Genetics-Driven Personalized Disease Progression Model

2025-02-24

AI Summary

Chronic disease progression exhibits substantial inter-individual heterogeneity, which conventional population-level, homogeneous modeling approaches fail to capture due to their inability to account for genetic and environmental drivers of divergent trajectories. To address this, we propose the first integrated framework jointly modeling genetic features and longitudinal clinical time series—moving beyond static subtype assumptions to simultaneously discover genetically informed subtypes and infer personalized disease progression trajectories. Methodologically, our approach employs a variational autoencoder (VAE) to learn low-dimensional, interpretable genetic representations, coupled with an RNN-driven state-space model to characterize multi-stage clinical dynamics. Evaluated on both real-world and synthetic datasets, the model achieves a 12.3% absolute improvement in disease stage prediction accuracy, while yielding subtypes with enhanced biological interpretability and clinical coherence. This work establishes a novel paradigm for precision prognosis and targeted intervention.

Technology Category

• Machine Learning: Deep Generative Models & Autoencoders
• Search and Optimization: Mixed Discrete/Continuous Search
• Reasoning under Uncertainty: Relational Probabilistic Models

Application Category

• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Web Mining and Content Analysis: Models for Web evolution

Abstract

Modeling disease progression through multiple stages is critical for clinical decision-making for chronic diseases, e.g., cancer, diabetes, chronic kidney diseases, and so on. Existing approaches often model the disease progression as a uniform trajectory pattern at the population level. However, chronic diseases are highly heterogeneous and often have multiple progression patterns depending on a patient's individual genetics and environmental effects due to lifestyles. We propose a personalized disease progression model to jointly learn the heterogeneous progression patterns and groups of genetic profiles. In particular, an end-to-end pipeline is designed to simultaneously infer the characteristics of patients from genetic markers using a variational autoencoder and how it drives the disease progressions using an RNN-based state-space model based on clinical observations. Our proposed model shows improvement on real-world and synthetic clinical data.

Problem

• Modeling heterogeneous disease progression patterns for chronic diseases.
• Integrating genetic profiles and environmental effects for personalized models.
• Improving clinical decision-making with data-driven disease progression predictions.

Innovation

• Personalized model using genetic and clinical data
• Variational autoencoder for genetic marker analysis
• RNN-based state-space model for disease progression