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

Abstract

BackgroundIVUS is widely used to quantitatively assess coronary artery disease. The purpose of this study was to automatically characterize dense calcium (DC) tissue in the gray scale intravascular ultrasound (IVUS) images using the image textural features.MethodsA total of 316 Gy-scale IVUS and corresponding virtual histology images from 26 patients with acute coronary syndrome who underwent IVUS along with X-ray angiography between October 2009 to September 2014 were retrospectively acquired and analyzed. One expert performed all procedures and assessed their IVUS scans. After image acquisition, the DC candidate and corresponding acoustic shadow regions were automatically determined. Then, nine image-base feature groups were extracted from the DC candidates. In order to reduce the dimensionalities, principal component analysis (PCA) was performed, and selected feature sets were utilized as an input for a deep belief network. Classification results were validated using 10-fold cross validation.ResultsThe dimensionality of the feature map was efficiently reduced by 50% (from 66 to 33) without any performance decrease using PCA method. Sensitivity, specificity, and accuracy of the proposed method were 92.8 ± 0.1%, 85.1 ± 0.1%, and 88.4 ± 0.1%, respectively (p < 0.05). We found that the window size could largely influence the characterization results, and selected the 5 × 5 size as the best condition. We also validated the performance superiority of the proposed method with traditional classification methods.ConclusionsThese experimental results suggest that the proposed method has significant clinical applicability for IVUS-based cardiovascular diagnosis.

The Prognostic Value of a Validated and Automated Intravascular Ultrasound-Derived Calcium Score

“…The concept of virtual histology (VH)-IVUS was first introduced in 2002 and demonstrated to be able to detect plaque characteristics in non-culprit coronary arteries that predict major adverse cardiovascular events [ 29 – 31 ]. At present, the use of VH-IVUS has practically been abandoned due to repetitive questions on the validity of the algorithm and lack of ability of VH-IVUS to alter patient management.Previous automated classification tools for detection of calcium in 20 MHz IVUS pullbacks and OCT have been proposed as well, but have mainly focused on tissue type segmentation in cross-sectional images [ 13 , 14 , 32 ]. This study is the first to derive a clinically relevant score and to validate this score against clinical outcome.…”

Recent Trends in Artificial Intelligence-Assisted Coronary Atherosclerotic Plaque Characterization

“…In [95], the Gabor transform for different scales and angles along with six measurements of entropy were explored. Likewise, first order statistics (FOS) can be used for texture analysis [96]. To extract coronary lumen and plaque features, an adjacent pattern method was implemented.…”

Ultrasound tissue classification: a review

“…In IVUS images, calcifications are demonstrated as hyperechoic areas, whereas hemorrage or fat deposition inside an atheromatic plaque is hypoechoic. Subsequently, the plaque can be classified as lipid, calcified and fibrous, according to its acoustic properties (Liu and Goldberg 1999) Tissue classification in IVUS images can automatically predict vulnerable plaques as well as quantify the amount of the different tissues; many approaches have been proposed in literature (Sathyanarayana et al 2009;Escalera et al 2009;Ciompi et al 2010;Seabra et al 2011;Ciompi 2012;Lee et al 2019), which will be discussed in detail in the following sections. Ciompi 2012) 1 3…”