Researchers at the University of Utah have developed RiskPath, an AI tool that predicts chronic and progressive diseases with impressive accuracy. This tool sifts through years of health data to pinpoint those at risk, potentially enabling earlier intervention and better health outcomes. If you’ve ever wondered how technology might help catch problems before they become severe, RiskPath shows promise by offering an accuracy rate between 85% and 99%.
RiskPath uses advanced time-series algorithms to track how risk factors evolve over time—a challenging task in long-term medical studies. It employs explainable AI to make its decision process clear, so clinicians can understand why a certain prediction was made. As Nina de Lacy, a psychiatry professor and executive committee member of the One-U Responsible AI Initiative, puts it, the goal is to offer explanations that make sense to human experts.
This tool marks a significant improvement over traditional methods, which often correctly identify at-risk individuals only about half to three-quarters of the time. By mapping changing risk factors, RiskPath not only detects high-risk individuals before symptoms appear but also highlights the life stages when key interventions are most effective.
The research team validated RiskPath using data from three major long-term patient cohorts, successfully predicting eight conditions ranging from depression and anxiety to hypertension and metabolic syndrome. The tool’s approach is both practical and insightful, offering clear visualisations of risk and focusing on the most critical factors in chronic disease development.
While RiskPath is currently a research tool, its developers hope it will soon transition into clinical practice, helping healthcare providers better manage and prevent chronic diseases. This shift towards preventative care comes at a time when keeping healthcare issues at bay before they escalate is more crucial than ever.
