top of page
Search
  • Clare Williams

Artificial Intelligence’s First Cardiologist: How Algorithms Can be Used to Improve Risk Assessment


How can AI help us tackle the persisting problems in the medical field today? Can AI diagnose more accurately than a trained physician? What is AI not capable of accomplishing? As we near the end of 2023, these questions appear to be more relevant within the world of healthcare than ever before. As demonstrated by recent applications in the field of cardiology, AI has the potential to revolutionize preventative care and treatment, but only if utilized carefully and equitably.


A new study from Cedars-Sinai, a prominent hospital in the heart of Los Angeles, has presented a groundbreaking AI coronary imaging device that is aiding cardiologists in plaque and stenosis quantification. Due to its quick response times and extreme precision, this new technology is providing promising insights into a future focused on cardiac risk prediction (1). Damini Dey, PhD., pioneer of the study and Director of the Biomedical Quantitative Image Analysis Lab at Cedars-Sinai, explains how fundamental these prevention-based practices are in reducing the number of cardiac-related deaths. She elaborates that the first sign of plaque buildup within coronary arteries is often chest pain that precedes a heart attack, which is often misattributed by the patient as anxiety or a panic attack. This makes it particularly challenging, time-consuming, and costly to identify high-risk patients with immediate cardiac complications. Consequently, plaque buildup can rapidly develop and cause arteries to narrow without proper treatment, which makes it difficult for blood to reach the heart and keep the body alive. For this reason, it is important to investigate the ways in which potential preventative measures can be implemented.


These statements are further supported by data from an international study called the SCOT-HEART trial, in which investigators confirmed the accuracy of measurements made by this AI algorithm from CTA (computed tomography angiography) images that predicted heart attack risk within five years for approximately 1,611 participants (2). The algorithm outlines the coronary arteries in 3D images, which allows it to identify the presence of blood and plaque buildup within the vessel (1). Results from the algorithm are also comparable to two existing invasive tests, intravascular ultrasound and catheter-based coronary angiography, that are used to analyze plaque deposits and artery narrowing.These deep-learning AI systems showed significant time reductions when compared to analysis by imaging specialists and physicians (1).


“This novel research could have broad applications,” claims Sumeet Chugh, M.D., Director of the Division of Artificial Intelligence in Medicine at Cedars-Sinai and the Center for Cardiac Arrest Prevention in the Smidt Heart Institute (3). Chugh further explains that “AI algorithms enable physicians to communicate more personalized information regarding potential timing of imminent heart disease events, allowing patients to engage more meaningfully in the shared decision-making process.” Even more importantly, this tool has placed a needed sense of prioritization and urgency on heart disease prevention efforts by both patients and providers.


However, in order for these data-led initiatives to make visible impacts, patients must be willing to instill complete trust in these new AI tools. Dr. Liu Nan, Associate Professor of Quantitative Medicine at Duke-NUS School of Medicine, credits the current lack of trust within the rapidly expanding medical technology to its absence within actual clinics (4). This is especially true for resource-constrained communities whose hospitals cannot afford this expensive equipment or specialized expertise.


How can we encourage trust in these new technologies for patients who face barriers in accessing it? With appropriate implementation into isolated and resource-lacking communities, we can use AI to facilitate the development of more affordable, higher quality, and accessible preventative healthcare. Amidst much uncertainty in the digitalization within the world of medicine, it is imperative that we do not lose sight or hope in our ability to provide equitable pathways to healthcare for patients who have been historically excluded from access to these preventative systems.


Reviewed by: Camille Krejdovsky

Graphic by: Ariha Mehta


Works Cited:


  1. Cedars-Sinai Medical Center. (2022, March 30). Artificial Intelligence Tool May Help Predict Heart Attacks. Artificial Intelligence Tool may help predict heart attacks. https://www.cedars-sinai.org/newsroom/artificial-intelligence-tool-may-help-predict-heart-attacks

  2. Lin, A., & Manral, N. (2022, April). Deep learning-enabled coronary CT angiography for plaque and stenosis ...A. The Lancet. https://www.thelancet.com/journals/landig/article/PIIS2589-7500(22)00022-X/fulltext

  3. Cedars-Sinai Medical Center. (2023, August 9). Can Artificial Intelligence Predict Heart Attack Risk?. Can artificial intelligence predict heart attack risk? https://www.cedars-sinai.org/newsroom/study-artificial-intelligence-may-predict-heart-attacks/

  4. Duke-NUS Medical School. (2023, April 13). Clinical trial of Duke-NUS, SGH co-developed AI system underway following HSA approval. https://www.duke-nus.edu.sg/allnews/news-highlights/aitriage


0 comments

Comentários


bottom of page