Evolving Patient Risk Profiles using integrated Clinical and Genomic risk
Identifying patients at higher risk of common diseases like Cardio Vascular Disease and Diabetes before they fall ill has recently become possible using a combination of artificial intelligence algorithms and patient 'big-data' like clinical histories, lifestyle and environmental information. At present such algorithms can typically provide clinicians with 6-9 months advance notice of diagnosis. Simultaneously the advent of large scale genomic studies have also led to breakthroughs in prediction of patient risk for the same common diseases, using purely genomic data. The predictive accuracy, however of these genomic scores currently falls short of the standards required for clinical use.
We propose to apply advanced data science methods previously used for disease prediction with clinical data alone to jointly analyse genomic and medical data to produce an integrated patient risk profile that can be used to make much earlier diagnostic decisions.