Can machine-learning improve cardiovascular risk prediction using routine clinical data?

Summary

Background Current approaches to predict cardiovascular risk fail to identify many people who would benefit from preventive treatment, while others receive unnecessary intervention. Machine-learning offers opportunity to improve accuracy by exploiting complex interactions between risk factors. We assessed whether machine-learning can improve cardiovascular risk prediction. Methods Prospective cohort study using routine clinical data of 378,256 patients from UK family practices, free from cardiovascular disease at outset. Four machine-learning algorithms (random forest, logistic regression, gradient boosting machines, neural networks) were compared to an established algorithm (American College of Cardiology guidelines) to predict first cardiovascular event over 10-years. Predictive accuracy was assessed by area under the ‘receiver operating curve’ (AUC); and sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) to predict 7.5% cardiovascular risk (threshold for initiating statins). Findings 24,970 incident cardiovascular events (6.6%) occurred. Compared to the established risk prediction algorithm (AUC 0.728, 95% CI 0.723–0.735), machine-learning algorithms improved prediction: random forest +1.7% (AUC 0.745, 95% CI 0.739–0.750), logistic regression +3.2% (AUC 0.760, 95% CI 0.755–0.766), gradient boosting +3.3% (AUC 0.761, 95% CI 0.755–0.766), neural networks +3.6% (AUC 0.764, 95% CI 0.759–0.769). The highest achieving (neural networks) algorithm predicted 4,998/7,404 cases (sensitivity 67.5%, PPV 18.4%) and 53,458/75,585 non-cases (specificity 70.7%, NPV 95.7%), correctly predicting 355 (+7.6%) more patients who developed cardiovascular disease compared to the established algorithm. Conclusions Machine-learning significantly improves accuracy of cardiovascular risk prediction, increasing the number of patients identified who could benefit from preventive treatment, while avoiding unnecessary treatment of others.

Key Points

• 1Machine-learning improves cardiovascular risk prediction accuracy
• 2The study analyzed data from over 378,000 patients in the UK
• 3Four machine-learning algorithms were compared to traditional methods
• 4Neural networks provided the highest improvement in predictive accuracy
• 5Enhanced prediction can lead to better-targeted preventive treatment strategies

Related articles

• Neonatal Disorders

  This document provides a comprehensive overview of neonatal disorders, including case scenarios and learning objectives targeted at third-year medical students. It outlines essential management plans and assessment procedures for common neonatal conditions.

• 13. Gross Anatomy: Muscles of the Hand

  This document outlines the anatomy of the intrinsic muscles of the hand, detailing their compartments and clinical significance, particularly in relation to carpal tunnel syndrome and recurrent branch injuries.

• 12. Gross Anatomy: RA Upper Extremities

  This document outlines a detailed guide on gross anatomy focused on the upper extremities, covering the shoulder, humerus, elbow, forearm, and hand/wrist series with an emphasis on radiographic examination techniques and anatomical considerations.

• 11. Gross Anatomy: Compartments of the Forearm

  This document outlines the anatomy of the forearm, detailing the anterior and posterior compartments, including the muscles, their origins, insertions, innervation, and actions. It serves as a comprehensive guide for students studying gross anatomy.

• 10. Gross Anatomy: Bones and Muscles of the Shoulder and Arm

  This document outlines the anatomy of the shoulder and arm, including details on the muscles and bones, their origins, insertions, innervations, and primary actions. It provides learning objectives that aim to enhance students' understanding of arm anatomy.