Clinical, demographic, behavioral, and social determinants combined into a single ensemble
Delivered to physicians, researchers, and population health teams for outreach and intervention
Risk drivers surfaced across ethnicities and races within the patient population
An academic medical center needed better decision support for physicians, population health teams, and researchers working on cardiovascular disease, the kind that could identify at-risk patients early enough for meaningful intervention.
Visibility into risk factors was limited, the relevant data was fragmented across clinical and operational systems, and the institution lacked a model that combined clinical signals with the behavioral and social determinants known to drive cardiovascular outcomes.
The team also needed to understand how risk associated with multilevel determinants varied across ethnicities and races in the patient population, the foundation for outreach and intervention that would reach the people most likely to benefit.
OneSix designed and built an AI/ML cardiovascular risk prediction system grounded in the institution's own clinical, demographic, behavioral, and social determinants data. The work began with a structured discovery phase to identify, acquire, and assess the available data sources for fit, followed by a data-specific hypothesis that subjects with identified predictors carry meaningfully higher CVD risk.
OneSix corrected data incompleteness and quality gaps as part of the build, identified the key predictor variables across clinical, demographic, behavioral, and social determinants, and engineered a diverse ensemble statistical model that distinguishes risk factors from protective factors and isolates potentially modifiable variables. Data ingestion, manipulation, analysis, and modeling were performed in R, with the output delivered as an AI/ML CVD prediction calculator for decision support, research, and population health outreach.
The model is positioned for embedding into production systems, including a Power BI dashboard inside Epic and the incorporation of real-time data from wearables, extending the calculator from a research instrument into a clinical and operational tool that supports early intervention at scale.
More to explore
OneSix implemented a scalable data warehouse solution for a digital health company, enabling efficient, accurate reporting and data-driven insights into member engagement across texting programs.
OneSix developed an AI-powered computer vision system for Sunsweet Growers, automating prune inspection with real-time defect detection while preserving high product quality standards.
OneSix built an AI product recommender that maximizes conversion and lets non-engineering teams tune it through low-code tooling.