Knowledge-driven generative subspaces for modeling multi-view dependencies in medical data
This addresses Alzheimer's disease detection for medical applications, but appears incremental as it builds on existing multi-view modeling approaches.
The paper tackles early detection of Alzheimer's disease by learning a disease model that combines genotypic, phenotypic, and cognitive data, proposing a probabilistic generative subspace to model multi-view dependencies, and shows it can lead to explainable clinical predictions and improved diagnoses.
Early detection of Alzheimer's disease (AD) and identification of potential risk/beneficial factors are important for planning and administering timely interventions or preventive measures. In this paper, we learn a disease model for AD that combines genotypic and phenotypic profiles, and cognitive health metrics of patients. We propose a probabilistic generative subspace that describes the correlative, complementary and domain-specific semantics of the dependencies in multi-view, multi-modality medical data. Guided by domain knowledge and using the latent consensus between abstractions of multi-view data, we model the fusion as a data generating process. We show that our approach can potentially lead to i) explainable clinical predictions and ii) improved AD diagnoses.