UKB-MDRMF: a multi-disease risk and multimorbidity framework based on UK biobank data

Area of Science:

• Biomedical data science
• Computational biology
• Epidemiology

Background:

• Biomedical cohort data is growing, offering insights into disease but often limited by narrow focus and fragmented analysis.
• Current research struggles to explore comprehensive risk factors and inter-disease correlations due to process fragmentation and time constraints.

Purpose of the Study:

• To develop a comprehensive framework for predicting and assessing health risks across a large number of diseases.
• To address limitations in current research by integrating multimodal data and considering multimorbidity mechanisms.

Main Methods:

• Integration of multimodal data from the UK Biobank, including basic, lifestyle, measurement, environment, genetic, and imaging data.
• Development of UKB-MDRMF, a framework designed for predicting and assessing health risks across 1560 diseases, incorporating multimorbidity.
• Comparison with single disease models to evaluate predictive accuracy and risk assessment performance.

Main Results:

• UKB-MDRMF demonstrates superior predictive accuracy compared to single disease models, with improved performance across all disease types.
• The framework successfully incorporates multimorbidity mechanisms, enhancing the assessment of health risks.
• Joint prediction and assessment of multiple diseases revealed shared and distinctive connections among risk factors and diseases.

Conclusions:

• UKB-MDRMF provides a comprehensive approach to health risk prediction and assessment by integrating diverse data and considering multimorbidity.
• The framework offers a broader perspective on health and multimorbidity mechanisms by uncovering complex disease interconnections.
• This approach enhances the utility of large-scale biomedical data for understanding disease etiology and developing personalized health strategies.