AI system predicts patient health issues 16 hours in advance: Study

New Delhi: Across India’s vast healthcare network, which relies heavily on more than two million hospital beds, traditional methods of patient monitoring in general wards largely depend on periodic manual checks. This conventional system often leads to delayed detection of health deterioration, leaving many vulnerable patients at risk.

A recent study published in Frontiers in Medical Technology (FMT) indicates that this scenario might soon change, as an AI-based Early Warning System (EWS) has shown promising potential to detect early signs of health deterioration up to 16 hours in advance. Conducted at King George’s Medical University (KGMU), the study highlights the benefits of continuous health monitoring in non-ICU settings, where periodic spot checks have historically limited early intervention.

The AI-powered EWS, which monitors vital signs continuously, issues alerts to healthcare providers well before a critical emergency. This advance notice could be life-saving, enabling faster responses and transforming treatment approaches in general wards.

The study suggests that healthcare workers will also benefit, as the system could reduce their daily workload by about 2.4 hours per person. Over 700 patients were monitored for more than 85,000 hours, with the EWS’s alerts demonstrating high sensitivity, specificity, and timely responsiveness.

Dr. Himanshu Dandu, Professor in the Department of Medicine at KGMU, noted, “The ability to detect signs of patient health deterioration can significantly improve their survival rates.”

Results showed that the EWS successfully predicted health issues in approximately 67-94 per cent of cases. Based on these findings, researchers estimate the system’s implementation could save up to 2.1 million lives annually and reduce healthcare costs by about ₹6,400 crore.