Modified GAN Augmentation Algorithms for the MRI-Classification of Myocardial Scar Tissue in Ischemic Cardiomyopathy.

Umesh C Sharma,Swati D Sonkawade,Kanhao Zhao,Leslie Ying,Jennifer K Lang,Kyle Mentkowski,Badri Karthikeyan

doi:10.3389/fcvm.2021.726943

Abstract

Contrast-enhanced cardiac magnetic resonance imaging (MRI) is routinely used to determine myocardial scar burden and make therapeutic decisions for coronary revascularization. Currently, there are no optimized deep-learning algorithms for the automated classification of scarred vs. normal myocardium. We report a modified Generative Adversarial Network (GAN) augmentation method to improve the binary classification of myocardial scar using both pre-clinical and clinical approaches. For the initial training of the MobileNetV2 platform, we used the images generated from a high-field (9.4T) cardiac MRI of a mouse model of acute myocardial infarction (MI). Once the system showed 100% accuracy for the classification of acute MI in mice, we tested the translational significance of this approach in 91 patients with an ischemic myocardial scar, and 31 control subjects without evidence of myocardial scarring. To obtain a comparable augmentation dataset, we rotated scar images 8-times and control images 72-times, generating a total of 6,684 scar images and 7,451 control images. In humans, the use of Progressive Growing GAN (PGGAN)-based augmentation showed 93% classification accuracy, which is far superior to conventional automated modules. The use of other attention modules in our CNN further improved the classification accuracy by up to 5%. These data are of high translational significance and warrant larger multicenter studies in the future to validate the clinical implications.

Highlights

Acute myocardial infarction (MI), commonly known as a heart attack, is an unpredictable complication of coronary artery disease (CAD)
The severity of myocardial injury following left anterior descending (LAD) ligation was calculated as the ratio of the area at risk (AAR) to left ventricular area (LV) (AAR/LV: 32.18% ± 13.32, n = 6), assessed across the heart at 1 mm intervals from base to apex
We report that Generative Adversarial Network (GAN) is an effective method to mitigate the data imbalance and present a comparative data analysis algorithm to show which GAN-type is must suitable to augment myocardial scar imaging

Summary

Introduction

Acute myocardial infarction (MI), commonly known as a heart attack, is an unpredictable complication of coronary artery disease (CAD). The location, size, density and heterogeneity of myocardial scarring provides both diagnostic and prognostic information for patient management. Such information is critical to manage patients at risk for heart failure (HF) and lethal cardiac arrhythmias [1, 2]. HF and cardiac arrhythmias usually result from diseased myocardium and electrically unstable scars [3,4,5]. Myocardial scar classification using emerging data augmentation methods is of great clinical significance.

Methods

Results

Discussion

Conclusion