Mono+ algorithm assessment of the diagnostic value of dual-energy CT for high-risk factors for colorectal cancer: a preliminary study
Jun-Fan Chen,Jing Yang,Wei-Juan Chen,Xin Wei,Xiang-Ling Yu,Dou-Dou Huang,Hao Deng,Yin-Deng Luo,Xin-Jie Liu
DOI: https://doi.org/10.21037/qims-23-291
2024-01-03
Abstract:Background: Risk factors for colorectal cancer (CRC) affect the way patients are subsequently treated and their prognosis. Dual-energy computerized tomography (DECT) is an advanced imaging technique that enables the quantitative evaluation of lesions. This study aimed to evaluate the quality of DECT images based on the Mono+ algorithm in CRC, and based on this, to assess the value of DECT in the diagnosis of CRC risk factors. Methods: This prospective study was performed from 2021 to 2023. A dual-phase DECT protocol was established for consecutive patients with primary CRC. The signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), overall image quality, lesion delineation, and image noise of the dual-phase DECT images were assessed. Next, the optimal energy-level image was selected to analyze the iodine concentration (IC), normalized iodine concentration (NIC), effective atomic number, electron density, dual-energy index (DEI), and slope of the energy spectrum curve within the tumor for the high- and low-risk CRC groups. A multifactor binary logistic regression analysis was used to construct a differential diagnostic regression model for high- and low-risk CRC, receiver operating characteristic (ROC) curves were plotted, and the area under the curve (AUC) was calculated to assess the diagnostic value of the model. Results: A total of 74 patients were enrolled in this study, of whom 41 had high-risk factors and 33 had low-risk factors. The SNR and CNR were best at 40 keV virtual monoenergetic imaging (VMI) based on the Mono+ algorithm (VMI+) (SNR 8.79±1.27, P<0.001; CNR 14.89±1.77, P=0.027). The overall image quality and lesion contours were best at 60 keV VMI+ and 40 keV VMI+, respectively (P=0.001). Among all the DECT parameters, the arterial phase (AP)-IC, NIC, DEI, energy spectrum curve, and venous phase-NIC differed significantly between the two groups. The AP-IC was the optimal DECT parameter for predicting high- and low-risk CRC with AUC, sensitivity, specificity, and cut-off values of 0.96, 97.06%, 87.80%, and 2.94, respectively, and the 95% confidence interval (CI) of the AUC was 0.88-0.99. Integrating the clinical factors and DECT parameters, the AUC, sensitivity, specificity, and predictive accuracy of the model were 0.99, 100.00%, 92.68%, and 94.67%, respectively, and the 95% CI of the AUC was 0.93-1.00. Conclusions: The DECT parameters based on 40 keV noise-optimized VMI+ reconstruction images depicted the CRC tumors best, and the clinical DECT model may have significant implications for the preoperative prediction of high-risk factors in CRC patients.