Radiomics nomogram for prediction of microvascular invasion in hepatocellular carcinoma based on MR imaging with Gd-EOB-DTPA

Abstract

ObjectiveTo develop a radiomics nomogram for predicting microvascular invasion (MVI) before surgery in hepatocellular carcinoma (HCC) patients.Materials and MethodsThe data from a total of 189 HCC patients (training cohort: n = 141; validation cohort: n = 48) were collected, involving the clinical data and imaging characteristics. Radiomics features of all patients were extracted from hepatobiliary phase (HBP) in 15 min. Least absolute shrinkage selection operator (LASSO) regression and logistic regression were utilized to reduce data dimensions, feature selection, and to construct a radiomics signature. Clinicoradiological factors were identified according to the univariate and multivariate analyses, which were incorporated into the final predicted nomogram. A nomogram was developed to predict MVI of HCC by combining radiomics signatures and clinicoradiological factors. Radiomics nomograms were evaluated for their discrimination capability, calibration, and clinical usefulness.ResultsIn the clinicoradiological factors, gender, alpha-fetoprotein (AFP) level, tumor shape and halo sign served as the independent risk factors of MVI, with which the area under the curve (AUC) is 0.802. Radiomics signatures covering 14 features at HBP 15 min can effectively predict MVI in HCC, to construct radiomics signature model, with the AUC of 0.732. In the final nomogram model the clinicoradiological factors and radiomics signatures were integrated, outperforming the clinicoradiological model (AUC 0.884 vs. 0.802; p <0.001) and radiomics signatures model (AUC 0.884 vs. 0.732; p < 0.001) according to Delong test results. A robust calibration and discrimination were demonstrated in the nomogram model. The results of decision curve analysis (DCA) showed more significantly clinical efficiency of the nomogram model in comparison to the clinicoradiological model and the radiomic signature model.ConclusionsDepending on the clinicoradiological factors and radiological features on HBP 15 min images, nomograms can effectively predict MVI status in HCC patients.

Predicting Extrathyroidal Extension in Papillary Thyroid Carcinoma Using a Clinical-Radiomics Nomogram Based on B-Mode and Contrast-Enhanced Ultrasound

“…Nomograms combining clinical risk factors and the radiomics score have been widely used to solve practical clinical problems. Our findings are consistent with many previous studies showing that the combination of clinical risk factors and radiomics improves a model's performance, producing robust results [59][60][61][62]. In this study, a clinical-radiomics nomogram was constructed by integrating age, CEUS-reported ETE, BMUS Radscore, and CEUS Radscore.…”

Preoperative prediction of microvascular invasion classification in hepatocellular carcinoma based on clinical features and MRI parameters

“…In the present study, nomogram models capable of preoperatively predicting MVI-positivity and its M2 classification in patients with HCC were developed. These models incorporated both clinical variables and preoperative MRI features, and demonstrated strong performance in predicting MVI status, with AUC values ranging between 0.700 and 0.920, consistent with previous findings (19,21,22). A notable advantage of the present nomogram model is its reliance on relatively simple and easily obtainable clinical parameters and MRI features.…”

“…Xu et al (25) reported no notable disparity in early recurrence after curative resection between patients with M0 HCC and patients with M1 HCC; however, the M2 grade was potentially associated with a poorer prognosis in patients with HCC. Although numerous studies (6,7,19,21,22,(26)(27)(28) have focused on predicting MVI status in patients with HCC, few have explored the preoperative grading of M2. To the best of our knowledge, no previous attempts have been made to use clinical and MRI features for preoperative prediction of the M2 classification in patients with HCC.…”

“…The evaluated characteristics of the lesions (Fig. S1) included: i) Tumor boundary, clear or unclear; ii) tumor margin, smooth or non-smooth; iii) tumor shape, round, defined as a long diameter/short diameter ratio ≤1.2, or non-round (21); iv) tumor size on MRI, recorded as the maximum diameter of the tumor that was measured in the transverse view on the equilibrium phase (delayed phase) of MRI; v) peritumoral enhancement, defined as presence of enhancement in the peritumoral region during the arterial phase; vi) peritumoral hypointensity, defined as hypointensity in the peritumoral region during the PVP or delayed phase; and vii) ADC value, calculated using regions of interest (30-500 mm 2 ) placed at the level of the maximum diameter of the lesion and inside the visually perceived lowest portions of the tumors on the ADC maps. Notably, areas characterized by hemorrhagic, cystic, necrotic and calcification features were excluded.…”