- The Smidt Heart Institute at Cedars-Sinai has trained an AI model to analyze pre-operative electrocardiograms (ECGs) to predict how patients will fare post-surgery.
- The AI algorithm was used on patients undergoing a variety of surgical procedures and was correlated with post-operative outcomes.
- While the model identified most patients as low risk, others flagged as high risk had almost nine times the likelihood of post-operative mortality.
- The AI tool could be integrated into a web-based application, making it accessible to medical professionals and patients alike.
Researchers at the Smidt Heart Institute at Cedars-Sinai have developed an artificial intelligence (AI) algorithm that uses electrocardiograms (ECG) to accurately predict post-operative outcomes. The AI model analyses pre-operative ECGs, a standard test that records the heart’s electrical activity, to predict how patients will fare after surgery.
The AI model was put to the test by correlating pre-surgical or pre-procedural ECGs of patients with their subsequent post-operative outcomes. The AI algorithm was tasked with detecting correlations or patterns within the ECG waveforms. Patients flagged as high risk by the AI model had an almost nine times higher likelihood of post-operative mortality, compared to those categorized as low risk.
As it stands, clinicians tend to predict the risk of surgery based on medical society guidelines, but these tools are deemed “insufficient” by Cedars-Sinai researchers. They believe the AI model could greatly enhance these predictions. The researchers are now exploring how to adapt this AI algorithm into a web-based application to make it universally accessible.
“This AI model could potentially be used to determine exactly which patients should undergo an intervention and which patients might be too sick,” added Dr. David Ouyang, a cardiologist in the Department of Cardiology in the Smidt Heart Institute at Cedars-Sinai.