Abstract:ObjectivesTo identify image features of macrotrabecular-massive (MTM) hepatocellular carcinoma (HCC) and to determine its role in predicting MTM-HCC.MethodsPatients who underwent preoperative gadoxetic acid-enhanced MRI and with surgery proven HCC were retrospectively included. Imaging features were assessed according to Liver Imaging Reporting and Data System. Quantitative measurements were recorded. Clinical characteristics and imaging findings were compared between MTM-HCCs and non-MTM-HCCs. Predictive factors of MTM-HCC were screened with univariate analyses and then identified with multivariate logistic regression. A regression-based diagnostic model was constructed. ROC analyses were used to determine cutoff values, AUC, and corresponding 95% confidence interval (CI) of findings. The diagnostic performance was validated by 10-fold cross-validation.ResultsOne hundred and forty-one patients with 37 MTM-HCCs were included. Multivariate analyses identified high platelet count (≥ 163.5 × 103/ul, odds ratio = 3.20; 95% CI: 1.29, 7.96; p = 0.012), low tumor-to-liver ADC ratio (≤ 1.05, odds ratio = 3.05; 95% CI, 1.23 - 7.55; p = 0.016), and necrosis or severe ischemia (odds ratio = 11.61; 95% CI, 3.99 - 33.76, p < 0.001) as independent predictors of MTM-HCC. Necrosis or severe ischemia alone helped identify 86% MTM-HCCs with a specificity of 66%. The average AUCs were 0.81 (95% CI: 0.71, 0.90) for the regression-based diagnostic model, with a sensitivity of 57% and specificity of 92%.ConclusionsNecrosis or severe ischemia was a sensitive imaging feature of MTM-HCC. Noninvasive prediction of this subtype can be achieved with good accuracy and excellent specificity when findings were combined.Key Points• The macrotrabecular-massive (MTM) hepatocellular carcinoma (HCC) represents an aggressive subtype of HCC and is associated with poor prognosis.• Imaging features of necrosis or severe ischemia alone helped identify 86% MTM-HCCs with a specificity of 66%.• A regression-based diagnostic model including high platelet count (≥ 163.5 × 103/ul), low tumor-to-liver ADC ratio (≤ 1.05), and necrosis or severe ischemia can provide noninvasive assessment of MTM-HCC with good accuracy and high specificity.

A multitask deep learning radiomics model for predicting the macrotrabecular-massive subtype and prognosis of hepatocellular carcinoma after hepatic arterial infusion chemotherapy

Deep Learning-Based Radiomics Models for Early Recurrence Prediction of Hepatocellular Carcinoma with Multi-phase CT Images and Clinical Data.

Radiomics Model of Dynamic Contrast-Enhanced MRI for Evaluating Vessels Encapsulating Tumor Clusters and Microvascular Invasion in Hepatocellular Carcinoma

A Transformer-Based Model for Preoperative Early Recurrence Prediction of Hepatocellular Carcinoma with Muti-modality MRI

Prediction of Macrotrabecular-Massive Hepatocellular Carcinoma by Using MR-based Models and Their Prognostic Implications

Dual-Energy CT Deep Learning Radiomics to Predict Macrotrabecular-Massive Hepatocellular Carcinoma

Macrotrabecular-massive hepatocellular carcinoma: imaging identification and prediction based on gadoxetic acid–enhanced magnetic resonance imaging

A radiomics nomogram based on contrast-enhanced MRI for preoperative prediction of macrotrabecular-massive hepatocellular carcinoma

Multitask deep learning for prediction of microvascular invasion and recurrence-free survival in hepatocellular carcinoma based on MRI images

Identification of macrotrabecular‐massive hepatocellular carcinoma through multiphasic CT‐based representation learning method

Deep Learning Radiomics Model of Dynamic Contrast-Enhanced MRI for Evaluating Vessels Encapsulating Tumor Clusters and Prognosis in Hepatocellular Carcinoma

Multi-task deep learning network to predict future macrovascular invasion in hepatocellular carcinoma

A Combined Deep Learning and Radiomics Model Predicts Ki-67 Expression in HCC Using MRI and Multifrequency MR Elastography

Multiparametric MRI‐based model for prediction of local progression of hepatocellular carcinoma after thermal ablation

Development of a magnetic resonance imaging-derived radiomics model to predict microvascular invasion in patients with hepatocellular carcinoma

Preoperative prediction model for macrotrabecular-massive hepatocellular carcinoma based on contrast-enhanced CT and clinical characteristics: a retrospective study

A deep learning model based on MRI for prediction of vessels encapsulating tumour clusters and prognosis in hepatocellular carcinoma

Application of a Convolutional Neural Network for Multitask Learning to Simultaneously Predict Microvascular Invasion and Vessels that Encapsulate Tumor Clusters in Hepatocellular Carcinoma

Radiomic analysis based on multi-phase magnetic resonance imaging to predict preoperatively microvascular invasion in hepatocellular carcinoma.

Magnetic Resonance Deep Learning Radiomic Model Based on Distinct Metastatic Vascular Patterns for Evaluating Recurrence‐Free Survival in Hepatocellular Carcinoma

Prediction of TACE Treatment Response in a Preoperative MRI via Analysis of Integrating Deep Learning and Radiomics Features