RESUMO
AIM: This study aimed to assess the effectiveness of using MRI-apparent diffusion coefficient (ADC) map-driven radiomics to differentiate between hepatocellular adenoma (HCA) and hepatocellular carcinoma (HCC) features. MATERIALS AND METHODS: The study involved 55 patients with liver tumors (20 with HCA and 35 with HCC), featuring 106 lesions equally distributed between hepatic carcinoma and hepatic adenoma who underwent texture analysis on ADC map MR images. The analysis identified several imaging features that significantly differed between the HCA and HCC groups. Four classification models were compared for distinguishing HCA from HCC including linear support vector machine (linear-SVM), radial basis function SVM (RBF-SVM), random forest (RF), and k-nearest neighbor (KNN). RESULTS: The k-nearest neighbor (KNN) classifier displayed the top accuracy (0.89) and specificity (0.90). Linear-SVM and KNN classifiers showcased the leading sensitivity (0.88) for both, with the KNN classifier achieving the highest precision (0.9). In comparison, the conventional interpretation had lower sensitivity (70.1%) and specificity (77.9%). CONCLUSION: The study found that utilizing ADC maps for texture analysis in MR images is a viable method to differentiate HCA from HCC, yielding promising results in identified texture features.
RESUMO
This review was undertaken to assess the diagnostic value of the Liver Imaging Reporting and Data System (LI-RADS) in patients with a high risk of hepatocellular carcinoma (HCC). Google Scholar, PubMed, Web of Science, Embase, PROQUEST, and Cochrane Library, as the international databases, were searched with appropriate keywords. Using the binomial distribution formula, the variance of all studies was calculated, and using Stata version 16 (StataCorp LLC, College Station, TX, USA), the obtained data were analyzed. Using a random-effect meta-analysis approach, we determined the pooled sensitivity and specificity. Utilizing the funnel plot and Begg's and Egger's tests, we assessed publication bias. The results exhibited pooled sensitivity and pooled specificity of 0.80% and 0.89%, respectively, with a 95% confidence interval (CI) of 0.76-0.84 and 0.87-0.92, respectively. The 2018 version of LI-RADS showed the greatest sensitivity (0.83%; 95% CI 0.79-0.87; I 2 = 80.6%; P < 0.001 for heterogeneity; T 2 = 0.001). The maximum pooled specificity was detected in LI-RADS version 2014 (American College of Radiology, Reston, VA, USA; 93.0%; 95% CI 89.0-96.0; I 2 = 81.7%; P < 0.001 for heterogeneity; T 2 = 0.001). In this review, the results of estimated sensitivity and specificity were satisfactory. Therefore, this strategy can serve as an appropriate tool for identifying HCC.
