Quantification of Right and Left Ventricular Function in Cardiac MR Imaging: Comparison of Semiautomatic and Manual Segmentation Algorithms

Abstract

The purpose of this study was to evaluate the performance of a semiautomatic segmentation method for the anatomical and functional assessment of both ventricles from cardiac cine magnetic resonance (MR) examinations, reducing user interaction to a “mouse-click”. Fifty-two patients with cardiovascular diseases were examined using a 1.5-T MR imaging unit. Several parameters of both ventricles, such as end-diastolic volume (EDV), end-systolic volume (ESV) and ejection fraction (EF), were quantified by an experienced operator using the conventional method based on manually-defined contours, as the standard of reference; and a novel semiautomatic segmentation method based on edge detection, iterative thresholding and region growing techniques, for evaluation purposes. No statistically significant differences were found between the two measurement values obtained for each parameter (p > 0.05). Correlation to estimate right ventricular function was good (r > 0.8) and turned out to be excellent (r > 0.9) for the left ventricle (LV). Bland-Altman plots revealed acceptable limits of agreement between the two methods (95%). Our study findings indicate that the proposed technique allows a fast and accurate assessment of both ventricles. However, further improvements are needed to equal results achieved for the right ventricle (RV) using the conventional methodology.

Multi-centre validation of an automatic algorithm for fast 4D myocardial segmentation in cine CMR datasets

“…b 𝑝 < 0.05, two-tailed F-test against "Automatic vs Manual" analysis. (39) Semi-automated 20 -5.5 ± 9.7 -3.6 ± 6.5 -7.3 ± 20.6 1.7 ± 4.1 van der Geest et al (27) Automatic 17 -2.9 ± 13.2 -5.1 ± 18.9 --1.2 ± 14.1 0.1 ± 6.7 van Geuns et al (7) Semi-automated 25 -8.15 ± 11.46 -5.95 ± 6.34 -7.19 ± 15.00 1.6 ± 3.5 Hayes et al (40) Automatic 20 -8.29 ± 10.38 -5.05 ± 8.32 --0.76 ± 3.91 Brossaud et al (10) Fully automatic 130 -11.9 ± 10.2 -8.4 ± 6.9 -3.3 ± 8.3 -3.7 ± 4.5 Codella et al (14) Automatic † 151 4.0 ± 6.8 1.4 ± 5.5 2.6 ± 5.3 -0.6 ± 2.3 Souto et al (15) Semi-automated 52 -4.1 ± 19.0 -3.7 ± 13.5 --1.1 ± 7.0 Lu et al (16) Automatic 133 -1.69 ± 12.76 -1.51 ± 11.30 --0.66 ± 14.72 0.02 ± 5.93 Tufvesson et al (17) Automatic * 150 -14.4 ± 9.0 -10.8 ± 8.7 -11.3 ± 14.4 2.5 ± 2.7 Tufvesson et al (18) Automatic 49 -11 ± 11 1 ± 10 -12 ± 8 4 ± 15 -3 ± 4…”

“…In addition, several commercial software packages are currently available for manual, semi-automated or automated analysis of these datasets (5)(6)(7)(8)(9)(10)(11) . Notwithstanding, due to the difficulties in designing a solution able to deal with the segmentation challenges present in CMR images (12) , fast, automatic and accurate assessment of LV boundaries from base to apex is still lacking, often requiring manual correction of imperfect contours (3, supported by the fact that it receives significant amount of attention in contemporary literature (12,(14)(15)(16)(17)(18) .…”

Automatic Segmentation of the Left Ventricle in Cardiac MRI Using Local Binary Fitting Model and Dynamic Programming Techniques

“…However, this work is a long and tedious task, with inter- and intra-observer variability. Therefore, it is attractive to develop algorithms that are accurate and require as little user interaction as possible for clinical applications [3] .…”

A Hybrid Method for Endocardial Contour Extraction of Right Ventricle in 4-Slices from 3D Echocardiography Dataset

“…Moreover, it is one of the important challenging issues for most researchers in the field of diagnostic of the heart abnormality. In most reviewed literatures, a number of RV automatic and semiautomatic segmentation methods have been presented for MRI [ 9 – 13 ]. The important objective in most methods is to develop the segmentation process by distinguishing the image into distinct regions (i.e., ventricular cavity and myocardial wall tissue) [ 14 , 15 ].…”