The Effect of Sample Age and Prediction Resolution on Myocardial Infarction Risk Prediction

Abstract

Myocardial infarction (MI) is one of the leading causes of death in many developed countries. Hence, early detection of MI events is critical for effective preventative therapies, potentially reducing avoidable mortality. One approach for early disease prediction is the use of risk prediction models developed using machine learning techniques. One important component of these models is to provide clinicians with the flexibility to customize (e.g., the prediction range) and use the risk prediction model that they deemed most beneficial for their patients. Therefore, in this paper, we develop MI prediction models and investigate the effect of sample age and prediction resolution on the performance of MI risk prediction models. The cardiovascular health study dataset was used in this study. Results indicate that the prediction model developed using SVM algorithm is capable of achieving high sensitivity, specificity, and balanced accuracy of 95.3%, 84.8%, and 90.1%, respectively, over a time span of 6 years. Both sample age and prediction resolution were found not to have a significant impact on the performance of MI risk prediction models developed using subjects aged 65 and above. This implies that risk prediction models developed using different sample age and prediction resolution is a feasible approach. These models can be integrated into a computer aided screening tool which clinicians can use to interpret and predict the MI risk status of the individual patients after performing the necessary clinical assessments (e.g., cognitive function, physical function, electrocardiography, general changes to health/lifestyle, and medications) required by the models. This could offer a means for clinicians to screen the patients at risk of having MI in the near future and prescribe early medical intervention to reduce the risk.

Prediction of incident myocardial infarction using machine learning applied to harmonized electronic health record data

“…While our study was ultimately negative a variety of useful insights emerged that would help guide future research in integrating ML methods with real-world clinical decision-making. Our study with over 2 million subjects and 52,000 features has the largest sample size and feature space for MI prediction to date [ 4 – 6 , 8 – 10 , 14 , 15 , 19 , 20 , 23 ]. As prior studies that have applied deep learning to MI prediction have focused on predicting recurring events or mortality risk in patients that presented to the ED or cath lab with ACS [ 14 , 15 , 19 , 20 ], to our knowledge, this is the first investigation to attempt to use deep learning to predict ‘first-MI’ events in an otherwise undifferentiated patient population.…”

“…ML methods offer an approach that contrasts with typical statistical tools to extract relationships between variables in a training dataset to predict various outcomes, including mortality. Numerous studies have applied these techniques to predicting cardiac events primarily in patients presenting with acute coronary syndrome (ACS), in some cases showing outperformance of traditional risk methods and statistical techniques [4][5][6][7][8][9][10]. A specific class of these techniques, deep learning, has received significant interest in recent years.…”

A Predictive Performance Analysis of Vitamin D Deficiency Severity Using Machine Learning Methods

“…The prediction model was developed by using multivariate logistic regression [8]. To estimate the best prediction model, statistical test like McNemar's test were conducted [21]. This will help us to validate the empirical results.…”

Risk Prediction Analysis of Cardiovascular Disease Using Supervised Machine Learning Techniques

“…Olga Troyanskaya et al [6] deal with a comparative study of several methods for the estimation of missing values in gene microarray data. They implemented and evaluated three methods, namely, Singular Value Decomposition (SVDimpute), weighted K-Nearest Neighbors (KNNimpute), and row average.…”