Facing gaps in off-the-shelf solutions, some health systems are building their own artificial intelligence tools to better address clinical and operational needs.
At Rochester, Minn.-based Mayo Clinic, researchers have created AI tools to identify young children at highest risk for severe asthma and acute respiratory infections. The tools use machine learning and natural language processing to analyze clinical data and apply two standard diagnostic checklists for asthma.
Mayo has also developed the Nurse Virtual Assistant, an in-house AI tool designed to streamline clinical workflows. The system, built by nurses and informatics teams, integrates directly into the electronic health record and provides a nurse-specific patient summary with links to resources such as Lippincott guidelines, intravenous administration protocols and the health system’s clinical policy library. Development began in 2024 as part of Mayo’s strategy to ease administrative burdens in increasingly complex digital environments.
Meanwhile, at New York City-based Icahn School of Medicine at Mount Sinai, researchers built AEquity, a tool aimed at detecting and mitigating biases in datasets used to train machine-learning algorithms. The system identified both well-known and previously overlooked biases.
Mount Sinai researchers also designed an AI model to help determine which atrial fibrillation patients benefit from blood thinners to prevent stroke. The model accurately assessed stroke and bleeding risks at the individual level and recommended against anticoagulation in up to half of patients who would otherwise have received it under current guidelines, according to the health system.
These efforts reflect a broader trend in which health systems are opting to develop in-house AI tools tailored to their patient populations and workflows, rather than relying solely on vendor products that may not fully meet their needs.
The post Health systems take AI development into their own hands appeared first on Becker’s Hospital Review | Healthcare News & Analysis.
Health IT
