Assessment of deep learning segmentation for real-time free-breathing cardiac magnetic resonance imaging at rest and under exercise stress

Abstract

In recent years, a variety of deep learning networks for cardiac MRI (CMR) segmentation have been developed and analyzed. However, nearly all of them are focused on cine CMR under breathold. In this work, accuracy of deep learning methods is assessed for volumetric analysis (via segmentation) of the left ventricle in real-time free-breathing CMR at rest and under exercise stress. Data from healthy volunteers (n = 15) for cine and real-time free-breathing CMR at rest and under exercise stress were analyzed retrospectively. Exercise stress was performed using an ergometer in the supine position. Segmentations of two deep learning methods, a commercially available technique (comDL) and an openly available network (nnU-Net), were compared to a reference model created via the manual correction of segmentations obtained with comDL. Segmentations of left ventricular endocardium (LV), left ventricular myocardium (MYO), and right ventricle (RV) are compared for both end-systolic and end-diastolic phases and analyzed with Dice's coefficient. The volumetric analysis includes the cardiac function parameters LV end-diastolic volume (EDV), LV end-systolic volume (ESV), and LV ejection fraction (EF), evaluated with respect to both absolute and relative differences. For cine CMR, nnU-Net and comDL achieve Dice's coefficients above 0.95 for LV and 0.9 for MYO, and RV. For real-time CMR, the accuracy of nnU-Net exceeds that of comDL overall. For real-time CMR at rest, nnU-Net achieves Dice's coefficients of 0.94 for LV, 0.89 for MYO, and 0.90 for RV and the mean absolute differences between nnU-Net and the reference are 2.9 mL for EDV, 3.5 mL for ESV, and 2.6% for EF. For real-time CMR under exercise stress, nnU-Net achieves Dice's coefficients of 0.92 for LV, 0.85 for MYO, and 0.83 for RV and the mean absolute differences between nnU-Net and reference are 11.4 mL for EDV, 2.9 mL for ESV, and 3.6% for EF. Deep learning methods designed or trained for cine CMR segmentation can perform well on real-time CMR. For real-time free-breathing CMR at rest, the performance of deep learning methods is comparable to inter-observer variability in cine CMR and is usable for fully automatic segmentation. For real-time CMR under exercise stress, the performance of nnU-Net could promise a higher degree of automation in the future.

Figure 1

Comparison of different measurement types. Mid-ventricular slices in ED phase of the same volunteer for cine and real-time free-breathing at different heart rates (top) and corresponding manually corrected segmentation (bottom) in a short axis view are shown. Image quality decreases and reconstruction artifacts increase with an increasing heart rate. The left ventricular endocard (red), the left ventricular myocardium (green), and the right ventricle (blue) are segmented.

Figure 2

Representative segmentations of manually corrected contours and deep learning methods. Mid-ventricular slices in ES phase of a volunteer for cine and real-time free-breathing at different heart rates (first row) with corresponding manually corrected (second row), comDL (third row), and nnU-Net segmentation (fourth row). Accuracy of segmentation is measured with Dice’s coefficient (DC). DC for left ventricular endocard (LV), left ventricular myocardium (MYO), and right ventricle (RV) are given for each segmentation.

Figure 3

Dice’s coefficient of real-time CMR measurements plotted against heart rate. DC values of LV, MYO, and RV are calculated for (a) nnU-Net and (b) comDL segmentation in respect to manually corrected contours. Each data point presents the average DC of a segmentation class for a single real-time measurement of a volunteer. Real-time CMR at rest, under exercise stress and maximal exercise stress are presented by their average calculated heart rate.

Figure 4

Example segmentation failures of nnU-Net for real-time CMR under exercise stress. Incomplete segmentation of the right ventricle in the apical region (first and second column). Anatomically incoherent segmentation of the myocardium and right ventricle in the basal region (third and fourth column).