Knowledge-driven generative subspaces for modeling multi-view dependencies in medical data

TL;DR

This paper introduces a probabilistic generative subspace model that integrates genotypic, phenotypic, and cognitive data to improve Alzheimer's disease diagnosis and provide explainable predictions.

Contribution

It presents a novel knowledge-driven generative framework for modeling multi-view dependencies in medical data, enhancing interpretability and diagnostic accuracy.

Findings

Potential for explainable clinical predictions

Improved Alzheimer's disease diagnosis accuracy

Effective modeling of multi-view medical data

Abstract

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.