Automatic segmentation of coronary lumen and external elastic membrane in intravascular ultrasound images using 8-layer U-Net

Liang Dong,Xiangfen Song,Xiaochang Leng,Ya Zhao,Jianping Xiang,Changling Li,Hang Zhao,Wei Lu,Wenbing Jiang,Jun Jiang,Jian’An Wang

doi:10.1186/s12938-021-00852-0

Abstract

BackgroundIntravascular ultrasound (IVUS) is the golden standard in accessing the coronary lesions, stenosis, and atherosclerosis plaques. In this paper, a fully automatic approach by an 8-layer U-Net is developed to segment the coronary artery lumen and the area bounded by external elastic membrane (EEM), i.e., cross-sectional area (EEM-CSA). The database comprises single-vendor and single-frequency IVUS data. Particularly, the proposed data augmentation of MeshGrid combined with flip and rotation operations is implemented, improving the model performance without pre- or post-processing of the raw IVUS images.ResultsThe mean intersection of union (MIoU) of 0.937 and 0.804 for the lumen and EEM-CSA, respectively, were achieved, which exceeded the manual labeling accuracy of the clinician.ConclusionThe accuracy shown by the proposed method is sufficient for subsequent reconstruction of 3D-IVUS images, which is essential for doctors’ diagnosis in the tissue characterization of coronary artery walls and plaque compositions, qualitatively and quantitatively.

Highlights

Intravascular ultrasound (IVUS) is the golden standard in accessing the coronary lesions, stenosis, and atherosclerosis plaques
A typical IVUS pullback contains more than 3000 images, so an accurate, fast, and fully automatic segmentation of lumen and elastic membrane cross-sectional area (EEM-CSA) is highly desirable, but remains a challenging task due to the relative complexity of the IVUS images
We develop a U-Net [11] and evaluate the modified U-Net-based pipeline that automatically segments the lumen and elastic membrane (EEM)-CSA from 2D IVUS images

Summary

Introduction

Intravascular ultrasound (IVUS) is the golden standard in accessing the coronary lesions, stenosis, and atherosclerosis plaques. Conclusion: The accuracy shown by the proposed method is sufficient for subsequent reconstruction of 3D-IVUS images, which is essential for doctors’ diagnosis in the tissue characterization of coronary artery walls and plaque compositions, qualitatively and quantitatively. Dong et al BioMed Eng OnLine (2021) 20:16 atherosclerosis plaque and its vulnerability by measuring lumen diameter, plaque eccentricity, plaque burden, etc., has crucial clinical significance. It is time-consuming and experience-dependent for doctors to manually delineate the lumen and EEM contours on the 2D IVUS images. A typical IVUS pullback contains more than 3000 images, so an accurate, fast, and fully automatic segmentation of lumen and EEM-CSA is highly desirable, but remains a challenging task due to the relative complexity of the IVUS images

Methods

Results

Conclusion