A Machine Learning Approach to Identifying Depression in Coronary Artery Disease Patients Using Radial Artery Pulse Wave Analysis
Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
Background Coronary artery disease (CAD) is a major global cardiovascular health threat and the leading cause of death in many countries. The disease has a significant impact in China, where it has become the leading cause of death. More than 200 studies have evaluated depression as a risk factor for cardiac events in patients with established CAD. There is an urgent need to develop objective, simple, and cost-effective techniques for the detection of potential depression in CAD patients using machine learning (ML). Methods 228 participants were divided into three groups: healthy, CAD, and depressed CAD. The raw data of pulse wave from those participants was collected. The data were de-noised, normalized, and analyzed using several applications. Seven ML classifiers were used to model the processed data, including Decision Tree (DT), Random Forest (RF), Gradient Boosting (GB), Extra Trees (ET), eXtreme Gradient Boosting (XGBoost), Light Gradient Boosting (LightGBM), and Unbiased Boosting with Categorical Features (CatBoost). Results The ET classifier demonstrated the best classification performance. After tuning hyperparameters, the results performance evaluation on test set are: 0.8261 accuracy, 0.9187 AUC, 0.8245 recall, 0.8255 precision, 0.825 F1-score, and 0.7398 MCC. The top 10 feature importances of tuned ET model are h f /4 , t 3 / t max , t f /6 / t 4 , t f /5 , t 4 / t max , t max / t , w , As , t 4 / t 1 , t 3 / t 1 . The top 20 features of SHAP value are: t 3 / t max , t f /6 / t 4 , h f /4 , t 3 / t 1 , t 4 / t max , t f /5 , w / t max , w / t 1 , w , t max / t , t 4 / t 1 , h f /3 , t 5 / t max , As , h f /5 , h f /6 , t f /3 / t max , t f /6 / t 1 , t f /4 / t 1 , and h 4 . Conclusion Radial artery pulse wave can be used to identify healthy, CAD and depressed CAD participants by using ET classifier. This method provides a potential pathway to recognize depressed CAD patients by using an objective, simple, and cost-effective technique.
