Answers are grounded in this paper and cite their sources.
Purpose To evaluate quantitative non-contrast dual-layer detector spectral CT (DLCT) for differentiating hepatic hemangioma (HH) from hepatocellular carcinoma (HCC) in chronic liver disease (CLD). Methods This retrospective study included 90 patients with CLD and 92 focal liver lesions (35 HHs, 57 HCCs) who underwent non-contrast and multiphase contrast-enhanced DLCT between September 2023 and April 2026. Diagnoses were established by pathology or typical CT/MRI findings with clinical follow-up. Clinical variables and DLCT parameters, including virtual monoenergetic CT attenuation, spectral attenuation curve slopes, effective atomic number (Zeff), and electron density (ED), were assessed. Clinical, DLCT, and combined logistic regression models were evaluated using repeated nested cross-validation. Results Compared with HHs, HCCs were associated with older age, more frequent cirrhosis and alpha-fetoprotein positivity, larger maximum lesion diameter, and more frequent abnormal albumin and total bilirubin levels (all p < 0.05). HU40, λHU40–100, and Zeff were higher in HCCs, whereas HU100, HU140, and ED were lower (all p < 0.05). Alpha-fetoprotein, maximum lesion diameter, Zeff, and ED were independent predictors. In outer cross-validation, the combined model showed the highest performance (AUC, 0.814; 95% CI, 0.806–0.823), with sensitivity, specificity, and accuracy of 0.723, 0.761, and 0.738. It outperformed the clinical and DLCT models (both p < 0.001) and improved reclassification over both. Conclusion Non-contrast DLCT parameters, particularly Zeff and ED, help differentiate HH from HCC in CLD. Their combination with clinical variables improved discrimination and risk reclassification.
No takes yet. Share an insight, caveat, or question.
Yan et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: