AI Breakthrough: Predicting Sudden Cardiac Death and Saving Lives
A groundbreaking study on predicting sudden cardiac death using artificial intelligence (AI) has given rise to the possibility of saving lives and advancing global health strategies. Initial research, set to be presented at the American Heart Association’s Resuscitation Science Symposium 2023 in Philadelphia, highlights the potential of AI in identifying high-risk individuals and preventing sudden cardiac death.
Sudden cardiac death poses a significant public health burden, accounting for 10% to 20% of all deaths. Traditional approaches to predicting such events often fail to accurately identify individuals at risk, particularly on an individual level. However, this new study proposes an innovative approach that looks beyond standard cardiovascular risk factors and takes into account all available medical information from electronic health records.
Led by Dr. Xavier Jouven, a professor of cardiology and epidemiology at the Paris Cardiovascular Research Center, the research team utilized AI to analyze medical information from registries and databases in Paris, France, and Seattle. The data comprised 25,000 individuals who had experienced sudden cardiac arrest and 70,000 individuals from the general population, with matching age, sex, and residential area. In total, over 1 million hospital diagnoses and 10 million medication prescriptions were included in the analysis, covering a period of up to ten years prior to each death.
The AI analysis involved building approximately 25,000 equations using personalized health factors to identify individuals at high risk of sudden cardiac death. Furthermore, the researchers developed a customized risk profile for each participant in the study. By considering various medical details such as treatment for high blood pressure, history of heart disease, and mental and behavioral disorders like alcohol abuse, the AI system could identify factors that would increase or decrease the risk of sudden cardiac death within a specific timeframe. For example, the analysis could indicate an 89% risk of sudden cardiac death within three months.
Remarkably, the AI analysis successfully identified individuals with a more than 90% risk of sudden cardiac death, accounting for over one-fourth of all cases.
Dr. Xavier Jouven expressed surprise at the level of accuracy achieved by the study, stating, We have been working for almost 30 years in the field of sudden cardiac death prediction, however, we did not expect to reach such a high level of accuracy. We also discovered that the personalized risk factors are very different between the participants and are often issued from different medical fields. Jouven emphasized that while medical treatments to address risk factors exist, the use of AI is crucial to identify a series of medical information over time that corresponds to an increased risk of sudden cardiac death. The hope is that with a personalized list of risk factors, patients can work with their healthcare providers to reduce these risks and ultimately decrease the likelihood of sudden cardiac death.
Although the study presents promising findings, there are limitations to consider. These include the applicability of the prediction models beyond this specific research and the potential differences in medical data collected among countries, requiring adaptation of the models.
The potential impact of AI in predicting and preventing sudden cardiac death cannot be understated. By harnessing the power of AI and leveraging comprehensive medical data, healthcare providers may be able to intervene earlier, saving lives and improving global health outcomes. The convergence of AI and cardiovascular medicine paves the way for a future where proactive strategies and personalized risk reduction can significantly reduce the global burden of sudden cardiac death.
