A team of University of Florida Health researchers aims to reduce health inequities faced by people living with dementia by using artificial intelligence and electronic health records to develop a machine learning-based social risk management platform.
The development of iSMART, or Intelligent Social Risk Management in Alzheimer’s disease/Alzheimer’s disease-related dementia Patients, is led by principal investigators Serena Jingchuan Guo, M.D. Ph.D., an assistant professor of pharmaceutical outcomes and policy in the UF College of Pharmacy; Jiang Bian, Ph.D., a professor and division director of biomedical informatics in the UF College of Medicine Department of Health Outcomes and Biomedical Informatics and chief data scientist/chief research information officer at UF Health; and Ramzi Salloum, Ph.D., a professor and director of the Division of Implementation Science and Health Interventions in the UF College of Medicine. The National Institutes of Health’s National Institute on Aging awarded the team $3 million for this R01 project.
The research team intends iSMART to be embedded into health systems’ electronic health records in order to improve the quality of care and quality of life experienced by people living with dementia. Guo said this work is critical for people living with dementia from racial and ethnic minority groups, as well as those from socioeconomically disadvantaged environments, who are more likely to face barriers to diagnosis and care.
To create iSMART, the study team will scour real-world data from the OneFlorida+ network, which contains electronic health records of more than 20 million people, as well as leverage their prior efforts in establishing an external exposome database that can extract person-level social determinants of health from clinical narratives. The team’s intervention mapping approach involves partnering with a stakeholder advisory committee to develop machine learning-based social risk management algorithms for dementia care and outcomes, allowing health care professionals to screen for unmet social needs and offer specialized interventions.
iSMART will enable the study of contextual and person-level social determinants of health to identify people living with dementia who possess an increased social risk. The project seeks to improve patient care and health outcomes over the long term, while reducing the burden on physicians and caregivers for customizing their care as their condition and needs change over time.
“The integration of real-world data with powerful analytics allows for a revolutionary change to provide individuals with personalized health care that can respond instantly to the patient’s needs at any time,” Guo said. “The iSMART platform will offer personalized care for patients’ unmet social needs, which are essential to their health outcomes.”