Accurate Prediction of Coronary Heart Disease for Patients With Hypertension From Electronic Health Records With Big Data and Machine-Learning Methods: Model Development and Performance Evaluation

Abstract

<jats:sec> <jats:title>Background</jats:title> <jats:p>Predictions of cardiovascular disease risks based on health records have long attracted broad research interests. Despite extensive efforts, the prediction accuracy has remained unsatisfactory. This raises the question as to whether the data insufficiency, statistical and machine-learning methods, or intrinsic noise have hindered the performance of previous approaches, and how these issues can be alleviated.</jats:p> </jats:sec> <jats:sec> <jats:title>Objective</jats:title> <jats:p>Based on a large population of patients with hypertension in Shenzhen, China, we aimed to establish a high-precision coronary heart disease (CHD) prediction model through big data and machine-learning</jats:p> </jats:sec> <jats:sec> <jats:title>Methods</jats:title> <jats:p>Data from a large cohort of 42,676 patients with hypertension, including 20,156 patients with CHD onset, were investigated from electronic health records (EHRs) 1-3 years prior to CHD onset (for CHD-positive cases) or during a disease-free follow-up period of more than 3 years (for CHD-negative cases). The population was divided evenly into independent training and test datasets. Various machine-learning methods were adopted on the training set to achieve high-accuracy prediction models and the results were compared with traditional statistical methods and well-known risk scales. Comparison analyses were performed to investigate the effects of training sample size, factor sets, and modeling approaches on the prediction performance.</jats:p> </jats:sec> <jats:sec> <jats:title>Results</jats:title> <jats:p>An ensemble method, XGBoost, achieved high accuracy in predicting 3-year CHD onset for the independent test dataset with an area under the receiver operating characteristic curve (AUC) value of 0.943. Comparison analysis showed that nonlinear models (K-nearest neighbor AUC 0.908, random forest AUC 0.938) outperform linear models (logistic regression AUC 0.865) on the same datasets, and machine-learning methods significantly surpassed traditional risk scales or fixed models (eg, Framingham cardiovascular disease risk models). Further analyses revealed that using time-dependent features obtained from multiple records, including both statistical variables and changing-trend variables, helped to improve the performance compared to using only static features. Subpopulation analysis showed that the impact of feature design had a more significant effect on model accuracy than the population size. Marginal effect analysis showed that both traditional and EHR factors exhibited highly nonlinear characteristics with respect to the risk scores.</jats:p> </jats:sec> <jats:sec> <jats:title>Conclusions</jats:title> <jats:p>We demonstrated that accurate risk prediction of CHD from EHRs is possible given a sufficiently large population of training data. Sophisticated machine-learning methods played an important role in tackling the heterogeneity and nonlinear nature of disease prediction. Moreover, accumulated EHR data over multiple time points provided additional features that were valuable for risk prediction. Our study highlights the importance of accumulating big data from EHRs for accurate disease predictions.</jats:p> </jats:sec>

Machine Learning and IoT Applied to Cardiovascular Diseases Identification through Heart Sounds: A Literature Review

“…Table 5 presents the list of selected articles, containing a numerical identification, reference, authors' countries, databases, year of publication, and a short description of the research work. India IEEE Xplore Noise detection on sign Ayari et al [54] Tunisia/USA IEEE Xplore Mathematical component analysis algorithm for separation of cardiac sounds from pulmonary sounds Udawatta et al [55] Sri Lanka IEEE Xplore Digital stethoscope to amplify signal Malek et al [56] Malaysia IEEE Xplore Digital stethoscope in Arduino, ZigBee and signal processing by MatLab Singh and Singh [36] India IEEE Xplore Convolutional Neural Networks Das et al [57] India IEEE Xplore Algorithm to remove signal noise, regardless of sensor quality Gjoreski et al [58] Slovenia/Macedônia IEEE Xplore Machine Learning Pereira et al [59] Portugal/Brasil IEEE Xplore Machine Learning Banerjee et al [60] India IEEE Xplore Convolutional Neural Networks Suhn et al [61] Germany IEEE Xplore Carotid auscultation equipment Gautam and kumar [62] India IEEE Xplore Multilayer Multilayer Perceptron Artificial Neural Network Zhang et al [63] Singapore IEEE Xplore Heart rate estimation algorithm Doshi et al [64] India IEEE Xplore Neural Network Prasad et al [65] Switzerland IEEE Xplore Processing in the time domain employing a low-pass filter Rao et al [66] Switzerland IEEE Xplore Neural Network Hui et al [67] USA IEEE Xplore Investigates transient movement and heartbeat Humayun et al [25] Bangladesh/USA IEEE Xplore Use of convolutional neural network to detect abnormality of cardiac sound with stethoscope Shuvo et al [68] Bangladesh/Saudi Arabia/Yemen IEEE Xplore Convolutional Neural Network for automatic detection of different classes of cardiovascular diseases, direct by phonocardiography signal Tiwari et al [27] India/Saudi Arabia IEEE Xplore Hybrid model, with signal processing using the constant Q transform and Convolutional Neural Network Du et al [69] China JMIR Big Data and Machine Learning Chowdhury et al [26] Qatar/Malaysia PubMed Central Processing and classification using MATLAB Leng et al [30] Singapore PubMed Central Machine Learning Techniques Elgendi et al [70] Canada/India PubMed Central Developed a Wavelet-based algorithm SwarupandMakaryus [71] USA PubMed Central Use of digital stethoscope and mobile computing Raza et al…”

“…Du et al [69] (34) demonstrated the use of big data, statistical methods, and machine learning methods through electronic health records. Because health records have nonlinear characteristics, the authors created a CVD development risk score and tested machine learning algorithms such as Extreme Gradient Boosting (XGBoost), Logistic Regression, Decision Tree, k-Nearest Neighbors (KNN), Random Forest, Missing Data, and Support Vector Machine (SVM).…”

The Application of Projection Word Embeddings on Medical Records Scoring System

“…Because the medical record is the historical record of the patient’s health care; it is also the basis of care, and its content records the patient’s condition during the care process, the reason and result of the inspection, and the treatment method and result. In recent studies, it is feasible to use electronic health records (EHR) to predict disease risk, such as atrial fibrillation (AF) [ 49 ], coronary heart disease in patients with hypertension [ 50 ], fall risk [ 51 ], multiple sclerosis disease [ 52 ], and cervical cancer [ 53 ]. Over the past two decades, the investigation of genetic variation underlying disease susceptibility has increased considerably.…”

Machine Learning and IoT Applied to Cardiovascular Diseases Identification through Heart Sounds: A Literature Review

“…Du et al [71] (34) demonstrated the use of big data, statistical methods, and machine learning methods through electronic health records. Because health records have nonlinear characteristics, the authors created a CVD development risk score and tested machine learning algorithms such as Extreme Gradient Boosting (XGBoost), Logistic Regression, Decision Tree, k-Nearest Neighbors (KNN), Random Forest, Missing Data, and Support Vector Machine (SVM).…”