Exploring the Utility of Cardiovascular Magnetic Resonance Radiomic Feature Extraction for Evaluation of Cardiac Sarcoidosis

• Published in: Diagnostics (Basel) Vol. 13; no. 11; p. 1865
• Main Authors: Mushari, Nouf A, Soultanidis, Georgios, Duff, Lisa, Trivieri, Maria G, Fayad, Zahi A, Robson, Philip M, Tsoumpas, Charalampos
• Format: Journal Article
• Language: English
• Published: Switzerland MDPI AG 26.05.2023; MDPI
• Subjects: Accuracy; Cardiovascular diseases; feature selection; Granulomas; imaging; Machine learning; Patients; PET-MRI; Radiomics; Regression analysis; Sarcoidosis; Software; texture analysis; United Kingdom; Germany; United States > US; PET-MRI; machine learning; imaging; feature selection; texture analysis
• ISSN: 2075-4418; 2075-4418
• DOI: 10.3390/diagnostics13111865

The aim of this study is to explore the utility of cardiac magnetic resonance (CMR) imaging of radiomic features to distinguish active and inactive cardiac sarcoidosis (CS). Subjects were classified into active cardiac sarcoidosis (CS ) and inactive cardiac sarcoidosis (CS ) based on PET-CMR imaging. CS was classified as featuring patchy [ F]fluorodeoxyglucose ([ F]FDG) uptake on PET and presence of late gadolinium enhancement (LGE) on CMR, while CS was classified as featuring no [ F]FDG uptake in the presence of LGE on CMR. Among those screened, thirty CS and thirty-one CS patients met these criteria. A total of 94 radiomic features were subsequently extracted using PyRadiomics. The values of individual features were compared between CS and CS using the Mann-Whitney U test. Subsequently, machine learning (ML) approaches were tested. ML was applied to two sub-sets of radiomic features (signatures A and B) that were selected by logistic regression and PCA, respectively. Univariate analysis of individual features showed no significant differences. Of all features, gray level co-occurrence matrix (GLCM) joint entropy had a good area under the curve (AUC) and accuracy with the smallest confidence interval, suggesting it may be a good target for further investigation. Some ML classifiers achieved reasonable discrimination between CS and CS patients. With signature A, support vector machine and k-neighbors showed good performance with AUC (0.77 and 0.73) and accuracy (0.67 and 0.72), respectively. With signature B, decision tree demonstrated AUC and accuracy around 0.7; Conclusion: CMR radiomic analysis in CS provides promising results to distinguish patients with active and inactive disease.