Researchers from Michigan Medicine have created an artificial intelligence (AI) tool that is capable of predicting the outcomes following percutaneous coronary intervention (PCI) — the treatment for clearing clogged arteries.

The PCI process — which often involves inserting a small balloon or a metal mesh tube dubbed a stent into a patient’s blood vessel to improve blood flow to the heart — is not without risks, however. The process can sometimes be linked to bleeding, kidney damage and even death.

Source: Pixabay

The team created a tool based on a machine learning algorithm called XGBoost that relies on data from roughly 20 factors — including age, cholesterol levels and blood pressure, among others — to determine a patient’s risk score, and, subsequently, their chances of dying or developing complications following a PCI procedure.

The team tested the tool on about 100,000 patients who underwent PCI procedures in Michigan between 2018 and 2021. According to its developers, the tool outperformed similar methods for predicting outcomes like death, the need for blood transfusion or bleeding following a PCI.

The tool, which is available as a web app and a mobile app, is detailed in the article, “Merging machine learning and patient preference: a novel tool for risk prediction of percutaneous coronary interventions,” which appears in the European Heart Journal.