Effect of vessel wall segmentation on volumetric and radiomic parameters of coronary plaques with adverse characteristics

Abstract

Background: Quantitative coronary plaque parameters are increasingly being utilized as surrogate endpoints of pharmaceutical trials. However, little is known whether differences in segmentation significantly alter parameter values. Methods: Overall, 100 coronary plaques with adverse imaging characteristics were segmented automatically, by two experts (R 1-R 2) and three nonexperts (R 3-R 5). Low attenuation noncalcified (LANCP), noncalcified and calcified plaque volume were calculated and 4310 radiomic features were extracted. Intraclass correlation coefficient (ICC) values were calculated between the segmentations. Results: ICC values between expert readers were 0.84 [

Analysis of vascular architecture and parenchymal damage generated by reduced blood perfusion in decellularized porcine kidneys using a gray level co-occurrence matrix

“…In angiography, the GLCM method is much less frequently applied, and so far, only a handful of studies have been published on this topic. These mainly include the use of GLCM features for assessment of low attenuation noncalcified (LANCP), noncalcified and calcified coronary plaques (Kolossvary et al, 2019;Kolossvary et al, 2021), or for computer-aided diagnosis-specific cases of endovascular aortic aneurysms (Garcia et al, 2014). In optical coherence tomography angiography, as demonstrated earlier, some GLCM indicators can also be applied to quantify choriocapillaris in healthy and diseased eyes (Khan et al, 2020).…”

“…The technique uses second-order statistics to determine indicators that reflect image features such as textural homogeneity, uniformity, and level of disorder. Previously, some of these indicators, such as angular second moment and inverse difference moment, have proven to be sensitive in assessing data obtained as the result of various X-ray digital image transformations (Chen et al, 2021;Kolossvary et al, 2021;Shankar et al, 2021).…”

“…In recent years, many innovative techniques, and algorithms have been proposed and tested, often with limited success regarding their potential for integration in current diagnostic and research protocols. Future developments in information technology ensure that many of these techniques will significantly improve diagnostic and prognostic accuracies in X-ray computed tomography, fluoroscopy, and angiography (Cao et al, 2019;Kolossvary et al, 2021;V et al, 2021). Computational methods that use statistical analyses in evaluating image texture are potentially instrumental in X-ray imaging since they may enable fast, objective, and accurate detection of subtle changes in tissue architecture that are occasionally hard to notice during the conventional assessment.…”

“…moment, have proven to be sensitive in assessing data obtained as the result of various X-ray digital image transformations (Chen et al, 2021;Kolossvary et al, 2021;Shankar et al, 2021).…”

Quantitative plaque assessment by coronary computed tomography angiography: An up-to-date review

“…Controversial results can be found in the literature concerning the reproducibility of fully automated plaque quantification methods. The study of Kolossvary et al [40] found that the automated segmentation method compared with expert readers had poor agreement regarding different quantitative plaque features with the exception of calcified plaques. On the contrary, an IVUS validated dedicated automated quantitative software analysis performed with excellent reproducibility in determining quantitative plaque parameters with intraclass correlation coefficient (ICC) of 0.880 for PB [41].…”