Prediction Of Microvascular Invasion In Hepatocellular Carcinoma With A Multi-Disciplinary Team-like Radiomics Fusion Model On Dynamic Contrast-enhanced Computed Tomography

Objective To investigate microvascular invasion（MVI）of HCC preoperatively through a noninvasive multi-disciplinary team（MDT）-like radiomics fusion model on dynamic contrast enhanced（DCE）computed tomography（CT）.Methods This retrospective study included 111 patients with pathologically proven hepatocellular carcinoma,which comprised 57 MVI-positive and 54 MVI-negative patients.Target volume of interest（VOI）was delineated slice-by-slice on the four phases（early arterial phase[EAP];late arterial phase[LAP],portal venous phase[PVP],and equilibrium phase[EP]）manually,obtaining volume of tumor core(V_tc),and seven peripheral tumor regions(V_pt,with varying distances of 2,4,6,8,10,12,and 14mm to tumor margin)were obtained,which were automatically acquired based on an automatic dilation algorithm.Radiomics features extracted from different combinations of phase（s）and VOI（s）were fed into 150 classification models（built with 10 classifiers and 15 feature selection methods）.The best phase and VOI（or combinations）were determined.The top predictive models were ranked and screened by cross-validation on the training/validation set.The model fusion,a procedure analogous to multi-disciplinary consultation,was performed on the top-3 models to generate a final model,which was validated on an independent testing set.Results Image features extracted from V_tc+V_pt（12mm）in portal venous phase（PVP）showed dominant predictive performances over features from other tumor regions combined with other three phases.Model fusion outperformed any a single model in MVI prediction.The weighted fusion（WF）method achieved the best predictive performance with an AUC of 0.81,accuracy of 78.3%,sensitivity of 81.8%,and specificity of 75%on the independent testing set,based on the top 3 radiomics models.Net Reclassification Improvement（NRI）demonstrated that the ensemble methods,WF and plurality voting（PV）can improve predictive performance compared with each single classification model,with positive benefit,yet with no significant statistical difference.The top-5 ranked features from V_tc+V_pt（12mm）in PVP included one Gray Level Size Zone Matrix（GLSZM）feature and four first-order features demonstrating superior predictive capabilities of MVI status in HCC.Conclusion Image features extracted from the PVP with V_tc+V_pt（12mm）are the most reliable features indicative of MVI based on radiomic analysis of dynamic contrast-enhanced CT.The MDT-like radiomics fusion model is a promising tool to generate accurate and reproducible results in MVI status prediction of HCC noninvasively.