Researchers at the University of Utah's Department of Psychiatry and Huntsman Mental Health Institute have developed RiskPath, an open-source AI toolkit that utilizes Explainable Artificial Intelligence (XAI) to predict chronic diseases before symptoms appear.

RiskPath offers significant advantages, including the ability to map disease progression, streamline risk assessment with just 10 key factors for most conditions, and provide practical visualizations of risk factors throughout a person's life.

The toolkit boasts an impressive prediction accuracy of 85-99%, far surpassing current medical systems that only identify at-risk patients 50-75% of the time.

The research team validated RiskPath using three major long-term patient cohorts, successfully predicting conditions such as depression, anxiety, ADHD, hypertension, and metabolic syndrome.

Nina de Lacy, the study's first author, highlights that early identification of high-risk individuals can lead to more effective preventive healthcare strategies.

The team is exploring the integration of RiskPath into clinical decision support systems and aims to expand its application to additional diseases and diverse populations.

RiskPath employs advanced time-series AI algorithms to analyze extensive health data over multiple years, providing insights into the interaction and significance of various risk factors over time.

The Huntsman Mental Health Institute is committed to addressing mental health and substance use disorders through innovative research and integrated care, supported by a substantial $150 million gift from the Huntsman family.

The full study detailing RiskPath was published in the April 2025 issue of Patterns and received support from the National Institute of Mental Health.