Last updated:
ID:
773784
Start date:
19 November 2025
Project status:
Current
Principal investigator:
Professor Lixin Na
Lead institution:
Shanghai University of Medicine and Health Sciences, China

Research questions
1. What novel, multimodal risk factors are associated with the development of specific comorbidities in individuals?
2. How do interactions between traditional risk factors and emerging factors influence comorbidity trajectories?

Objectives
1. To identify potential risk factors through genomics, biochemical profiles, and environmental exposures associated with comorbidity development.
2. To develop a machine learning model integrating multimodal data to predict 5-year comorbidity risks with >85% accuracy.
3. To validate the clinical utility of identified factors through longitudinal cohort analysis.

Scientific rationale for the research
1. Multimodal data capture hidden interactions: Comorbidities likely emerge from nonlinear interactions between factors such as genetic susceptibility, lifestyle behaviors and environmental triggers. Single-modality analyses cannot detect these cross-domain synergies.
2. Machine learning enables pattern discovery: Deep learning architectures, such as graph neural networks, can model high-dimensional relationships in multimodal data, identifying novel risk clusters.
3. Early prediction enables precision prevention: By shifting focus from disease treatment to pre-symptomatic risk mitigation.