Automatic Detection of Dense Calcium and Acoustic Shadow in Intravascular Ultrasound Images by Dual-threshold-based Segmentation Approach

Abstract

The purpose of this study was to automatically detect dense calcium (DC) and acoustic shadow regions in intravascular ultrasound (IVUS) images by a dual-threshold-based segmentation approach. Three hundred grayscale IVUS and corresponding virtual histology (VH)-IVUS images of human coronary arteries were obtained using a 20 MHz commercial catheter. Plaque regions between intima and media-adventitial borders were manually extracted from all IVUS images. To detect DC and acoustic shadow regions automatically, DC candidates were first selected from plaque regions on the basis of intensity. The shadow mask of each DC candidate was then obtained by calculating its centroid. A DC candidate involving acoustic shadow was finally selected as DC tissue. The segmentation performance of the proposed approach was quantitatively evaluated using the area difference, DC ratio, Hausdorff distance, and Dice similarity coefficient. Quantitative results indicated that all the parameters for the proposed approach were highly similar to those of VH-IVUS. Despite the relatively low agreement (64.1%) for the DC tissue, reliable performance was found for the proposed approach. These experimental results suggest that the proposed method has clinical applicability for diagnosing cardiovascular diseases in IVUS images.

Automated classification of dense calcium tissues in gray-scale intravascular ultrasound images using a deep belief network

“…In our previous study in [32], the DC and corresponding acoustic shadow regions were automatically detected from the gray-scale IVUS images using only the dual-threshold-based segmentation without any specific classification model. Despite the significant similarities, the output regions contained a substantial part of NC tissue (28.4%), since the NC frequently accompanies acoustic shadows.…”

“…The DC candidate and corresponding acoustic shadow regions were automatically detected by the dual-threshold-based segmentation method [32] prior to tissue characterization. The original IVUS image in Cartesian coordinates is relatively difficult to handle due to the circular trait of vessels (Fig.…”

“…Therefore, it is not allowed to analyze all image slices and construct a three-dimensional map for the whole vessel region. On the other hand, the image texture-based approaches are able to analyze all the gray-scale image frames [1, 2, 21–32]. Textural analysis approaches attempt to quantify the perceived texture of lesion mostly based on the spatial characteristics or echogenicity.…”