Fully automated segmentation of left ventricular scar from 3D late gadolinium enhancement magnetic resonance imaging using a cascaded multi‐planar U‐Net (CMPU‐Net). Issue 4 (10th February 2020)

• Record Type: Journal Article
• Authors: Zabihollahy, Fatemeh; Rajchl, Martin; White, James A.; Ukwatta, Eranga
• Is Part Of: Medical physics. Volume 47:Issue 4(2020)
• Page Start: 1645
• Page End: 1655
• Publication Date: 2020-02-10
• Subjects: convolutional neural network -- late gadolinium enhancement magnetic resonance imaging -- left ventricle myocardium -- left ventricular scar -- U‐Net; Medical physics -- Periodicals; Medical physics; Geneeskunde; Natuurkunde; Toepassingen; Biophysics; Periodicals; Periodicals; Electronic journals; 610.153
• DOI: 10.1002/mp.14022
• Languages: English
• ISSNs: 0094-2405

Abstract

Abstract : Purpose: Three‐dimensional (3D) late gadolinium enhancement magnetic resonance (LGE‐MR) imaging enables the quantification of myocardial scar at high resolution with unprecedented volumetric visualization. Automated segmentation of myocardial scar is critical for the potential clinical translation of this technique given the number of tomographic images acquired. Methods: In this paper, we describe the development of cascaded multi‐planar U‐Net (CMPU‐Net) to efficiently segment the boundary of the left ventricle (LV) myocardium and scar from 3D LGE‐MR images. In this approach, two subnets, each containing three U‐Nets, were cascaded to first segment the LV myocardium and then segment the scar within the presegmented LV myocardium. The U‐Nets were trained separately using two‐dimensional (2D) slices extracted from axial, sagittal, and coronal slices of 3D LGE‐MR images. We used 3D LGE‐MR images from 34 subjects with chronic ischemic cardiomyopathy. The U‐Nets were trained using 8430 slices, extracted in three orthogonal directions from 18 images. In the testing phase, the outputs of U‐Nets of each subnet were combined using the majority voting system for final label prediction of each voxel in the image. The developed method was tested for accuracy by comparing its results to manual segmentations of LV myocardium and LV scar from 7250 slices extracted from 16 3D LGE‐MR images. Our method was also compared to numerous alternative methods based on machine learning, energy minimization, and intensity‐thresholds. Results: Our algorithm reported a mean dice similarity coefficient (DSC), absolute volume difference (AVD), and Hausdorff distance (HD) of 85.14% ± 3.36%, 43.72 ± 27.18 cm 3, and 19.21 ± 4.74 mm for determining the boundaries of LV myocardium from LGE‐MR images. Our method also yielded a mean DSC, AVD, and HD of 88.61% ± 2.54%, 9.33 ± 7.24 cm 3, and 17.04 ± 9.93 mm for LV scar segmentation on the unobserved test dataset. Our method significantly outperformed the alternative techniques in segmentation accuracy ( P < 0.05). Conclusions: The CMPU‐Net method provided fully automated segmentation of LV scar from 3D LGE‐MR images and outperformed the alternative techniques.