PracticeUpdate Cardiology June 2019

EDITOR’S PICKS 13

Identification of Imaging Markers of Coronary Plaque Vulnerability Using Radiomic Analysis of Coronary CTA European Heart Journal – Cardiovascular Imaging

" …the day may soon be arriving when the humans are just overseeing the computers. "

Take-home message • This study assessed if radiomic analysis outperforms conventional assessment of coronary CTA images in identifying invasive and radionuclide imaging markers of plaque vulnerability. The goal was to determine if a single, widely available noninva- sive technique could identify vulnerable plaques and so be extended to vulnerable patients across broad populations. The study included 25 patients (44 plaques) who underwent coronary CTA, OCT, IVUS, and 18 NaF-PET. The researchers included 935 radiomic parameters and 7 conventional plaque features from CTA images and found that the best radiomic parameters significantly outperformed the best conventional CTA parameters to identify thin-cap fibroatheroma by OCT, attenuated plaque by IVUS, and 18 NaF-positivity. • Although larger studies are needed, these findings suggest that coronary CTA radiomics may be superior to other approaches in accurately identifying plaque vulnerability. T he technique of coronary CT angiography (CCTA) has evolved substantially over the past 20 years since the early reports. Although we initially focused on the ability of CCTA to detect or rule out obstructive plaques, in more recent years, the ability to identify characteristics of atherosclerotic plaques that underlie plaque vulnerability has been a major focus of the literature. Patients whose plaques have features such as positive remodeling, low attenuation, and spotty calcification are at higher risk of subsequent plaque instability and clinical acute coronary syndrome. However, those features are qualitatively assessed and the data emanate from expert labs, raising questions of reproducibility and generalizability. In this paper, the authors used sophisticated quantitative image assessment tools to interrogate the CCTA images, which they refer to as “radiomics,” which, in their words, is “…the process of obtaining quantitative metrics from radiologic images to create big datasets, where each lesion is characterized by hundreds of different parameters.” After data processing, 935 radiomics parameters were calculated. The goal was to assess whether these more quantitative parameters outperformed the usual visual features in identifying high-risk plaques. The latter were defined by three techniques: intravascular ultrasound, optical coherence tomography, and F18-sodium fluoride PET imaging (the latter images inflammation and microcalcification in plaques, suggest- ing vulnerability). The results showed that the best radiomics parameters consistently out-performed the visual assessments of the CCTA images in identifying high-risk plaque by all three methods. This is quite an interesting observation. Before we start discarding our imaging colleagues and turning our images over to purely quantitative assessment, however, this result emanating from a small sample (25 patients and 44 lesions) would need to be externally validated in a much larger patient sample, done elsewhere. It is a reminder that, although we have considered computer-driven quantitative assessment of our noninvasive images as an aid to human eyeball readings, the day may soon be arriving when the humans are just overseeing the computers. If the results of this study are generalizable, that might actually be a good thing for patients! COMMENT By James E. Udelson MD

Abstract AIMS Identification of invasive and radionuclide imaging markers of coronary plaque vulnera- bility by a single, widely available non-invasive technique may provide the opportunity to iden- tify vulnerable plaques and vulnerable patients in broad populations. Our aim was to assess whether radiomic analysis outperforms con- ventional assessment of coronary computed tomography angiography (CTA) images to iden- tify invasive and radionuclide imaging markers of plaque vulnerability. METHODS AND RESULTS We prospectively included patients who underwent coronary CTA, sodium-fluoride positron emission tomography (NaF18-PET), intravascular ultrasound (IVUS), and optical coherence tomography (OCT). We assessed seven conventional plaque features and calculated 935 radiomic parameters from CTA images. In total, 44 plaques of 25 patients were analysed. The best radiomic parameters significantly outperformed the best conven- tional CT parameters to identify attenuated plaque by IVUS [fractal box counting dimen- sion of high attenuation voxels vs. non-calcified plaque volume, area under the curve (AUC): 0.72, confidence interval (CI): 0.65-0.78 vs. 0.59, CI: 0.57-0.62; P<0.001], thin-cap fibroatheroma by OCT (fractal box counting dimension of high attenuation voxels vs. presence of low attenua- tion voxels, AUC: 0.80, CI: 0.72-0.88 vs. 0.66, CI: 0.58-0.73; P<0.001), and NaF18-positivity (sur- face of high attenuation voxels vs. presence of two high-risk features, AUC: 0.87, CI: 0.82-0.91 vs. 0.65, CI: 0.64-0.66; P<0.001). CONCLUSION Coronary CTA radiomics identified invasive and radionuclide imaging markers of plaque vulnerability with good to excellent diagnostic accuracy, significantly outperforming conventional quantitative and qualitative high- risk plaque features. Coronary CTA radiomics may provide a more accurate tool to identify vulnerable plaques compared with conventional methods. Further larger population studies are warranted. Identification of Invasive and Radionuclide Imaging Markers of Coronary Plaque Vulner- ability Using Radiomic Analysis of Coronary Computed Tomography Angiography. Eur Heart J Cardiovasc Imaging 2019 Mar 05;[EPub Ahead of Print], M Kolossváry, J Park, JI Bang, et al. www.practiceupdate.com/c/80848

VOL. 4 • NO. 2 • 2019