Artificial intelligence (AI)—the use of algorithms and software to interpret and understand complex data—has proven to be a disruptive force in modern medicine and has impacted everything from biomedical research to clinical practice. The rapid development of AI tech has seen its incorporation into increasingly specific applications and clinical settings. In particular, the field of cardiology has seen AI integrated with a variety of tools and devices in recent years and presents new opportunities for data-driven improvements to patient care.
Today’s AI applications in cardiology aim to enhance physician decision-making through data analysis. Interpretation of test results such as electrocardiograms or angiograms can be guided by AI algorithms trained to detect abnormalities and reduce diagnosis time. AI can also be used to determine patient risk for chronic conditions like heart failure, promoting earlier treatment and informing care plans.
Investment into AI-driven cardiology products has seen significant activity in recent years, with over 151 deals involving AI in the cardiovascular devices market since 2020. Cardiovascular devices AI deals hit an all-time high last year, totalling nearly $1.6bn. In the previous month alone, several notable industry deals and partnerships centred around AI in cardiology. Developer Cleerly announced that it had raised $192m in a series C financing round for its AI-driven diagnostic platform for coronary atherosclerosis. Earlier in the month, Anumana partnered with Novartis to drive the development of its electrocardiogram (ECG) AI algorithms for the detection of subclinical heart disease. Researchers from Cornell University also announced the launch of the Cardiovascular AI Initiative: a three-year, $15m collaboration with the New York Presbyterian Hospital that aims to employ AI and machine learning techniques to inform heart failure treatment and prevention strategies.
The continued investment in the development of AI-based cardiology tools reflects the recognised potential of AI among manufacturers and cardiologists alike. Challenges will remain in validating AI algorithms in large, diverse study populations to avoid biases, potential regulatory barriers and data privacy concerns. Significant research efforts will be needed to support its adoption into practice, as patient benefits will need to be clearly demonstrated, and diagnostic accuracy will need to be exceptionally high.