Mortality risk for individual cardiac surgery patients can be predicted using a machine learning-based model developed by researchers at Mount Sinai Hospital in New York.
This new data-driven algorithm builds on a stockpile of electronic health records (EHRs) and is said to be the first institution-specific model for assessing risk in cardiac patients before surgery. , which allows healthcare providers to pursue the best possible treatment. Adapt actions to the individual patient. For more information on team work, Open Journal of Thoracic and Cardiovascular Surgery (JTCVS).
“Standard-of-care risk models in use today are limited by their applicability to certain types of surgery, excluding a significant number of patients undergoing complex or combination surgeries for which models do not exist. said lead author Dr. Ravi Iyengar in the paper. statement. “Our team is the first to demonstrate how an individual hospital can build its own risk model for post-cardiac surgery mortality using a rigorous combination of electronic medical record data and machine learning techniques.”
Predictive models based on machine learning algorithms have been generated across different areas of medicine, and some predictive models have shown improved results over their standard-of-care counterparts.
In cardiac surgery, the Society of Thoracic Surgeons (STS) risk score is considered the gold standard and is routinely used to assess surgical risk in cardiac surgery patients. While these continue to provide important benchmarks for hospitals to assess and improve their performance, they are based on population-level data, requiring customized preoperative assessments and complex surgeries. It may not be possible to accurately predict risk for certain patients with complex medical conditions.
Medical and healthcare details
Cardiovascular surgeons and data science experts at Mount Sinai Hospital from their own institution under the supervision of Dr. Gauraf Pandey, co-senior author and Associate Professor of Genetics and Genomic Sciences at Icahn Mount Sinai. A machine learning-based model using the EHR data of provides an effective solution.
To this end, they created a rigorous machine learning framework using regularly collected EHR data to develop a hospital-specific postoperative mortality risk prediction model for each patient. Did. It incorporates important information about the Mount Sinai patient population, including demographics and socioeconomic data. Factors and health features.
According to Mount Sinai, this contrasts with population-derived models like the STS, which are based on data from diverse healthcare systems across the United States. Further driving the performance of this technique was an open-source prediction algorithm called XGBoost. The algorithm builds a series of decision trees that progressively focus on less predictable subsets of the training data.
The research team used XGBoost to model 6,392 heart surgeries, including heart valve surgeries, performed at Mount Sinai Hospital between 2011 and 2016. Coronary artery bypass graft. Aortic resection, replacement, or anastomosis. and re-heart surgery. The team then compared the model’s performance with the STS model on the same patient set.
This study showed that the XGBoost model outperformed the STS risk score in terms of mortality in all commonly performed categories of cardiac surgery for which the STS score was designed. The predictive performance of his XGBoost model across all surgical types is also high, demonstrating the potential of machine learning and his EHR data for building effective site-specific models.
“Accurately predicting postoperative mortality is critical to ensuring the best outcome for cardiac surgery patients, and our study favors site-specific models rather than clinical standards based on population data.” It shows that it may be favorable,” Dr. Pandey said. “Equally important, we have demonstrated that it is practical for healthcare organizations to develop their own predictive models through sophisticated machine learning algorithms to replace or complement established STS templates. .”
