Prediction of Myocardial Infarction From Patient Features With Machine Learning

Abstract

This study proposes machine learning-based models to automatically evaluate the severity of myocardial infarction (MI) from physiological, clinical, and paraclinical features. Two types of machine learning models are investigated for the MI assessment: the classification models classify the presence of the infarct and the persistent microvascular obstruction (PMO), and the regression models quantify the Percentage of Infarcted Myocardium (PIM) of patients suspected of having an acute MI during their reception in the emergency department. The ground truth labels for these supervised models are derived from the corresponding Delayed Enhancement MRI (DE-MRI) exams and manual annotations of the myocardium and scar tissues. Experiments were conducted on 150 cases and evaluated with cross-validation. Results showed that for the MI (PMO inclusive) and the PMO (infarct exclusive), the best models obtained respectively a mean error of 0.056 and 0.012 for the quantification, and 88.67 and 77.33% for the classification accuracy of the state of the myocardium. The study of the features' importance also revealed that the troponin value had the strongest correlation to the severity of the MI among the 12 selected features. For the proposal's translational perspective, in cardiac emergencies, qualitative and quantitative analysis can be obtained prior to the achievement of MRI by relying only on conventional tests and patient features, thus, providing an objective reference for further treatment by physicians.

Early Changes in Acute Myocardial Infarction in Pigs: Achieving Early Detection with Wearable Devices

“…In a recent study, a machine learning algorithm was applied to analyze basic physiological, clinical, and paraclinical data from 150 patients. The algorithm obtained an MI prediction accuracy of 88.67% by using only a blood test, ECG readings, and echocardiography findings [ 39 ]. Although it is in early stages in AMI detection, the AI algorithm is booming in the field of arrhythmia diagnosis because of the large data pool.…”

Artificial Intelligence Applications in Cardiovascular Magnetic Resonance Imaging: Are We on the Path to Avoiding the Administration of Contrast Media?

“…Chen et al developed an ML model that combined physiological, clinical, and paraclinical features to evaluate the severity of myocardial infarction in 150 patients. The proposed model revealed with high accuracy the presence of infarction, persistent microvascular dysfunction, and the percentage of infarcted myocardium, demonstrating a mean error of 0.056 and 0.012 for the quantification, and 88.67 and 77.33% for the classification accuracy of the state of the myocardium [ 65 ]. In a retrospective study of 272 patients with diagnoses of myocardial infarction ( n = 108) and healthy controls ( n = 164), an AI-based model was investigated to predict post-contrast information (i.e., presence, location, and/or extent of MI scar) from non-contrast data [ 69 ].…”

Artificial Intelligence and Cardiovascular Magnetic Resonance Imaging in Myocardial Infarction Patients

“…In another study by Chen et al, they proposed ML-based models to automatically evaluate the severity of MI from physiological, clinical, and paraclinical features. (38) Two types of ML models were investigated for MI assessment: the classification models classify the presence of MI and persistent microvascular obstruction, and the regression models quantify the percentage of infarcted myocardium of patients suspected of having an acute MI in the emergency department. (38) The prediction accuracy for the classification of myocardial state and regression quantification of infarcted myocardium were encouraging.…”

“…(38) Two types of ML models were investigated for MI assessment: the classification models classify the presence of MI and persistent microvascular obstruction, and the regression models quantify the percentage of infarcted myocardium of patients suspected of having an acute MI in the emergency department. (38) The prediction accuracy for the classification of myocardial state and regression quantification of infarcted myocardium were encouraging. Rosa et al proposed a new automatic method for MI quantification from LGE sequences.…”