Machine learning model improves mortality risk prediction in cardiac surgery
Click Here to Manage Email Alerts
Key takeaways:
- A machine learning-based model more accurately predicted mortality after cardiac surgery than population-based models.
- The accuracy was consistent across five surgery types.
A machine learning-based model appeared to improve prediction of mortality risk for patients undergoing cardiac surgery compared with population-derived models, researchers reported.
“The standard-of-care risk models used today are limited by their applicability to specific types of surgeries, leaving out significant numbers of patients undergoing complex or combination procedures for which no models exist,” Ravi Iyengar, PhD, the Dorothy H. and Lewis Rosenstiel Professor of Pharmacological Sciences at the Icahn School of Medicine at Mount Sinai and director of the Mount Sinai Institute for Systems Biomedicine, said in a press release. “Our team rigorously combined electronic health record data and machine learning methods to demonstrate for the first time how individual institutions can build their own risk models for post-cardiac surgery mortality.”
From a cohort of patients who underwent cardiac surgery between 2011 and 2016, the researchers developed “a data-driven, institution-specific machine learning-based model inferred from multi-modal electronic health records and compared performance with the Society of Thoracic Surgeons (STS) models.”
The cohort included 6,392 patients who were described by 4,016 features and randomly assigned to the training/development cohort (75%) or the test/evaluation cohort (25%).
The researchers developed a machine learning framework using routinely collected EHR data to build a postsurgical mortality risk prediction model that would be personalized to a patient and specific to a hospital, in contrast to the STS population-based models that are derived from data from diverse populations from many different locations.
The researchers trained four machine learning classification algorithms on the imputed data.
The best-performing predictor among those was the eXtreme Gradient Boosting (XGBoost) algorithm, according to the researchers.
Iyengar and colleagues wrote that XGBoost performed well (F-measure = 0.775; precision = 0.756; recall = 0.795; accuracy = 0.986; area under the receiver operating characteristic curve = 0.978; area under the precision-recall curve = 0.804) and outperformed the STS population-based models for evaluating five different surgical procedures.
“Accurate prediction of postsurgical mortality is critical to ensure the best outcomes for cardiac surgery patients, and our study shows that institution-specific models may be preferable to the clinical standard based on population data,” Gaurav Pandey, PhD, associate professor of genetics and genomic sciences at Icahn Mount Sinai, said in the release. “Just as importantly, we’ve demonstrated that it’s practical for health care institutions to develop their own predictive models through sophisticated machine learning algorithms to replace or complement the established STS template.”
Reference:
- Mount Sinai. https://www.eurekalert.org/news-releases/989057. Published May 17, 2023. Accessed May 17, 2023.