Advanced biomedical imaging for accurate discrimination and prognostication of mediastinal masses

Background: To investigate the potential of radiomic features and dual-source dual-energy CT (DECT) parameters in differentiating between benign and malignant mediastinal masses and predicting patient outcomes. Methods: In this retrospective study, we analysed data from 90 patients (38 females, mean age 51 +/- 25 years) with confirmed mediastinal masses who underwent contrast-enhanced DECT. Attenuation, radiomic features and DECT-derived imaging parameters were evaluated by two experienced readers. We performed analysis of variance (ANOVA) and Chi-square statistic tests for data comparison. Receiver operating characteristic curve analysis and Cox regression tests were used to differentiate between mediastinal masses. Results: Of the 90 mediastinal masses, 49 (54%) were benign, including cases of thymic hyperplasia/thymic rebound (n = 10), mediastinitis (n = 16) and thymoma (n = 23). The remaining 41 (46%) lesions were classified as malignant, consisting of lymphoma (n = 28), mediastinal tumour (n = 4) and thymic carcinoma (n = 9). Significant differences were observed between benign and malignant mediastinal masses in all DECT-derived parameters (p =.001) and 38 radiomic features (p =.044) obtained from contrast-enhanced DECT. The combination of these methods achieved an area under the curve of.98 (95% CI,.893-1.000; p <.001) to differentiate between benign and malignant masses, with 100% sensitivity and 91% specificity. Throughout a follow-up of 1800 days, a multiparametric model incorporating radiomic features, DECT parameters and gender showed promising prognostic power in predicting all-cause mortality (c-index =.8 [95% CI,.702-.890], p <.001). Conclusions: A multiparametric approach combining radiomic features and DECT-derived imaging biomarkers allows for accurate and noninvasive differentiation between benign and malignant masses in the anterior mediastinum.