Fully Automatic Myocardial Segmentation of Contrast Echocardiography Sequence Using Random Forests Guided by Shape Model

Abstract

Myocardial contrast echocardiography (MCE) is an imaging technique that assesses left ventricle function and myocardial perfusion for the detection of coronary artery diseases. Automatic MCE perfusion quantification is challenging and requires accurate segmentation of the myocardium from noisy and time-varying images. Random forests (RF) have been successfully applied to many medical image segmentation tasks. However, the pixel-wise RF classifier ignores contextual relationships between label outputs of individual pixels. RF which only utilizes local appearance features is also susceptible to data suffering from large intensity variations. In this paper, we demonstrate how to overcome the above limitations of classic RF by presenting a fully automatic segmentation pipeline for myocardial segmentation in full-cycle 2-D MCE data. Specifically, a statistical shape model is used to provide shape prior information that guide the RF segmentation in two ways. First, a novel shape model (SM) feature is incorporated into the RF framework to generate a more accurate RF probability map. Second, the shape model is fitted to the RF probability map to refine and constrain the final segmentation to plausible myocardial shapes. We further improve the performance by introducing a bounding box detection algorithm as a preprocessing step in the segmentation pipeline. Our approach on 2-D image is further extended to 2-D+t sequences which ensures temporal consistency in the final sequence segmentations. When evaluated on clinical MCE data sets, our proposed method achieves notable improvement in segmentation accuracy and outperforms other state-of-the-art methods, including the classic RF and its variants, active shape model and image registration.

Clinical quantitative cardiac imaging for the assessment of myocardial ischaemia

“…Although various physi cal and human factors affecting quantification are known and can be accounted for 135 , automated myocardial segmentation remains challenging owing to speckle noise and 3D deformations. Artificial intelligence solutions using machine learning for automated myocardial segmentation on contrast echocardiography are being developed 136 . These techniques promise automated and fast segmentation and perfusion quantification.…”

High-Frame-Rate Contrast Echocardiography Using Diverging Waves: Initial In Vitro and In Vivo Evaluation

“…8). A recent random forest segmentation method is shown to be able to segment the HFR images and produce an oscillating pattern in the TIC curves corresponding to the heart rate of the sheep [46]. The lognormal fit is able to fit the TIC even if in the presence of this motion [47].…”

“…For in vivo data, the ROI and the tissue correspond to the cardiac chamber containing UCA and the myocardium divided into different segments, respectively. The myocardium segmentation was obtained using a fully automatic myocardial segmentation based on random forests guided by shape model [46]. Then, the myocardium segmentation is divided into six segments in order to evaluate the contrast at several depths.…”

Differentiation of patients with two-dimensional echocardiography false positive non-compaction of ventricular myocardium by contrast echocardiography

“…In order to reduce measurement error, the blind method was used to measure the ultrasonic images. According to the guidelines outlined by the American Society of Echocardiography, 17 myocardial segments utilizing the wall images of conventional ultrasound and LVO echocardiography were (segmentation method) evaluated (Li et al, 2018). According to the 17-segment method of left ventricle, the number of noncompact segments of 2D echocar-diography was determined independently by an attending physi-cian and a deputy chief physician in the Department of Echocar-diography.…”