterogeneity even among subjects at similar clinical impairment levels. Code is available at https://github.com/Nancy-Zhang-0/SC-TauPath

Links to Paper and Supplementary Materials

Main Paper (Open Access Version): https://papers.miccai.org/miccai-2026-sat/paper/MLMI_010.pdf

SharedIt Link: Not yet available

SpringerLink (DOI): Not yet available

Supplementary Material: Not Submitted

Link to Open Review

Open Review Page: https://openreview.net/forum?id=xldo6VJXyY

BibTex

@InProceedings{ZhaJin_SCTauPath_MICCAISAT2026,
        author = { Zhang, Jing AND Scheel, Norman AND Chen, Minheng AND Chen, Tong AND Lyu, Yanjun AND Zhu, David C. AND Zhang, Rong AND Zhu, Dajiang},
        title = { { SC-TauPath: How Structural Connectivity Differences Shape Tau Distribution in Alzheimer’s Disease } },
        booktitle = {Medical Image Computing and Computer Assisted Intervention -- MICCAI 2026 Workshops and Challenges},
        year = {2026},
        publisher = {Springer Nature Switzerland},
        volume = {LNCS 17272},
        month = {pending},
        page = {pending}
}


back to top