Network regularization in imaging genetics improves prediction performances and model interpretability on Alzheimers's disease

Nicolas Guigui (1, 2) , C. Philippe (1) , A. Gloaguen (1, 3) , S. Karkar (1) , V. Guillemot (4) , T. Löfstedt (5) , V. Frouin (1)

1 UNATI - Unité Analyse et Traitement de l'Information
2 CentraleSupélec
3 L2S - Laboratoire des signaux et systèmes
4 C3BI - Centre de Bioinformatique, Biostatistique et Biologie Intégrative
5 Department of Radiation Sciences [Umeå]

Imaging-genetics is a growing popular research avenue which aims to find genetic variants associated with quantitative phenotypes that characterize a disease. In this work, we combine structural MRI with genetic data structured by prior knowledge of interactions in a Canonical Correlation Analysis (CCA) model with graph regularization. This results in improved prediction performance and yields a more inter-pretable model.

Mots clés: Index Terms-Imaging-Genetics; Networks; Structured constraints; Generalized Canonical Correlation Analysis

Domaines: Imagerie; Bio-informatique [q-bio.QM]; Génétique des populations [q-bio.PE]; Machine Learning [stat.ML]

Nicolas Guigui, C. Philippe, A. Gloaguen, S. Karkar, V. Guillemot, et al.. Network regularization in imaging genetics improves prediction performances and model interpretability on Alzheimers's disease. ISBI 2019 - Proceedings of the IEEE International Symposium on Biomedical Imaging, Apr 2019, Venice, Italy. ⟨10.1109/ISBI.2019.8759593⟩. ⟨cea-02016625⟩