AI Predicts Death and Complications in Angioplasty and Stent Patients
Posted on 17 Jan 2024
Percutaneous coronary intervention (PCI) is a minimally invasive procedure used to treat blocked heart arteries. Traditionally, during PCI, blocked arteries are cleared by inflating a balloon and potentially inserting a stent to enhance blood flow from the heart. Although this procedure is less risky than open-heart surgery, it can still lead to complications such as bleeding and kidney injury. Recognizing these risks, a team of researchers has developed a new AI-powered algorithm that can accurately predict mortality and complications following a PCI. This innovative tool holds promise for aiding clinicians in making more informed treatment decisions.
Several risk stratification tools have been developed to identify risk after PCI, although most are modestly accurate and were created without involving patients. The research team at Michigan Medicine (Ann Arbor, MI, USA) set out to develop a more accurate risk stratification tool, incorporating patient data into the design process, unlike previous models. The research team gathered comprehensive data on all adult patients who underwent PCI from April 2018 to the end of 2021. This data was sourced from the Blue Cross Blue Shield of Michigan Cardiovascular Consortium (BMC2) registry, a network of hospitals throughout Michigan that uses collective data to enhance care quality and patient outcomes.
Utilizing over 20 pre-procedural characteristics, including factors like age, blood pressure, and total cholesterol, the team employed the machine learning software "XGBoost" to construct a risk prediction model. This AI-driven algorithm demonstrated high accuracy in predicting deaths, major bleeding events, and the necessity for blood transfusions, surpassing other models that used similar pre-procedural characteristics. To make this advanced technology widely accessible, it has been integrated into both computer and phone applications, available for free use. This development represents a significant step forward in improving clinical decision-making for patients undergoing PCI.
"Precise risk prediction is critical to treatment selection and the shared decision-making process,” said lead David E. Hamilton, M.D., a cardiology-critical care fellow at Michigan Medicine. “Our tool can recognize a wide array of outcomes after PCI and can be used by care providers and patients together to decide the best course of treatment."
"In the age of widespread smartphones and electronic medical records, this computerized risk score could be integrated into electronic health systems and made easy to use at the bedside,” added senior author Hitinder Gurm, MBBS, interim chief medical officer at U-M Health. “It would not only help relay complex information to the provider quickly, but it could also be used to enhance patient education on the risks related to PCI."
Related Links:
Michigan Medicine
