AI Predicts Cardiac Arrest 10-15 Minutes Early | Medical Xpress

An artificial intelligence (AI) model developed by researchers at the Perelman School of Medicine at the University of Pennsylvania is showing promise in predicting cardiac arrest up to 10 to 15 minutes before it occurs. The model analyzes electrocardiographic (ECG) data to identify subtle patterns indicative of impending cardiac arrest, potentially offering a critical window for intervention.

The research, led by cardiologist Rajat Deo, builds on nearly two decades of study focused on ECG data and cardiac rhythms. According to reporting from Medical Xpress, Deo stated that the ability to predict cardiac arrest even a few minutes in advance could significantly improve patient outcomes.

Predicting Cardiac Arrest with AI

The AI model, known as CAMEL, forecasts dangerous cardiac rhythms before they strike by analyzing long-underused heart imaging data alongside a comprehensive range of medical records. This approach allows the system to reveal previously undetected indicators of risk. The findings, published earlier this year, suggest a new era of real-time, predictive heart care is on the horizon.

Cardiac arrest occurs when the heart suddenly stops beating, cutting off blood flow to the brain and other vital organs. It is a leading cause of death worldwide, and rapid intervention is crucial for survival. Current methods for predicting cardiac arrest rely on identifying patients with known risk factors, such as heart disease or a history of arrhythmias, but these methods are not always accurate.

How the AI Model Works

The CAMEL model utilizes machine learning algorithms to analyze ECG data, which records the electrical activity of the heart. By identifying subtle changes in the ECG signal that precede cardiac arrest, the AI can provide an early warning, potentially allowing medical personnel to intervene before the heart stops beating. The model’s ability to integrate various data points—including heart imaging and complete medical histories—is a key factor in its predictive accuracy.

The development of this AI model represents a significant advancement in cardiac care. Traditional methods often struggle to identify individuals at imminent risk, leading to delayed interventions. This new technology aims to bridge that gap by providing a proactive approach to preventing sudden cardiac death.

Potential Applications and Future Research

The potential applications of this AI model are far-reaching. In hospitals, it could be used to continuously monitor patients at high risk of cardiac arrest, alerting medical staff to potential problems before they escalate. Emergency medical services could also utilize the technology to identify individuals who may be experiencing a pre-arrest condition, allowing for faster and more targeted responses.

Researchers are continuing to refine the AI model and explore its potential for broader applications. Future studies will focus on validating the model’s performance in diverse patient populations and integrating it into clinical workflows. The team also plans to investigate the possibility of using the AI to personalize treatment plans for patients at risk of cardiac arrest.

While the AI model shows considerable promise, it is not a foolproof solution. The technology is still under development, and further research is needed to fully understand its capabilities and limitations. However, the initial findings suggest that AI has the potential to revolutionize the way we approach cardiac arrest prevention and treatment.

The research builds on previous work in the field. A 2021 study demonstrated that machine learning could accurately predict cardiac arrest risk, highlighting the potential for resource deployment and scheduling within emergency medical services. Further research in 2025 underscored the importance of analyzing comprehensive medical records alongside heart imaging to reveal previously hidden risk factors.