Cardiothoracic surgeons use risk analysis to decide which patients would benefit from Surgery to replace a defective heart valve, and artificial intelligence (AI) now provides them a leg up when it comes to assessing risk vs. reward. Dr. Robert Hagberg, Chief of Cardiac Surgery at Hartford Hospital and a member of the Hartford HealthCare Heart & Vascular Institute, was part of a research group that published “Machine learning models for mitral valve replacement: A comparative analysis with the Society of Thoracic Surgeons risk score” in the Journal of Cardiac Surgery recently.
It’s one of the first studies to look into the use of artificial intelligence (AI) to predict surgical outcome. The researchers used an online database of mitral valve replacement cases and outcomes to build algorithms for evaluating patient outcomes, using different AI techniques. The results were then compared to those obtained by a previously developed predictive tool made available by the Society of Thoracic Surgeons (STS).
“The proposed risk models supplemented existing STS models in predicting mortality, extended ventilation, and renal failure, allowing healthcare practitioners to better properly assess a patient’s risk of morbidity and death after having mitral valve Surgery,” the study concludes. The results showed a 3% increase in the accuracy of surgical outcome prediction.
After mitral valve replacement Surgery, he noted, sicker patients are more likely to experience problems like stroke, extended ventilation, renal failure, and death. For these reasons, he believes they should rethink their decision to get. The use of the new prediction models may become normal practise for surgeons in the future, but according to Dr. Hagberg, the research has shown that “we can make improvements.”