Radiomics Analysis of Clinical Myocardial Perfusion Stress SPECT Images to Identify Coronary Artery Calcification

Abstract

AbstractPurposeMyocardial perfusion stress SPECT (MPSS) is an established diagnostic test for patients suspected with coronary artery disease (CAD). Meanwhile, coronary artery calcification (CAC) scoring obtained from diagnostic CT is a highly specific test, offering incremental diagnostic information in identifying patients with significant CAD yet normal MPSS scans. However, after decades of wide utilization of MPSS, CAC is not commonly reimbursed (e.g. by the CMS), nor widely deployed in community settings. We aimed to perform radiomics analysis of normal MPSS scans to investigate the potential to predict the CAC score.MethodsWe collected data from 428 patients with normal (non-ischemic) MPSS (99mTc-Sestamibi; consensus reading). A nuclear medicine physician verified iteratively reconstructed images (attenuation-corrected) to be free from fixed perfusion defects and artifactual attenuation. 3D images were automatically segmented into 4 regions of interest (ROIs), including myocardium and 3 vascular segments (LAD-LCX-RCA). We used our software package, standardized environment for radiomics analysis (SERA), to extract 487 radiomic features in compliance with the image biomarker standardization initiative (IBSI). Isotropic cubic voxels were discretized using fixed bin-number discretization (8 schemes). We first performed blind-to-outcome feature selection focusing on a priori usefulness, dynamic range, and redundancy of features. Subsequently, we performed univariate and multivariate machine learning analyses to predict CAC scores from i) selected radiomic features, ii) 10 clinical features, iii) combined radiomics + clinical features. Univariate analysis invoked Spearman correlation with Benjamini-Hotchberg false-discovery correction. The multivariate analysis incorporated stepwise linear regression, where we randomly selected a 15% test set and divided the other 85% of data into 70% training and 30% validation sets. Training started from a constant (intercept) model, iteratively adding/removing features (stepwise regression), invoking Akaike information criterion (AIC) to discourage overfitting. Validation was run similarly, except that the training output model was used as the initial model. We randomized training/validation sets 20 times, selecting the best model using log-likelihood for evaluation in the test set. Assessment in the test set was performed thoroughly by running the entire operation 50 times, subsequently employing Fisher’s method to verify the significance of independent tests.ResultsUnsupervised feature selection significantly reduced 8×487 features to 56. In univariate analysis, no feature survived FDR to directly correlate with CAC scores. Applying Fisher’s method to the multivariate regression results demonstrated combining radiomics with the clinical features to enhance the significance of the prediction model across all cardiac segments. The median absolute Pearson’s coefficient values / p-values for the three feature-pools (radiomics, clinical, combined) were: (0.15, 0.38, 0.41)/(0.1, 0.001, 0.0006) for myocardium, (0.24, 0.35, 0.41)/(0.05, 0.004, 0.0007) for LAD, (0.07, 0.24, 0.28)/(0.4, 0.06, 0.02), for LCX, and (0.06, 0.16, 0.24)/(0.4, 0.2, 0.05) for RCA, demonstrating consistently enhanced correlation and significance for combined radiomics and clinical features across all cardiac segments.ConclusionsOur standardized and statistically robust multivariate analysis demonstrated significant prediction of the CAC score for all cardiac segments when combining MPSS radiomic features with clinical features, suggesting radiomics analysis can add diagnostic or prognostic value to standard MPSS for wide clinical usage.