Artificial intelligence is rapidly moving from concept to clinic in cardiovascular prevention, prompting the American Society for Preventive Cardiology (ASPC) to launch a new working group on artificial intelligence (AI) and health technology aimed at shaping how these tools are safely and effectively integrated into practice. As AI-driven risk prediction, imaging analysis and wearable technologies gain traction, the group seeks to position AI at the forefront of innovation in preventive cardiology, while balancing enthusiasm with rigorous evidence and clinical oversight. Central to its mission is fostering a community of experts to generate high-impact research, provide practical guidance on validation and implementation, and mentor the next generation of clinicians navigating this evolving landscape. With growing recognition of challenges such as algorithmic bias, data limitations and regulatory uncertainty, the working group also emphasizes the need for thoughtful guardrails to ensure equitable and accountable use of AI. Co-Chair of the new working group, Dr Ashish Sarraju, Associate Staff in Preventive Cardiology at the Miller Heart, Vascular and Thoracic Institute at the Cleveland Clinic, and a researcher with the Cleveland Clinic Coordinating Center for Clinical Research (C5research), Cleveland, OH, USA, spoke to us about the new initiative, the complexity of the use AI in the field and his excitement for the future of healthcare technology in preventive medicine.
What was the basis of the creation of the American Society for Preventive Cardiology (ASPC) working group on artificial intelligence (AI) and healthcare technology, and what do you hope to achieve?
AI and health technologies are here in preventive cardiology. They have potentially key roles to play in improving the detection and management of cardiovascular risk factors. I believe that preventive cardiology is the perfect field for the integration of these technologies into evidence generation and healthcare delivery. So, I think the ASPC, as the leading national preventive cardiology society, can be at the forefront of using innovation to improve preventive cardiology care. The goal with this working group is establish a community where leaders can think about where and how AI can be incorporated, write and disseminate thoughtful and high-impact evidence and opinions that educate the field widely, develop thought leadership in AI and health technologies, and mentor the next generation of preventive cardiologists.
As AI-driven risk prediction models become more integrated into preventive cardiology, how is the ASPC guiding clinicians on validating and implementing these tools in routine cardiovascular care?
The key here is to balance enthusiasm with rigorous evidence generation. We are ensuring that we come together to discuss not only the potential for AI and health tech to improve care, but also the pitfalls and guardrails needed for successful care. For example, the American Journal of Preventive Cardiology has published data on how generative AI can easily be prompted to provide inaccurate information regarding preventive cardiology topics that we know our patients are already asking chatbots.1
What evidence do we currently have that AI-enabled imaging analysis such as automated coronary artery calcium scoring or echocardiographic interpretation improves diagnostic accuracy and clinical outcomes, compared with standard cardiology workflows?
There is lots of excitement about AI to improve workflows and interpretation for cardiac imaging. Key data, for example, include a blinded, randomized trial studying AI versus sonographers for left ventricular ejection fraction assessment, finding that AI assessment was non-inferior.2 We need more such trials that directly evaluate AI within “real-world” workflows against the standard of care, so that we can get a direct assessment of AI’s strengths and limitations.
How should clinicians think about algorithmic bias and health equity when deploying AI tools in cardiology, particularly in populations historically underrepresented in cardiovascular research datasets?
This is an area that I think deserves continuous highlighting. Algorithmic bias has been something that many of our clinical tools, such as risk scores, as susceptible to, and AI is not immune to this. Clinicians should be, at minimum, familiar with the idea that AI technologies may be limited by their training and validation data. So, clinicians should exercise care and clinical judgement when applying AI models broadly, just as we would with risk scores that may not have included historically underrepresented populations. For example, if a model is deployed to predict a certain outcome or flag “high-risk” patients, clinicians should ask whom the models were developed on and whom they may not apply to well. The synthesis and critical lens of a clinician is the ultimate requirement when applying any technology to the care of an individual patient.
In the era of wearable devices and remote patient monitoring, what are the most promising applications of AI for translating continuous physiologic data into actionable insights for preventive cardiology practice?
Using AI and health technologies, one “low-hanging fruit” is to encourage people to improve risk factors. For example, we need to leverage the use of app-based exercise and nutrition coaching, blood pressure monitoring, remote management of lipids, and virtual/mobile cardiac rehabilitation. Beyond this, there is potential for studies to understand whether novel, continuous physiologic data can be used to improve an individual’s health. For example, can wearable devices help understand an individuals’ environmental exposures, i.e. their “exposome”, to improve cardiovascular risk? Can sleep data be leveraged to understand additional opportunities to improve an important risk factor that is often ignored? These are exciting questions.
From a regulatory, reimbursement, and medico-legal standpoint, what frameworks are emerging to define accountability when AI-supported clinical decision tools influence cardiovascular treatment decisions?
I think that ultimately, clinicians should still expect to be held responsible for individual medical decisions whether made with or without the help of AI and other technologies. The regulatory environment will probably evolve around AI models depending on their scope. The US Food and Drug Administration (FDA)’s 2026 guidance on clinical decision support software, for example, notes that software that is not intended to replace a healthcare provider (HCP)’s judgement and allows HCPs to assess the underlying basis of any recommendation may NOT fall under the FDA’s enforcement as a device, whereas a software that provides a recommendation for which the HCP is unable to access the underlying basis would fall under the FDA’s enforcement.3 From a reimbursement standpoint, we are already seeing AI-enabled devices incorporated into reimbursement models, for example, for AI-enabled plaque analysis.
References
- Al-Dalakta A, Honnekeri B, Rodriguez F, et al. Inaccurate information regarding cardiovascular disease prevention enabled by generative artificial intelligence. Am J Prev Cardiol. 2026;25:101404. DOI: 10.1016/j.ajpc.2025.101404
- He B, Kwan AC, Cho JH, et al. Blinded, randomized trial of sonographer versus AI cardiac function assessment. Nature. 2023;616:520-4. DOI: 10.1038/s41586-023-05947-3.
- US Food & Drug Administration. Clinical Decision Support Software – Guidance for Industry and Food and Drug Administration Staff. 2026. Available at: https://www.fda.gov/media/109618/download (accessed 24 March 2026).
Cite: Sarraju A. Harnessing AI in Preventive Cardiology: ASPC Introduce new Working Group on AI and Health Technology. touchCARDIO. 30 March 2026.Editor: Heather Hall, Managing Editor
Disclosures: Ashish Sarraju wishes to disclose collaboration with Hippocratic AI for a pilot study (no personal compensation). This interview was conducted in collaboration with the American Society for Preventive Cardiology (ASPC). This article was edited by the touchCARDIO team utilizing AI as an editorial tool (ChatGPT (GPT-4o) [Large language model]. https://chat.openai.com/chat.) The content was developed and edited by human editors. No funding was received in the publication of this article.
The American Society for Preventive Cardiology (ASPC) is a nonprofit organisation dedicated to promoting the prevention of cardiovascular disease by educating healthcare professionals and patients. It works to advance cardiovascular health by providing evidence-based information, training, and advocacy to improve prevention, diagnosis, and management of heart disease risk factors. Their upcoming congress, the ASPC 2026 Congress on CVD Prevention is 31 July to 2nd August, 2026. Find out more at their website: https://www.aspconline.org/
Related content:
Already registered? Login below.
SIGN UP to touchCARDIOLOGY!
Join our global community today for access to thousands of peer-reviewed articles, expert insights and learn-on-the-go education across 150+ specialties, plus concise email updates and newsletters so you never miss out.
