Search

Dual-phenotype hepatocellular carcinoma (DPHCC) is highly aggressive and difficult to distinguish from hepatocellular carcinoma (HCC).To develop and validate clinical and radiomics models based on contrast-enhanced MRI for the preoperative diagnosis of DPHCC.Retrospective.A total of 87 patients with DPHCC and 92 patients with non-DPHCC randomly divided into a training cohort (n = 125: 64 non-DPHCC; 61 DPHCC) and a validation cohort (n = 54: 28 non-DPHCC; 26 DPHCC).A 3.0 T; dynamic contrast-enhanced MRI with time-resolved T1-weighted imaging sequence.In the clinical model, the maximum tumor diameter and hepatitis B virus (HBV) were independent risk factors of DPHCC. In the radiomics model, a total of 1781 radiomics features were extracted from tumor volumes of interest (VOIs) in the arterial phase (AP) and portal venous phase (PP) images. For feature reduction and selection, Pearson correlation coefficient (PCC) and recursive feature elimination (RFE) were used. Clinical, AP, PP, and combined radiomics models were established using machine learning algorithms (support vector machine [SVM], logistic regression [LR], and logistic regression-least absolute shrinkage and selection operator [LR-LASSO]) and their discriminatory efficacy assessed and compared.The independent sample t test, Mann-Whitney U test, Chi-square test, regression analysis, receiver operating characteristic curve (ROC) analysis, Pearson correlation analysis, the Delong test. A P value 0.05 was considered statistically significant.In the validation cohort, the combined radiomics model (area under the curve [AUC] = 0.908, 95% confidence interval [CI]: 0.831-0.985) showed the highest diagnostic performance. The AUCs of the PP (AUC = 0.879, 95% CI: 0.779-0.979) and combined radiomics models were significantly higher than that of clinical model (AUC = 0.685, 95% CI: 0.526-0.844). There were no significant differences in AUC between AP or PP radiomics model and combined radiomics model (P = 0.286, 0.180 and 0.543).MRI radiomics models may be useful for discriminating DPHCC from non-DPHCC before surgery.4 TECHNICAL EFFICACY: Stage 2.

Purpose: Early detection of liver cancer improves patient prognosis dramatically. We propose a novel approach in which liver deformation is tracked during the cardiac cycle using tagged MRI. Strain concentrations that arise from material mismatches between the tumor and non-tumor regions were compared in patients with liver tumors. Methods: Tagged MR images were acquired from four liver cancer patients in the coronal plane in a 3.0T scanner. In each cardiac cycle, a total of 25 time frames were acquired with a FOV of 289×359 mm2 and a resolution of 180×224. Lagrangian strain fields were calculated from displacement fields estimated from the tag lines using the HARP algorithm. Tumor and non-tumor regions of interest (ROIs) were delineated manually by a trained physician. The non-tumor ROI was defined as a region surrounding the tumor but away from the left lobe adjacent to the heart. A one-tail paired T-test was used to compare the strain distributions from the tumor and non-tumor ROIs. Results: The strain fields reflected heterogeneity of liver mechanical properties and the nature of the loading. The left lobe had the highest strain magnitude because of its proximity to the heart. However, the maximum principal strain (E1) of the tumor ROI was significantly lower than that of the non-tumor ROI. Conclusion: Cardiac contraction produced measureable strains in the liver, with statistically significant differences between tumor and non-tumor ROIs. Lower strains in the tumor ROIs, observed for all patients, reflected the higher stiffness of the tumor, and showed the potential of using biomechanical properties as a biomarker for tumor diagnosis. Future work will focus on the question of whether tumors that are not visible to a trained physician can be identified using these strain fields. Funding supports are provided by grant BK20140356 (YF) from Jiangsu Province, grant 61503267 (YF) from National Natural Science Foundation, and by grant K511701515 (YF) from Scientific Research Foundation for the Returned Overseas Chinese Scholars, State Education Ministry.