Radiomic analysis of contrast-enhanced CT predicts microvascular invasion and outcome in hepatocellular carcinoma

Establishment and Validation of Diagnostic Model of Microvascular Invasion in Solitary Hepatocellular Carcinoma

BACKGROUND

The microvascular invasion (MVI) score evaluates the presence of MVI in patients with hepatocellular carcinoma (HCC) by integrating multiple factors associated with MVI. We aimed to establish a MVI scoring system for HCC based on the clinical characteristics and serum biomarkers of patients with HCC.

METHODS

A total of 1027 patients with HCC hospitalized at Shandong Provincial Hospital from January 2016 to August 2021 were included and randomly divided into the development group and validation group at a ratio of 3:1. Univariable and multivariable logistic regression analyses were conducted to identify independent risk factors for MVI in HCC patients. Based on these independent risk factors, the preoperative MVI scoring system (diagnostic model) for HCC was established and verified. The receiver operating characteristic (ROC) curves, calibration curves and decision curve analyses (DCA) were employed to evaluate the discrimination and clinical application of the diagnostic model.

RESULTS

Independent risk factors for MVI of HCC involved Hepatitis B virus infection (HBV), large tumor diameter, higher logarithm of Alpha-fetoprotein (Log AFP), higher logarithm of AFP-L3% (Log AFP-L3%), higher logarithm of protein induced by vitamin K absence or antagonist-II (Log PIVKA-II) and higher logarithm of Carbohydrate antigen 125 (Log CA125). The diagnostic model incorporating these six independent risk factors was finally established. The areas under the ROC curve (AUC) assessed by the nomogram in the development cohort and validation cohort were 0.806 (95% CI, 0.773-0.839) and 0.818 (95% CI, 0.763-0.874) respectively. The calibration curve revealed that the results predicted by our diagnostic model for MVI in HCC were highly consistent with the postoperative pathological outcomes. The DCA further indicated promising clinical application of the diagnostic model.

CONCLUSION

An effective preoperative diagnostic model for MVI of HCC based on readily available tumor markers and clinical characteristics has been established, which is both clinically significant and easy to implement for diagnosing MVI.

Modality-Aware Distillation Network for Microvascular Invasion Prediction of Hepatocellar Carcinoma From MRI Images

Microvascular invasion (MVI) of hepatocellular carcinoma (HCC) is a crucial histopathologic prognostic factor associated with cancer recurrence after liver transplantation or hepatectomy. Recently, clinicoradiologic characteristics are combined with medical images to enhance the HCC prediction. However, compared to medical imaging data, the clinicoradiologic characteristics (e.g., APOe4 genotyping) is not easy to collect or even unavailable, as it requires more efforts of clinicians and more medical instruments for collecting diverse measurements. This work explores how to transfer the knowledge of a teacher network learned from non-image clinical data and image data to a student network with only image data such that the student network can leverage the transferred clinical information to boost HCC classification with only imaging data as input. Specifically, we present a modality-aware distillation network (MD-Net) to transform non-image clinicoradiologic from the teacher network to the student network. The teacher network integrates non-image clinicoradiologic characteristics with two 3D MRI modality images via two MRI-clinical-fusion modules and a symmetric attention (SA) module, while the student network extracts features from two modality MRI data via two MRI-only modules and then refine these two MRI features via a SA module. A classification-level distillation and a feature-level distillation are jointly utilized to transfer the clinical information between teacher and student networks. Furthermore, we design a novel self-supervised task to predict clinicoradiologic characteristics from the imaging data to further enhance the downstream HCC classification. The experimental results from our collected dataset and a multi-modal sarcasm detection dataset have demonstrated the effectiveness of our approach. Specifically, we achieved an AUC score of 71.86% and 75.51% respectively, surpassing the performance of the state-of-the-art classification methods.

Prediction of lymph node metastasis in papillary thyroid carcinoma using non-contrast CT-based radiomics and deep learning with thyroid lobe segmentation: A dual-center study

Objectives

This study aimed to develop a predictive model for lymph node metastasis (LNM) in papillary thyroid carcinoma (PTC) patients by deep learning radiomic (DLRad) and clinical features.

Methods

This study included 271 thyroid lobes from 228 PTC patients who underwent preoperative neck non-contrast CT at Center 1 (May 2021-April 2024). LNM status was confirmed via postoperative pathology, with each thyroid lobe labeled accordingly. The cohort was divided into training (n = 189) and validation (n = 82) cohorts, with additional temporal (n = 59 lobes, Center 1, May-August 2024) and external (n = 66 lobes, Center 2) test cohorts. Thyroid lobes were manually segmented from the isthmus midline, ensuring interobserver consistency (ICC ≥ 0.8). Deep learning and radiomics features were selected using LASSO algorithms to compute DLRad scores. Logistic regression identified independent predictors, forming DLRad, clinical, and combined models. Model performance was evaluated using AUC, calibration, decision curves, and the DeLong test, compared against radiologists' assessments.

Results

Independent predictors of LNM included age, gender, multiple nodules, tumor size group, and DLRad. The combined model demonstrated superior diagnostic performance with AUCs of 0.830 (training), 0.799 (validation), 0.819 (temporal test), and 0.756 (external test), outperforming the DLRad model (AUCs: 0.786, 0.730, 0.753, 0.642), clinical model (AUCs: 0.723, 0.745, 0.671, 0.660), and radiologist evaluations (AUCs: 0.529, 0.606, 0.620, 0.503). It also achieved the lowest Brier scores (0.167, 0.184, 0.175, 0.201) and the highest net benefit in decision-curve analysis at threshold probabilities > 20 %.

Conclusions

The combined model integrating DLRad and clinical features exhibits good performance in predicting LNM in PTC patients.

Prediction of Ki-67 expression in hepatocellular carcinoma with machine learning models based on intratumoral and peritumoral radiomic features

BACKGROUND

Hepatocellular carcinoma (HCC) is one of the most common malignant tumours of the digestive system worldwide. The expression of Ki-67 is crucial for the diagnosis, treatment, and prognostic evaluation of HCC.

AIM

To construct a machine learning model for the preoperative evaluation of Ki-67 expression in HCC and to assist in clinical decision-making.

METHODS

This study included 164 pathologically confirmed HCC patients. Radiomic features were extracted from the computed tomography images reconstructed by superresolution of the intratumoral and peritumoral regions. Features were selected via the intraclass correlation coefficient, t tests, Pearson correlation coefficients and least absolute shrinkage and selection operator regression methods, and models were constructed via various machine learning methods. The best model was selected, and the radiomics score (Radscore) was calculated. A nomogram incorporating the Radscore and clinical risk factors was constructed. The predictive performance of each model was evaluated via receiver operating characteristic (ROC) curves and calibration curves, and decision curve analysis was used to assess the clinical benefits.

RESULTS

In total, 164 HCC patients, namely, 104 patients with high Ki-67 expression and 60 with low Ki-67 expression, were included. Compared with the models in which only intratumoral or peritumoral features were used, the fusion model in which intratumoral and peritumoral features were combined demonstrated stronger predictive ability. Moreover, the clinical-radiomics model including the Radscore and clinical features had higher predictive performance than did the fusion model (area under the ROC curve = 0.848 vs 0.780 in the training group, area under the ROC curve = 0.830 vs 0.760 in the validation group). The calibration curve showed good consistency between the predicted probability and the actual probability, and the decision curve further confirmed its clinical benefit.

CONCLUSION

A machine learning model based on the radiomic features of the intratumoral and peritumoral regions on superresolution computed tomography in conjunction with clinical factors can accurately evaluate Ki-67 expression. The model provides valuable assistance in selecting treatment strategies for HCC patients and contributes to research on neoadjuvant therapy for liver cancer.

Predicting early recurrence in hepatocellular carcinoma after hepatectomy using GD-EOB-DTPA enhanced MRI-based model

PURPOSE

To develop and validate a comprehensive model for predicting postoperative early recurrence of hepatocellular carcinoma (HCC) based on gadoxetate disodium (Gd-EOB-DTPA)-enhanced MRI.

METHODS

239 patients with HCC who underwent curative surgical resection were recruited from two centers between April 2017 and December 2022. Radiomics features were extracted from the region of interest (ROI) on preoperative Gd-EOB-DTPA-enhanced MR images, and consistency analysis was performed to select stable radiomics features. Significant variables in the univariate and multivariate logistic regression analysis were included in clinical-radiologic model. Nomograms were constructed by combining the best performing radiologic and clinical-radiologic characteristics. Recurrence-free survival (RFS) comparisons were conducted using the log-rank test based on high versus low model-derived scores.

RESULTS

The radiomics model based on multiple phases MR outperformed all other radiomics models and had the best discrimination for early recurrence, with AUC of 0.799 and 0.743 in the training and validation sets, respectively. In the entire cohort, high-risk patients exhibited significantly lower RFS compared to low-risk patients.

CONCLUSION

The nomogram integrating Gd-EOB-DTPA enhanced MRI radiomics features and clinical-radiologic characteristics demonstrate superior predictive performance with postoperative early recurrence in patients with HCC. The model can identify patients at high risk and provide support for individualized treatment planning.

Preoperative assessment in lymph node metastasis of pancreatic ductal adenocarcinoma: a transformer model based on dual-energy CT

BACKGROUND

Deep learning(DL) models can improve significantly discrimination of lymph node metastasis(LNM) of pancreatic ductal adenocarcinoma(PDAC), but have not been systematically assessed.

PURPOSE

To develop and test a transformer model utilizing dual-energy computed tomography (DECT) for predicting LNM in patients with PDAC.

MATERIALS AND METHODS

This retrospective study examined patients who had undergone surgical resection and had pathologically confirmed PDAC, with DECT performed between August 2016 and October 2022. Six predictive models were constructed: a DECT report model, a clinical model, 100 keV DL model, 150 keV DL model, a combined 100 + 150 keV DL model, and a model that integrated clinical information with DL-derived signatures. Multivariable logistic regression analysis was employed to develop the integrated model. The efficacy of these models was assessed by comparing their areas under the receiver operating characteristic curve (AUC) using the Delong test. Survival analysis was conducted using Kaplan-Meier curves.

RESULTS

In brief, 223 patients (mean age, 57 years ± 11 standard deviation; 93 men) were evaluated. All patients were divided into training (n = 160) and test (n = 63) sets. Patients with LNM accounted for 96 of the 223 patients (43%). In the test set, the integrated model, which integrated DECT parameters such as IC and Z, CA- 199 levels, DECT reports, and DL signatures, demonstrated the highest performance in predicting LNM, with an AUC of 0.93. In contrast, the radiologists'assessment and the clinical model yielded AUCs of 0.60 and 0.62, respectively. The integrated model-predicted positive LNM was associated with worse overall survival (hazard ratio, 1.75; 95% confidence interval: 1.22 - 2.83; P =.023).

CONCLUSION

A transformer-based model outperformed radiologists and clinical model for prediction of LNM at DECT in patients with PDAC.

CCL26 as a prognostic biomarker in hepatocellular carcinoma: integrating bioinformatics analysis, clinical validation, and radiomics score

BACKGROUND

CCL26 has been identified as a potential prognostic biomarker in hepatocellular carcinoma (HCC). This study aimed to assess the prognostic significance of CCL26 and develop a radiomics score (Rad-score) for predicting outcomes in HCC patients.

METHODS

Data from 316 HCC patients, including genomic information, computed tomography (CT) images, and clinicopathological data, were analyzed. The prognostic value of CCL26 was evaluated in 295 TCGA patients using Kaplan-Meier and Cox regression analyses, and validated in 21 patients from Jiujiang No. 1 People's Hospital. Gene set variation and immune cell infiltration analyses were conducted to elucidate the biological functions of CCL26. Radiomic models for predicting CCL26 expression were constructed using CT images and genomic data from 34 TCGA patients. Radiomic features were extracted from tumor regions and screened using maximum relevance minimum redundancy (mRMR) and recursive feature elimination (RFE). Two Rad-scores were generated via logistic regression and validated using internal fivefold cross-validation. A prognostic nomogram incorporating the optimal Rad-score, gender, and hepatic inflammation was developed using Cox proportional hazards regression.

RESULTS

Elevated CCL26 levels correlated with poor prognosis, as confirmed by immunohistochemistry. The optimal Rad-score, combined with gender and hepatic inflammation, accurately predicted overall survival (OS), with areas under the receiver operating characteristic curve (AUCs) of 0.819, 0.902, and 0.982 for 24-, 36-, and 48 month survival, respectively. Calibration curves and decision curve analysis (DCA) demonstrated the accuracy and clinical utility of the model.

CONCLUSIONS

CCL26 serves as a significant prognostic biomarker in HCC. The developed Rad-score provides an effective, non-invasive tool for predicting patient outcomes and enhancing clinical decision-making. This study not only highlights the prognostic role of CCL26 but also offers a novel approach for evaluating HCC patient prognosis through radiomics.

Radiomics Diagnosis of Microvascular Invasion in Hepatocellular Carcinoma Using 3D Ultrasound and Whole-Slide Image Fusion

This study aims to develop a machine learning model that accurately diagnoses microvascular invasion (MVI) in hepatocellular carcinoma by using radiomic features from MVI-positive regions of interest (ROIs). Unlike previous studies, which do not account for the location and distribution of MVI, this research focuses on correlating preoperative imaging with postoperative pathological MVI. This study involves obtaining ex vivo 3D ultrasound images of 36 hepatic specimens from nine rabbits. These images are fused with whole-slide images to localize MVI regions precisely. The identified MVI regions are segmented into MVI-positive ROIs, with a 1:3 ratio of positive to negative ROIs. Radiomic features are extracted from each ROI, and 30 features highly associated with MVI are selected for model development. The performance of several machine learning models is evaluated using metrics such as sensitivity, specificity, accuracy, the area under the curve (AUC), and F1 score. The GBDT model achieves the best results, with an AUC of 0.91, an F1 score of 0.85, a sensitivity of 0.76, a specificity of 0.92, and an accuracy of 0.86. The high diagnostic accuracy of these models highlights the potential for future clinical application in the precise diagnosis of MVI using radiomic features from MVI-positive ROIs.

Natural Killer Cell-Associated Radiogenomics Model for Hepatocellular Carcinoma: Integrating CD2 and Enhanced CT-Derived Radiomics Signatures

RATIONALE AND OBJECTIVES

Hepatocellular carcinoma (HCC) is a leading cause of cancer mortality. Natural Killer (NK) cells play a crucial role in immune defense against HCC, but their activity is often impaired by the tumor microenvironment (TME). This study aims to integrate radiomics and transcriptomics to develop a prognostic model linking NK cell characteristics to clinical outcomes in HCC.

METHODS

Transcriptomic data from five cohorts (734 HCC patients) from the Gene Expression Omnibus and The Cancer Genome Atlas databases were analyzed using the Microenvironment Cell Populations-counter algorithm. NK cell-related prognostic biomarkers were identified via weighted gene co-expression network analysis and LASSO-Cox regression. Radiomics models were established using CT imaging features from 239 patients in three datasets from The Cancer Imaging Archive and Shanghai East Hospital. HCC radiogenomic subtypes were proposed by integrating genetic biomarkers and radiomics models.

RESULTS

CD2 expression was identified as an independent NK cell-related prognostic biomarker, with a positive impact on prognosis and a strong correlation with NK cell-associated biological processes in HCC. A robust radiomics model was constructed, and the integration of CD2 expression with radioscore identified potential radiogenomic subtypes of HCC.

CONCLUSION

Radiomics has potential to link TME immune phenotypes with HCC prognosis. CD2 is a key biomarker connecting NK cells with radiomic features, offering a new classification of HCC into radiogenomic subtypes. This approach supports the use of radiogenomics in personalized HCC treatment.

Role of Radiomics-based Multiomics Panel in the Microenvironment and Prognosis of Hepatocellular Carcinoma

Hepatocellular carcinoma (HCC) is the most prevalent form of liver tumor, characterized by restricted therapeutic options and typically low long-term survival rates. Recently, immunotherapy has revolutionized HCC treatment, making the tumor microenvironment (TME) a research focus. Radiomics is increasingly crucial in HCC clinical decisions, offering advanced tools for TME characterization and prognosis assessment. Meanwhile, advancements in pathomics provide new insights into HCC's comprehensive traits and details. Advancements in genomics and transcriptomics enable the integration of radiomics and pathomics with genetic data to better understand HCC heterogeneity and its microenvironment, aiding prognostic assessments. This review provides a comprehensive overview of pivotal radiomics studies focused on TME prediction, underscoring the synergistic effects of integrating multiomics approaches for TME analysis and HCC outcome prediction. It critically examines the challenges and opportunities inherent in multiomics research, emphasizing its substantial significance in both research and clinical contexts.

Automatic Diagnosis of Hepatocellular Carcinoma and Metastases Based on Computed Tomography Images

Liver cancer, a leading cause of cancer mortality, is often diagnosed by analyzing the grayscale variations in liver tissue across different computed tomography (CT) images. However, the intensity similarity can be strong, making it difficult for radiologists to visually identify hepatocellular carcinoma (HCC) and metastases. It is crucial for the management and prevention strategies to accurately differentiate between these two liver cancers. This study proposes an automated system using a convolutional neural network (CNN) to enhance diagnostic accuracy to detect HCC, metastasis, and healthy liver tissue. This system incorporates automatic segmentation and classification. The liver lesions segmentation model is implemented using residual attention U-Net. A 9-layer CNN classifier implements the lesions classification model. Its input is the combination of the results of the segmentation model with original images. The dataset included 300 patients, with 223 used to develop the segmentation model and 77 to test it. These 77 patients also served as inputs for the classification model, consisting of 20 HCC cases, 27 with metastasis, and 30 healthy. The system achieved a mean Dice score of 87.65 % in segmentation and a mean accuracy of 93.97 % in classification, both in the test phase. The proposed method is a preliminary study with great potential in helping radiologists diagnose liver cancers.

Precision surgery for hepatocellular carcinoma

Hepatocellular carcinoma arises in the setting of cirrhosis in most cases, requiring multidisciplinary input to define resectability. In this regard, more precise surgical management considers patient factors and anatomical states, including resection margins, tumour biology, and perioperative therapy. Together with advances in surgical techniques, this integrated approach has resulted in considerable improvements in patient morbidity and oncological outcomes. Despite this, recurrence rates in hepatocellular carcinoma remain high. As the systemic treatment landscape in hepatocellular carcinoma continues to evolve and locoregional options are increasingly used, we review current and future opportunities to individualise the surgical management of patients with hepatocellular carcinoma.

Decision Fusion Model for Predicting Microvascular Invasion in Hepatocellular Carcinoma Based on Multi-MR Habitat Imaging and Machine-Learning Classifiers

RATIONALE AND OBJECTIVES

Accurate prediction of microvascular invasion (MVI) in hepatocellular carcinoma (HCC) is crucial for guiding treatment. This study evaluates and compares the performance of clinicoradiologic, traditional radiomics, deep-learning radiomics, feature fusion, and decision fusion models based on multi-region MR habitat imaging using six machine-learning classifiers.

MATERIALS AND METHODS

We retrospectively included 300 HCC patients. The intratumoral and peritumoral regions were segmented into distinct habitats, from which radiomics and deep-learning features were extracted using arterial phase MR images. To reduce feature dimensionality, we applied intra-class correlation coefficient (ICC) analysis, Pearson correlation coefficient (PCC) filtering, and recursive feature elimination (RFE). Based on the selected optimal features, prediction models were constructed using decision tree (DT), K-nearest neighbors (KNN), logistic regression (LR), random forest (RF), support vector machine (SVM), and XGBoost (XGB) classifiers. Additionally, fusion models were developed utilizing both feature fusion and decision fusion strategies. The performance of these models was validated using the area under the receiver operating characteristic curve (ROC AUC), calibration curves, and decision curve analysis.

RESULTS

The decision fusion model (VOI-Peri10-1) using LR and integrating clinicoradiologic, radiomics, and deep-learning features achieved the highest AUC of 0.808 (95% confidence interval [CI]: 0.807-0.912) in the test cohort, with good calibration (Hosmer-Lemeshow test, P > 0.050) and clinical net benefit.

CONCLUSION

The LR-based decision fusion model integrating clinicoradiologic, radiomics, and deep-learning features shows promise for preoperative prediction of MVI in HCC, aiding in patient outcome predictions and personalized treatment planning.

Real-world efficacy of radiomics versus clinical predictors for microvascular invasion in patients with hepatocellular carcinoma: Large cohort study

AIM

Microvascular invasion (MVI) affects the prognosis and treatment of hepatocellular carcinoma (HCC); however, its preoperative diagnosis is challenging. Analysis of computed tomography (CT) images using radiomics can detect MVI, but its effectiveness depends on the imaging conditions. We compared the efficacies of radiomics, clinical, and combined models for predicting MVI in HCC using nonstandardized scanning protocols.

METHODS

This multicenter study included 533 patients who underwent hepatic resection for HCC. Patients were divided randomly into training (n = 426) and test groups (n = 107). We manually extracted 3D CT features in hepatic arterial, portal venous, and venous phases. The radiomics model was trained by machine learning. A logistic regression model was developed based on clinical information, and a fused model was created integrating clinical information and radiomics prediction score (Rad_Score). We calculated areas under the receiver operating characteristic curves (AUCs) for the radiomics, clinical, and mixed models in the test groups.

RESULTS

The clinical model incorporated hepatitis B virus surface antigen, tumor diameter, and log-transformed α-fetoprotein and des-gamma-carboxyprothrombin. The AUCs of the radiomics and clinical models were comparable (p = 0.76). Rad_Score was not an independent significant factor in the fused model (p = 0.40) and its addition did not improve the accuracy of the clinical model alone (p = 0.51).

CONCLUSIONS

A clinical model is as effective as a CT radiomics model for predicting MVI status in patients with HCC based on real-world scanning data, and integration of both models does not improve the predictive performance compared with a clinical model alone.

Contrast-Enhanced Computed Tomography Radiomics Predicts Colony-Stimulating Factor 3 Expression and Clinical Prognosis in Ovarian Cancer

RATIONALE AND OBJECTIVES

To develop a radiomics model for non-invasive prediction of colony-stimulating factor 3 (CSF3) expression in ovarian cancer (OC) and evaluate its prognostic value.

MATERIALS AND METHODS

We acquired clinical data, genetic information, and corresponding computed tomography (CT) scans of OC from The Cancer Genome Atlas and The Cancer Imaging Archive repositories. We assessed the prognostic significance of CSF3 and its association with clinical features through the utilization of Kaplan-Meier analysis, univariate and multivariate Cox regression analysis, along with subgroup analysis. To explore the potential molecular mechanisms associated with CSF3 expression, we utilized gene set enrichment analysis and conducted an analysis on immune-cell infiltration. The max-relevance and min-redundancy and recursive feature elimination (RFE) algorithms were used for feature screening. The CT-based radiomics prediction model was built using support vector machine (SVM) and logistic regression (LR).

RESULTS

The expression of CSF3 was found to be decreased in OC, and high expression of CSF3 was associated with poor overall survival. Moreover, it was noted that the expression of CSF3 exhibited a positive correlation with programmed death ligand 1 (PD-L1) and sialic acid-binding Ig-like lectin 15 (SIGLEC15). Patients with high CSF3 expression exhibited a decrease in tumor necrosis factor receptor superfamily member 7 (CD27) expression. The infiltration of neutrophils increased and CD8 +T cells decreased in CSF3 high expression group.

CONCLUSION

The radiomics model, which utilized both LR and SVM methods, demonstrated significant clinical applicability. The expression level of CSF3 was related to the prognosis of OC. Radiomics based on CT can serve as a novel tool for forecasting prognosis.

Diagnosis of parotid gland tumors using a ternary classification model based on ultrasound radiomics

Objective

This study aimed to evaluate the diagnostic value of two-step ultrasound radiomics models in distinguishing parotid malignancies from pleomorphic adenomas (PAs) and Warthin's tumors (WTs).

Methods

A retrospective analysis was conducted on patients who underwent parotidectomy at our institution between January 2015 and December 2022. Radiomics features were extracted from two-dimensional (2D) ultrasound images using 3D Slicer. Feature selection was performed using the Mann-Whitney U test and seven additional selection methods. Two-step LASSO-BNB and voting ensemble learning modeling algorithm with recursive feature elimination feature selection method (RFE-Voting) models were then applied for classification. Model performance was assessed using the area under the receiver operating characteristic curve (AUC), and internal validation was conducted through fivefold cross-validation.

Results

A total of 336 patients were included in the study, comprising 73 with malignant tumors and 263 with benign lesions (118 WT and 145 PA). The LASSO-NB model demonstrated excellent performance in distinguishing between benign and malignant parotid lesions, achieving an AUC of 0.910 (95% CI, 0.907-0.914), with an accuracy of 86.8%, sensitivity of 92.5%, and specificity of 66.7%, significantly outperforming experienced sonographers (accuracy of 61.90%). The RFE-Voting model also showed outstanding performance in differentiating PA from WT, with an AUC of 0.962 (95% CI, 0.959-0.963), accuracy of 83.0%, sensitivity of 84.0%, and specificity of 92.1%, exceeding the diagnostic capability of experienced sonographers (accuracy of 65.39%).

Conclusion

The two-step LASSO-BNB and RFE-Voting models based on ultrasound imaging performed well in distinguishing glandular malignant tumors from PA and WT and have good predictive capabilities, which can provide more useful information for non-invasive differentiation of parotid gland tumors before surgery.

Combining radiomics and imaging biomarkers with clinical variables for the prediction of HCC recurrence after liver transplantation

To develop and validate an integrated model that combines CT-based radiomics and imaging biomarkers with clinical variables to predict recurrence and recurrence-free survival in patients with HCC following liver transplantation (LT), this 2-center retrospective study includes 123 patients with HCC who underwent LT between 2007 and 2021. Radiomic features (RFs) were extracted from baseline CT liver tumor volume. Feature selection was performed using the Least Absolute Shrinkage and Selection Operator (LASSO) regression method with 10-fold cross-validation in the training cohort (n=48) to build a predictive radiomics signature for HCC recurrence. Combined diagnostic models were built based on the radiomics signature supplemented with imaging features beyond the Milan criteria, the AFP (alpha-fetoprotein) model, and Metroticket 2.0 before LT using multivariate logistic regression. Receiver operating characteristic analyses were performed in both internal (n=22) and external (n=53) validation cohorts, and patients were stratified into either high-risk or low-risk groups for HCC recurrence. Kaplan-Meier analysis was performed to analyze recurrence-free survival. LASSO and multivariate regression analysis revealed 4 independent predictors associated with an increased risk of HCC recurrence: radiomics signature of 5 RF, peritumoral enhancement, satellite nodules, and no bridging therapies. For the prediction of tumor recurrence, the highest AUC of the final integrated models combining clinical variables, non-radiomics imaging features, and radiomics was 0.990 and 0.900 for the internal and external validation sets, respectively, outperforming the Milan and clinical stand-alone models. In all integrated models, the high-risk groups had a shorter recurrence-free survival than the corresponding low-risk group. CT-based radiomics and imaging parameters beyond the Milan criteria representing aggressive behavior, along with the history of bridging therapies, show potential for predicting HCC recurrence after LT.

Multiregional Radiomics to Predict Microvascular Invasion in Hepatocellular Carcinoma Using Multisequence MRI

OBJECTIVES

This study aimed to develop a multiregional radiomics-based model using multisequence MRI to predict microvascular invasion in hepatocellular carcinoma.

METHODS

We enrolled 141 patients with hepatocellular carcinoma, including 61 with microvascular invasion, who were diagnosed between March 2017 and July 2022. Clinical data were compared using the Wilcoxon rank-sum test or χ2 test. Patients were randomly divided into training (n=112, 80%) and test (n=29, 20%) data sets. Four MRI sequences-including T2-weighted imaging, T2-weighted imaging with fat suppression, arterial phase-contrast enhancement, and portal venous phase contrast enhancement-were used to build the radiomics model. The tumor volumes of interest were manually delineated, and the expand-5 mm and expand-10 mm volumes of interest were automatically generated. A total of 1409 radiomic features were extracted from each volume of interest. Feature selection was performed using the least absolute shrinkage and selection operator and Spearman correlation analysis. Three logistic regression models (Tumor, Tumor-Expand5, and Tumor-Expand10) were established based on the radiomic features. Model performance was assessed using receiver operating characteristic analysis and Delong's test.

RESULTS

Maximum tumor diameter, hepatitis B virus DNA, and aspartate aminotransferase levels were significantly different between the groups. The Tumor-Expand5mm model exhibited the best performance among the 3 models, with areas under the curve of 0.90 and 0.84 in the training and test data sets.

CONCLUSIONS

The Tumor-Expand5 model based on multisequence MRI shows great potential for predicting microvascular invasion in patients with hepatocellular carcinoma, and may further contribute to personal clinical decision-making.

Integrating CT Radiomics and Clinical Features to Optimize TACE Technique Decision-Making in Hepatocellular Carcinoma

BACKGROUND/OBJECTIVES

To develop a decision framework integrating computed tomography (CT) radiomics and clinical factors to guide the selection of transarterial chemoembolization (TACE) technique for optimizing treatment response in non-resectable hepatocellular carcinoma (HCC).

METHODS

A retrospective analysis was performed on 151 patients [33 conventional TACE (cTACE), 69 drug-eluting bead TACE (DEB-TACE), 49 degradable starch microsphere TACE (DSM-TACE)] who underwent TACE for HCC at a single tertiary center. Pre-TACE contrast-enhanced CT images were used to extract radiomic features of the TACE-treated liver tumor volume. Patient clinical and laboratory data were combined with radiomics-derived predictors in an elastic net regularized logistic regression model to identify independent factors associated with early response at 4-6 weeks post-TACE. Predicted response probabilities under each TACE technique were compared with the actual techniques performed.

RESULTS

Elastic net modeling identified three independent predictors of response: radiomic feature "Contrast" (OR = 5.80), BCLC stage B (OR = 0.92), and viral hepatitis etiology (OR = 0.74). Interaction models indicated that the relative benefit of each TACE technique depended on the identified patient-specific predictors. Model-based recommendations differed from the actual treatment selected in 66.2% of cases, suggesting potential for improved patient-technique matching.

CONCLUSIONS

Integrating CT radiomics with clinical variables may help identify the optimal TACE technique for individual HCC patients. This approach holds promise for a more personalized therapy selection and improved response rates beyond standard clinical decision-making.

Integrating radiomics into predictive models for low nuclear grade DCIS using machine learning

Predicting low nuclear grade DCIS before surgery can improve treatment choices and patient care, thereby reducing unnecessary treatment. Due to the high heterogeneity of DCIS and the limitations of biopsies in fully characterizing tumors, current diagnostic methods relying on invasive biopsies face challenges. Here, we developed an ensemble machine learning model to assist in the preoperative diagnosis of low nuclear grade DCIS. We integrated preoperative clinical data, ultrasound images, mammography images, and Radiomic scores from 241 DCIS cases. The ensemble model, based on Elastic Net, Generalized Linear Models with Boosting (glmboost), and Ranger, improved the ability to predict low nuclear grade DCIS preoperatively, achieving an AUC of 0.92 on the validation set, outperforming the model using clinical data alone. The comprehensive model also demonstrated notable enhancements in integrated discrimination improvement and net reclassification improvement (p < 0.001). Furthermore, the Radiomic ensemble model effectively stratified DCIS patients by risk based on disease-free survival. Our findings emphasize the importance of integrating Radiomic into DCIS prediction models, offering fresh perspectives for personalized treatment and clinical management of DCIS.

The Association Between Tumor Radiomic Analysis and Peritumor Habitat-Derived Radiomic Analysis on Gadoxetate Disodium-Enhanced MRI With Microvascular Invasion in Hepatocellular Carcinoma

BACKGROUND

Hepatocellular carcinoma (HCC) has a poor prognosis, often characterized by microvascular invasion (MVI). Radiomics and habitat imaging offer potential for preoperative MVI assessment.

PURPOSE

To identify MVI in HCC by habitat imaging, tumor radiomic analysis, and peritumor habitat-derived radiomic analysis.

STUDY TYPE

Retrospective.

SUBJECTS

Three hundred eighteen patients (53 ± 11.42 years old; male = 276) with pathologically confirmed HCC (training:testing = 224:94).

FIELD STRENGTH/SEQUENCE

1.5 T, T2WI (spin echo), and precontrast and dynamic T1WI using three-dimensional gradient echo sequence.

ASSESSMENT

Clinical model, habitat model, single sequence radiomic models, the peritumor habitat-derived radiomic model, and the combined models were constructed for evaluating MVI. Follow-up clinical data were obtained by a review of medical records or telephone interviews.

STATISTICAL TESTS

Univariable and multivariable logistic regression, receiver operating characteristic (ROC) curve, calibration, decision curve, Delong test, K-M curves, log rank test. A P-value less than 0.05 (two sides) was considered to indicate statistical significance.

RESULTS

Habitat imaging revealed a positive correlation between the number of subregions and MVI probability. The Radiomic-Pre model demonstrated AUCs of 0.815 (95% CI: 0.752-0.878) and 0.708 (95% CI: 0.599-0.817) for detecting MVI in the training and testing cohorts, respectively. Similarly, the AUCs for MVI detection using Radiomic-HBP were 0.790 (95% CI: 0.724-0.855) for the training cohort and 0.712 (95% CI: 0.604-0.820) for the test cohort. Combination models exhibited improved performance, with the Radiomics + Habitat + Dilation + Habitat 2 + Clinical Model (Model 7) achieving the higher AUC than Model 1-4 and 6 (0.825 vs. 0.688, 0.726, 0.785, 0.757, 0.804, P = 0.013, 0.048, 0.035, 0.041, 0.039, respectively) in the testing cohort. High-risk patients (cutoff value >0.11) identified by this model showed shorter recurrence-free survival.

DATA CONCLUSION

The combined model including tumor size, habitat imaging, radiomic analysis exhibited the best performance in predicting MVI, while also assessing prognostic risk.

EVIDENCE LEVEL

3 TECHNICAL EFFICACY: Stage 2.

Development models to predict complication and prognosis following liver resection for hepatocellular carcinoma in patients with clinically significant portal hypertension

BACKGROUND

Postoperative complications are potential factors influencing the prognosis of patients with HCC combined with CSPH. This study aims to explore the risk factors affecting the occurrence of postoperative complications, investigate potential factors influencing long-term prognosis in these patients, and establish predictive models.

METHODS

From April 2018 to December 2021, a total of 190 patients with HCC combined with CSPH who underwent curative liver resection in our hospital were included, comprising 69 cases in the complication group and 121 cases in the non-complication group. LASSO-Logistic regression was employed to identify risk factors influencing postoperative complications and establish a predictive model. LASSO-Cox regression was used to determine prognostic factors for long-term outcomes in patients with HCC combined with CSPH and establish a predictive model.

RESULTS

LASSO regression selected variables including ALBI grade, preoperative ascites, major hepatectomy, and portal vein occlusion time >15 ​min. These variables were incorporated into logistic regression (P ​< ​0.05) to establish a nomogram for predicting postoperative complications, with a C-index of 0.723. Results from the multivariable Cox regression analysis showed that postoperative complications, maximum tumor diameter, and microvascular invasion were risk factors for recurrence, while postoperative complications, maximum tumor diameter, microvascular invasion, and prealbumin were risk factors for overall survival. The C-index values for the respective nomograms were 0.635 and 0.734. The calibration curves and ROC curves demonstrated good performance for all three nomograms.

CONCLUSIONS

The three nomograms achieved optimal predictive performance for postoperative complications, recurrence, and overall survival in patients with HCC combined with CSPH undergoing curative resection.

Predicting Microvascular Invasion and Early Recurrence in Hepatocellular Carcinoma Using DeepLab V3+ Segmentation of Multiregional MR Habitat Images

RATIONALE AND OBJECTIVES

Accurate identification of microvascular invasion (MVI) in hepatocellular carcinoma (HCC) is crucial for treatment and prognosis. Single-modality and feature fusion models using manual segmentation fail to provide insights into MVI. This study aims to develop a DeepLab V3+ model for automated segmentation of HCC magnetic resonance (MR) images and a decision fusion model to predict MVI and early recurrence (ER).

MATERIALS AND METHODS

This retrospective study included 209 HCC patients (146 in the training and 63 in the test cohorts). The performance of DeepLab V3+ for HCC MR image segmentation was evaluated using Dice Loss and F1 score. Intraclass correlation coefficients (ICCs) assessed feature extraction reliability. Spearman's correlation analyzed the relationship between tumor volumes from automated and manual segmentation, with agreement evaluated using Bland-Altman plots. Model performance was assessed using the area under the receiver operating characteristic curve (ROC AUC), calibration curves, and decision curve analysis. A nomogram predicted ER of HCC after surgery, with Kaplan-Meier analysis for 2-year recurrence-free survival (RFS).

RESULTS

The DeepLab V3+ model demonstrated high segmentation accuracy, with strong agreement in feature extraction (ICC: 0.802-0.999). The decision fusion model achieved AUCs of 0.968 and 0.878 for MVI prediction, and the nomogram for predicting ER yielded AUCs of 0.782 and 0.690 in the training and test cohorts, respectively, with significant RFS differences between the risk groups.

CONCLUSION

The DeepLab V3+ model accurately segmented HCC. The decision fusion model significantly improved MVI prediction, and the nomogram offered valuable insights into recurrence risk for clinical decision-making.

AVAILABILITY OF DATA AND MATERIALS

The datasets used and/or analyzed during the current study are available from the corresponding author upon reasonable request.

Preoperative clinical radiomics model based on deep learning in prognostic assessment of patients with gallbladder carcinoma

OBJECTIVE

We aimed to develop a preoperative clinical radiomics survival prediction model based on the radiomics features via deep learning to provide a reference basis for preoperative assessment and treatment decisions for patients with gallbladder carcinoma (GBC).

METHODS

A total of 168 GBC patients who underwent preoperative upper abdominal enhanced CT from one high-volume medical center between January 2011 to December 2020 were retrospectively analyzed. The region of interest (ROI) was manually outlined by two physicians using 3D Slicer software to establish a nnU-Net model. The DeepSurv survival prediction model was developed by combining radiomics features and preoperative clinical variables.

RESULTS

A total of 1502 radiomics features were extracted from the ROI results based on the nnU-Net model and manual segmentation, and 13 radiomics features were obtained through the 4-step dimensionality reduction methods, respectively. The C-index and AUC of 1-, 2-, and 3-year survival prediction for the nnU-Net based clinical radiomics DeepSurv model was higher than clinical and nnU-Net based radiomics DeepSurv models in the training and testing sets, and close to manual based clinical radiomics DeepSurv model. Delong-test was performed on the AUC of 1-, 2-, and 3-year survival prediction for the two preoperative clinical radiomics DeepSurv prediction models in the testing set, and the results showed that the two models had the same prediction efficiency (all P > 0.05).

CONCLUSIONS

By using the DeepSurv model via nnU-Net segmentation, postoperative survival outcomes for individual gallbladder carcinoma patients could be assessed and stratified, which can provide references for preoperative diagnosis and treatment decisions.

18F-FDG PET/CT-based intratumoral and peritumoral radiomics combining ensemble learning for prognosis prediction in hepatocellular carcinoma: a multi-center study

BACKGROUND

Radiomic models combining intratumoral with peritumoral features are potentially beneficial to enhance the predictive performance. This study aimed to identify the optimal 18F-FDG PET/CT-derived radiomic models for prediction of prognosis in hepatocellular carcinoma (HCC).

METHODS

A total of 135 HCC patients from two institutions were retrospectively included. Four peritumoral regions were defined by dilating tumor region with thicknesses of 2 mm, 4 mm, 6 mm, and 8 mm, respectively. Based on segmentation of intratumoral, peritumoral and integrated volume of interest (VOI), corresponding radiomic features were extracted respectively. After feature selection, a total of 15 intratumoral radiomic models were constructed based on five ensemble learning algorithms and radiomic features from three image modalities. Then, the optimal combination of ensemble learning algorithms and image modality in the intratumoral models was selected to develop subsequent peritumoral radiomic models and integrated radiomic models. Finally, a nomogram was developed incorporating the optimal radiomic model with clinical independent predictors to achieve an intuitive representation of the prediction model.

RESULTS

Among the intratumoral radiomic models, the one which combined PET/CT-based radiomic features with SVM classifier outperformed other models. With the addition of peritumoral information, the integrated model based on an integration of intratumoral and 2 mm-peritumoral VOI, was finally approved as the optimal radiomic model with a mean AUC of 0.831 in the internal validation, and a highest AUC of 0.839 (95%CI:0.718-0.960) in the external test. Furthermore, a nomogram incorporating the optimal radiomic model with HBV infection and TNM status, was able to predict the prognosis for HCC with an AUC of 0.889 (95%CI: 0.799-0.979).

CONCLUSIONS

The integrated intratumoral and peritumoral radiomic model, especially for a 2 mm peritumoral region, was verified as the optimal radiomic model to predict the overall survival of HCC. Furthermore, combination of integrated radiomic model with significant clinical parameter contributed to further enhance the prediction efficacy.

TRIAL REGISTRATION

This study was a retrospective study, so it was free from registration.

Development and clinical validation of a novel platelet count-based nomogram for predicting microvascular invasion in HCC

We aimed to develop a convenient nomogram to predict preoperative MVI in patients with hepatocellular carcinoma (HCC). Patients who underwent surgical resection due to HCC from June 2018 to June 2023 at the Third Affiliated Hospital of Sun Yat-sen University were retrospectively reviewed. Univariate and multivariable logistic linear regression analyses were used to investigate potential risk factors for MVI. A nomogram was plotted based on these risk factors. The tumor diameter (≥ 5 cm), BCLC stage, PLT (>127.50 × 109/L), AST (>29.50 U/L) and AFP (>10.07 ng/ml) were identified as independent preoperative risk factors for MVI by univariate and multivariable logistic analysis. The nomogram demonstrated decent accuracy in estimating the presence of MVI, with an AUC of 0.69 (95%CI: 0.64-0.73). The calibration curves exhibited a close match between the predicted probabilities and the actual estimates of MVI in the nomogram (p = 0.947). Decision curve analysis (DCA) revealed that the prediction model had a high net benefit if the threshold probability>20%. High platelet counts were strongly associated with the presence of MVI in HCC patients. Our convenient nomogram demonstrated decent accuracy in estimating the presence of MVI and had notable clinical application.

MRI Features and Neutrophil-to-Lymphocyte Ratio (NLR)-Based Nomogram to Predict Prognosis of Microvascular Invasion-Negative Hepatocellular Carcinoma

Purpose

This study aimed to develop a novel nomogram to predict recurrence-free survival (RFS) for microvascular invasion (MVI)-negative hepatocellular carcinoma (HCC) patients after curative resection.

Patients and Methods

A total of 143 pathologically confirmed MVI-negative HCC patients were analyzed retrospectively. Baseline MRI features and inflammatory markers were collected. We used univariable and multivariable Cox regression analysis to identify the independent risk factors for RFS. And we established a nomogram based on significant MRI features and inflammatory marker. The receiver operating characteristic (ROC) curve, concordance index (C-index) and calibration curve were used to evaluate the predictive accuracy and discriminative ability of the nomogram. The decision curve analysis (DCA) was performed to validate the clinical utility of the nomogram.

Results

In multivariate Cox regression analysis, neutrophil-to-lymphocyte ratio (NLR) (P = 0.018), tumor size (P = 0.002), and tumor capsule (P = 0.000) were independent significant variables associated with RFS. Nomogram with independent factors was developed and achieved a good C-index of 0.730 (95% confidence interval [CI]: 0.656-0.804) for predicting RFS. In ROC analysis, the areas under curve of the nomogram for 1-, 3- and 5-year RFS prediction were 0.725, 0.784 and 0.798, respectively. The risk score calculated by nomogram could divide MVI-negative HCC patients into high-risk group or low-risk group (P < 0.0001). DCA analysis revealed that the nomogram could increase net benefit and exhibited a wider range of threshold probabilities by the risk stratification than the independent risk factors in the prediction of MVI-negative HCC recurrence.

Conclusion

The nomogram prognostic model based on MRI features and NLR for predicting RFS showed high accuracy in MVI-negative HCC patients after curative resection. It can help clinicians make treatment decisions for MVI-negative HCC patients and identify high-risk patients for timely intervention.

Impact of microvascular invasion risk on tumor progression of hepatocellular carcinoma after conventional transarterial chemoembolization

OBJECTIVE

To assess tumor progression in patients with hepatocellular carcinoma (HCC) without macrovascular invasion who underwent treatment with conventional transarterial chemoembolization (cTACE) based on microvascular invasion (MVI) risk within 2 years.

METHODS

This retrospective investigation comprised adult patients with HCC who had either liver resection or cTACE as their first treatment from January 2016 to December 2021. A predictive model for MVI was developed and validated using preoperative clinical and MRI data from patients with HCC treated with liver resection. The MVI predictive model was applied to patients with HCC receiving cTACE, and differences in tumor progression between the MVI high- and low-risk groups were examined throughout 2 years.

RESULTS

The MVI prediction model incorporated nonsmooth margin, intratumoral artery, incomplete or absent tumor capsule, and tumor DWI/T2WI mismatch. The area under the receiver operating characteristic curve (AUC) for the prediction model, in the training cohort, was determined to be 0.904 (95% CI, 0.862-0.946), while in the validation cohort, it was 0.888 (0.782-0.994). Among patients with HCC undergoing cTACE, those classified as high risk for MVI possessed a lower rate of achieving a complete response after the first tumor therapy and a higher risk of tumor progression within 2 years.

CONCLUSIONS

The MVI prediction model developed in this study demonstrates a considerable degree of accuracy. Patients at high risk for MVI who underwent cTACE treatment exhibited a higher risk of tumor progression within 2 years.

Evolving Indications for Liver Transplantation for Hepatocellular Carcinoma Following the Milan Criteria

Background/Objectives: Since their introduction in the 1990s, the Milan criteria have been the gold standard of indication for liver transplantation (LT) in patients with hepatocellular carcinoma (HCC). Nevertheless, several institutions have reported wider indication criteria for LT with comparable survival outcomes. Methods: This paper summarizes the recent indications for LT for HCC through a literature review. Results: There are several criteria expanding the Milan criteria, which can be subdivided into the "based on tumor number and size only", "based on tumor number and size plus tumor markers", and "based on tumor differentiation" groups, with the outcomes being comparable to those of patients included within the Milan criteria. Besides the tumor size and number, which are included in the Milan criteria, recent criteria included biomarkers and tumor differentiation. Several retrospective studies have reported microvascular invasion (MVI) as a significant risk factor for postoperative recurrence, highlighting the importance of preoperatively predicting MVI. Several studies attempted to identify preoperative predictive factors for MVI using tumor markers or preoperative imaging findings. Patients with HCC who are LT candidates are often treated while on the waiting list to prevent the progression of HCC or to reduce the measurable disease burden of HCC. The expanding repertoire of chemotherapeutic regiments suitable for patients with HCC should be further investigated. Conclusions: There are several criteria expanding Milan criteria, with the outcomes being comparable to those of patients included within the Milan criteria.

A nomogram incorporating CT-based peri-hematoma radiomics features to predict functional outcome in patients with intracerebral hemorrhage

OBJECTIVE

To evaluate the ability of non-contrast computed tomography based peri-hematoma and intra-hematoma radiomic features to predict the 90-day poor functional outcome for spontaneous intracerebral hemorrhage (sICH) and to present an effective clinically relevant machine learning system to assist in prognosis prediction.

MATERIALS AND METHODS

We retrospectively analyzed the data of 691 patients diagnosed with sICH at two medical centers. Fifteen radiomic features from the intra- and peri-hematoma regions were extracted and selected to build six radiomics models. The clinical-semantic model and nomogram model were constructed to compare prediction abilities. The areas under the curve (AUC) and decision curve analysis were used to evaluate discriminative performance.

RESULTS

Combining radiomics of the intra-hematoma with peri-hematoma regions significantly improved the AUC to 0.843 compared with radiomics of the intra-hematoma region (AUC = 0.780, P ＜ 0.001) in the test set. A similar trend was observed in the external validation cohort (AUC, 0.769 vs. 0.793, P = 0.709). The nomogram, which integrates clinical-semantic signatures with intra-hematoma and peri-hematoma radiomics signatures, accurately predicted a 90-day poor functional outcome in both the test and external validation sets (AUC 0.879 and 0.901, respectively).

CONCLUSION

The nomogram constructed using clinical-semantic signatures and combined intra-hematoma and peri-hematoma radiomics signatures showed the potential to precisely predict 90-day poor functional outcomes for sICH.

A prognostic and predictive model based on deep learning to identify optimal candidates for intensity-modulated radiotherapy alone in patients with stage II nasopharyngeal carcinoma: A retrospective multicenter study

PURPOSE

To develop and validate a prognostic and predictive model integrating deep learning MRI features and clinical information in patients with stage II nasopharyngeal carcinoma (NPC) to identify patients with a low risk of progression for whom intensity-modulated radiotherapy (IMRT) alone is sufficient.

METHODS

This multicenter, retrospective study enrolled 999 patients with stage II NPC from two centers. 3DResNet was used to extract deep learning MRI features and eXtreme Gradient Boosting model was employed to integrate the pre-trained features and clinical information to obtain an overall score for each patient. Based on the optimal cutoff value of the overall score, patients were stratified into high- and low- risk groups. Model performance was evaluated using concordance indexes (C-indexes), the area under the curve (AUC) values and calibration tests. Survival curves were used to analyze the clinical benefits of additional chemotherapy in each risk group.

RESULTS

The combined model achieved a concordance index of 0.789 (95 % confidence interval [CI] 0.787-0.791), 0.768 (95 % CI 0.764-0.771), and 0.804 (95 % CI 0.801-0.807) for the training, internal validation, and external test cohorts, respectively, demonstrating a statistically significant improvement compared to the MRI model, T Stage, and N Stage. An overall score of < 0.405 in patients was significantly associated with a low risk of progression. In the low-risk group, patients treated with IMRT alone had comparable or even superior progression-free survival (PFS) compared to those who received additional chemotherapy.

CONCLUSION

The model demonstrated a satisfactory prognostic and predictive performance for PFS. Patients with stage II NPC were stratified into different risk groups to help identify optimal candidates who could benefit from IMRT alone.

Habitat radiomics based on CT images to predict survival and immune status in hepatocellular carcinoma, a multi-cohort validation study

BACKGROUND AND OBJECTIVE

Though several clinicopathological features are identified as prognostic indicators, potentially prognostic radiomic models are expected to preoperatively and noninvasively predict survival for HCC. Traditional radiomic models are lacking in a consideration for intratumoral regional heterogeneity. The study aimed to establish and validate the predictive power of multiple habitat radiomic models in predicting prognosis of hepatocellular carcinoma (HCC).

METHODS

A total of 232 HCC patients were retrospectively included, including a training/validation cohort and two external testing cohorts from 4 centers. For habitat radiomics, intratumoral habitat partitioning based on CT images was first performed by using Otsu thresholding method. Second, a total of 350 habitat radiomic models were constructed to select the optimal model. Then, both ROC curve analyses and Kaplan-Meier survival curve analyses were applied to assess the predictive performances. Ultimately, an immune status profiling was conducted based on bioinformatic analyses and multiplex immunohistochemistry (mIHC) assays to reveal the potential mechanisms.

RESULTS

A total of 4 habitats were segmented, and the corresponding habitat radiomic models were constructed based on each habitat and an integration of all the four habitats. Generally, habitat radiomic models outperformed traditional radiomic models in stratifying prognosis for HCC. The habitat radiomic model based on the segmented habitat 4 involving decision tree (DT) screening and random forest (RF) classifier was identified as the optimal model with an AUCmean of 0.806. Distinct resting natural killer (NK) cell infiltrations significantly contributed to the prognosis stratification of HCC by the optimal habitat radiomic model.

CONCLUSIONS

The habitat radiomic model based on CT images was potentially predictive of overall survival for HCC, with a superiority over the traditional radiomic model. The prognostic power of the habitat radiomic model was partly attributed to the distinct immune status captured in the CT images.

TRACE Model: Predicting Treatment Response to Transarterial Chemoembolization in Unresectable Hepatocellular Carcinoma

Purpose

To develop and validate a predictive model for predicting six-month outcome by integrating pretreatment MRI features and one-month treatment response after TACE.

Methods

A total of 108 patients with 160 hCCs from a single-arm, multicenter clinical trial (NCT03113955) were analyzed and served as the training cohort. An external multicenter dataset (ChiCTR2100046020) consisting of 63 patients with 99 hCCs served as the test dataset. Radiomics model was constructed based on the selected features from pretreatment MR images. Univariate and multivariate logistic regression analysis of clinical and radiological factors were used to identify the independent predictors for the 6-month treatment response. A combined model was further constructed by incorporating one-month treatment response, selected clinical and radiological factors and radiomics signature.

Results

Among all the clinical and radiological features, only corona enhancement and one-month treatment response were selected. The combined model, named TRACE model (Treatment response at 1 month, RAdiomics and Corona Enhancement), with AUCs of 0.91 (training cohort) and 0.84 (test cohort). The TRACE model demonstrated a significantly higher AUC than the radiomics model (P = 0.001). High-risk and low-risk groups stratified by using the TRACE model also exhibited significant differences in overall survival (OS) (P < 0.001). In contrast, none of the published scoring systems, including ART, SNACOR or ABCR score, demonstrated significant differences between the risk groups in OS prediction.

Conclusion

The TRACE model exhibited favorable predictive capability for six-month TACE response, and holds potential as a marker for long-term survival outcomes.

A clinical study exploring the prediction of microvascular invasion in hepatocellular carcinoma through the use of combined enhanced CT and MRI radiomics

OBJECTIVE

To develop a predictive model for microvascular invasion (MVI) in hepatocellular carcinoma (HCC) through radiomics analysis, integrating data from both enhanced computed tomography (CT) and magnetic resonance imaging (MRI).

METHODS

A retrospective analysis was conducted on 93 HCC patients who underwent partial hepatectomy. The gold standard for MVI was based on the histopathological diagnosis of the tissue. The 93 patients were randomly divided into training and validation groups in 7:3 ratio. The imaging data of patients, including CT and MRI, were collected and processed using 3D Slicer to delineate the region of interest (ROI) for each tumor. Radiomics features were extracted from CT and MRI of patients using Python. Lasso regression analysis was used to select optimal radiomics features for MVI in the training group. The optimal radiomics features of CT and MRI were selected to establish the prediction model. The predictive performance of the model was evaluated using the receiver operator characteristic curve (ROC), calibration curve, and decision curve analysis (DCA).

RESULTS

After univariate and multivariate analyses, it was found that tumor diameter was significantly different between the MVI positive and negative groups. After extracting 2153 imaging phenotyping features from the CT and MRI images of the 93 patients using Python, ten standardized coefficient non-zero imaging phenotyping features were finally determined by Lasso regression analysis in the CT and MRI images. A comprehensive predictive model with clinical variable and optimal radiomics features was established. The area under the curve (AUC) of the training group was 0.916 (95%CI: 0.843-1.000), sensitivity: 95.2%, specificity: 79.2%. In the validation group, the predictive model diagnosed MVI with AUC = 0.816 (95%CI: 0.642-0.990), sensitivity: 84.2%, and specificity: 75.0%.

CONCLUSION

The joint model that integrated the optimal radiomics features with clinical variables has good diagnostic performance for MVI of HCC and specific clinical applicability.

Prognostic and diagnostic value of [18F]FDG, 11C-acetate, and [68Ga]Ga-FAPI-04 PET/CT for hepatocellular carcinoma

OBJECTIVES

To assess the prognostic value of Fluorine 18-labeled fluorodeoxyglucose [18F]FDG, gallium 68-labeled fibroblast-activation protein inhibitor-04 [68Ga]Ga-FAPI-04, 11C-acetate in hepatocellular carcinoma (HCC) and evaluate the potential usefulness and advantages of different combinations for accurate diagnosis.

MATERIALS AND METHODS

Thirty-six patients with suspected hepatic masses were prospectively enrolled from May 2021 to September 2022 and underwent [18F]FDG, [68Ga]Ga-FAPI-04, and 11C-acetate PET/CT scans before surgery. PET/CT results and histopathologic examinations were independently interpreted by two radiologists and pathologists, respectively. Kaplan-Meier overall survival curves were calculated and the sensitivity among [18F]FDG, 11C-acetate, [68Ga]Ga-FAPI-04, and different combinations were compared.

RESULTS

Of the 36 included patients (mean age, 59 years ± 10 (standard deviation)), 29 were diagnosed with HCC, four with non-HCC malignant tumors, and three with benign tumors. Patients with HCC lesions negative for 11C-acetate or [68Ga]Ga-FAPI-04 exhibited poorer overall survival. Out of 36 patients, 44 HCC lesions were detected. The dual-tracer [68Ga]Ga-FAPI-04/11C-acetate exhibited the highest sensitivity (39 of 44 lesions (88.6%)) among all schemes. HCC lesions with higher histological grade and microvascular invasion (MVI) showed higher maximum standardized uptake value (SUVmax) and tumor-to-background ratio (TBR) of [18F]FDG, but no evidence of significant differences was found in [68Ga]Ga-FAPI-04 and 11C-acetate PET/CT. Higher expression of fibroblast activation protein (FAP) showed higher uptake of [68Ga]Ga-FAPI-04 and [18F]FDG.

CONCLUSION

[68Ga]Ga-FAPI-04 and 11C-acetate PET/CT exhibited good predictive value for HCC patients, with their combination showing the highest sensitivity for HCC detection, suggesting potential for improved diagnostic protocols.

KEY POINTS

Question What are the prognostic and diagnostic values of PET/CT tracers, including [18F]FDG, [68Ga]FAPI-04, and 11C-acetate? Findings Hepatocellular carcinoma, with differing findings across [18F]FDG, [68Ga]GaFAPI-04, and 11C-acetate PET/CT, showed varied prognoses; [68Ga]GaFAPI-04 and 11C-acetate combined offered the highest detection sensitivity. Clinical relevance Evaluating the prognostic value and diagnostic efficacy of different tracer combinations in patients with hepatocellular carcinoma helps to guide the optimal selection of tracers in clinical practice.

Predictive value of a constructed artificial neural network model for microvascular invasion in hepatocellular carcinoma: A retrospective study

BACKGROUND

Microvascular invasion (MVI) is a significant risk factor for recurrence and metastasis following hepatocellular carcinoma (HCC) surgery. Currently, there is a paucity of preoperative evaluation approaches for MVI.

AIM

To investigate the predictive value of texture features and radiological signs based on multiparametric magnetic resonance imaging in the non-invasive preoperative prediction of MVI in HCC.

METHODS

Clinical data from 97 HCC patients were retrospectively collected from January 2019 to July 2022 at our hospital. Patients were classified into two groups: MVI-positive (n = 57) and MVI-negative (n = 40), based on postoperative pathological results. The correlation between relevant radiological signs and MVI status was analyzed. MaZda4.6 software and the mutual information method were employed to identify the top 10 dominant texture features, which were combined with radiological signs to construct artificial neural network (ANN) models for MVI prediction. The predictive performance of the ANN models was evaluated using area under the curve, sensitivity, and specificity. ANN models with relatively high predictive performance were screened using the DeLong test, and the regression model of multilayer feedforward ANN with backpropagation and error backpropagation learning method was used to evaluate the models' stability.

RESULTS

The absence of a pseudocapsule, an incomplete pseudocapsule, and the presence of tumor blood vessels were identified as independent predictors of HCC MVI. The ANN model constructed using the dominant features of the combined group (pseudocapsule status + tumor blood vessels + arterial phase + venous phase) demonstrated the best predictive performance for MVI status and was found to be automated, highly operable, and very stable.

CONCLUSION

The ANN model constructed using the dominant features of the combined group can be recommended as a non-invasive method for preoperative prediction of HCC MVI status.

Predicting hepatocellular carcinoma outcomes and immune therapy response with ATP-dependent chromatin remodeling-related genes, highlighting MORF4L1 as a promising target

BACKGROUND

Hepatocellular carcinoma (HCC) continues to be a major cause of cancer-related death worldwide, primarily due to delays in diagnosis and resistance to existing treatments. Recent research has identified ATP-dependent chromatin remodeling-related genes (ACRRGs) as promising targets for therapeutic intervention across various types of cancer. This development offers potential new avenues for addressing the challenges in HCC management.

METHODS

This study integrated bioinformatics analyses and experimental approaches to explore the role of ACRRGs in HCC. We utilized data from The Cancer Genome Atlas (TCGA) and the Gene Expression Omnibus (GEO), applying machine learning algorithms to develop a prognostic model based on ACRRGs' expression. Experimental validation was conducted using quantitative real-time Polymerase Chain Reaction (qRT-PCR), Western blotting, and functional assays in HCC cell lines and xenograft models.

RESULTS

Our bioinformatics analysis identified four key ACRRGs-MORF4L1, HDAC1, VPS72, and RUVBL2-that serve as prognostic markers for HCC. The developed risk prediction model effectively distinguished between high-risk and low-risk patients, showing significant differences in survival outcomes and predicting responses to immunotherapy in HCC patients. Experimentally, MORF4L1 was demonstrated to enhance cancer stemness by activating the Hedgehog signaling pathway, as supported by both in vitro and in vivo assays.

CONCLUSION

ACRRGs, particularly MORF4L1, play crucial roles in modulating HCC progression, offering new insights into the molecular mechanisms driving HCC and potential therapeutic targets. Our findings advocate for the inclusion of chromatin remodeling dynamics in the strategic development of precision therapies for HCC.

Diagnostic Model for Proliferative HCC Using LI-RADS: Assessing Therapeutic Outcomes in Hepatectomy and TKI-ICI Combination

BACKGROUND

Proliferative hepatocellular carcinoma (HCC), aggressive with poor prognosis, and lacks reliable MRI diagnosis.

PURPOSE

To develop a diagnostic model for proliferative HCC using liver imaging reporting and data system (LI-RADS) and assess its prognostic value.

STUDY TYPE

Retrospective.

POPULATION

241 HCC patients underwent hepatectomy (90 proliferative HCCs: 151 nonproliferative HCCs), divided into the training (N = 167) and validation (N = 74) sets. 57 HCC patients received combination therapy with tyrosine kinase inhibitors (TKIs) and immune checkpoint inhibitors (ICIs).

FIELD STRENGTH/SEQUENCE

3.0 T, T1- and T2-weighted, diffusion-weighted, in- and out-phase, T1 high resolution isotropic volume excitation and dynamic gadoxetic acid-enhanced imaging.

ASSESSMENT

LI-RADS v2018 and other MRI features (intratumoral artery, substantial hypoenhancing component, hepatobiliary phase peritumoral hypointensity, and irregular tumor margin) were assessed. A diagnostic model for proliferative HCC was established, stratifying patients into high- and low-risk groups. Follow-up occurred every 3-6 months, and recurrence-free survival (RFS), progression-free survival (PFS) and overall survival (OS) in different groups were compared.

STATISTICAL TESTS

Fisher's test or chi-square test, t-test or Mann-Whitney test, logistic regression, Harrell's concordance index (C-index), Kaplan-Meier curves, and Cox proportional hazards. Significance level: P < 0.05.

RESULTS

The diagnostic model, incorporating corona enhancement, rim arterial phase hyperenhancement, infiltrative appearance, intratumoral artery, and substantial hypoenhancing component, achieved a C-index of 0.823 (training set) and 0.804 (validation set). Median follow-up was 32.5 months (interquartile range [IQR], 25.1 months) for postsurgery patients, and 16.8 months (IQR: 13.2 months) for combination-treated patients. 99 patients experienced recurrence, and 30 demonstrated tumor nonresponse. Differences were significant in RFS and OS rates between high-risk and low-risk groups post-surgery (40.3% vs. 65.8%, 62.3% vs. 90.1%, at 5 years). In combination-treated patients, PFS rates differed significantly (80.6% vs. 7.7% at 2 years).

DATA CONCLUSION

The MR-based model could pre-treatment identify proliferative HCC and assist in prognosis evaluation.

LEVEL OF EVIDENCE: 4

TECHNICAL EFFICACY

Stage 2.

Preoperative detection of hepatocellular carcinoma's microvascular invasion on CT-scan by machine learning and radiomics: A preliminary analysis

INTRODUCTION

Microvascular invasion (MVI) is the main risk factor for overall mortality and recurrence after surgery for hepatocellular carcinoma (HCC).The aim was to train machine-learning models to predict MVI on preoperative CT scan.

METHODS

3-phases CT scans were retrospectively collected among 4 Italian centers. DICOM files were manually segmented to detect the liver and the tumor(s). Radiomics features were extracted from the tumoral, peritumoral and healthy liver areas in each phase. Principal component analysis (PCA) was performed to reduce the dimensions of the dataset. Data were divided between training (70%) and test (30%) sets. Random-Forest (RF), fully connected MLP Artificial neural network (neuralnet) and extreme gradient boosting (XGB) models were fitted to predict MVI. Prediction accuracy was estimated in the test set.

RESULTS

Between 2008 and 2022, 218 preoperative CT scans were collected. At the histological specimen, 72(33.02%) patients had MVI. First and second order radiomics features were extracted, obtaining 672 variables. PCA selected 58 dimensions explaining >95% of the variance.In the test set, the XGB model obtained Accuracy = 68.7% (Sens: 38.1%, Spec: 83.7%, PPV: 53.3% and NPV: 73.4%). The neuralnet showed an Accuracy = 50% (Sens: 52.3%, Spec: 48.8%, PPV: 33.3%, NPV: 67.7%). RF was the best performer (Acc = 96.8%, 95%CI: 0.91-0.99, Sens: 95.2%, Spec: 97.6%, PPV: 95.2% and NPV: 97.6%).

CONCLUSION

Our model allowed a high prediction accuracy of the presence of MVI at the time of HCC diagnosis. This could lead to change the treatment allocation, the surgical extension and the follow-up strategy for those patients.

Prognostic value of enhanced cine cardiac MRI-based radiomics in dilated cardiomyopathy

BACKGROUND

Early precise identification of high-risk dilated cardiomyopathy (DCM) phenotype is essential for clinical decision-making and patient surveillance. The aim of the study was to assess the prognostic value of enhanced cine cardiac magnetic resonance (CMR)-based radiomics in DCM.

METHODS

We prospectively enrolled 401 (training set: 281; test set: 120) DCM patients. Radiomic features were extracted from enhanced cine images of entire left ventricular wall and selected by the least absolute shrinkage and selection operator. Different predictive models were built using logistic regression classifier to predict all-cause mortality and heart transplantation. Model performances were compared with the area under the receiver operating characteristic curves (AUCs). Kaplan-Meier curves, log-rank test, and Cox regression were used for survival analysis.

RESULTS

Endpoint events occurred in 65 patients over a median follow-up period of 25.4 months. 13 radiomic features were finally selected. The Rad_Combined model integrating clinical characteristics, CMR parameters and radiomics features achieved the best performance with an AUC of 0.836 and 0.835 in the training and test sets, respectively. High-risk groups with endpoint events defined by the Rad_Combined model had significantly shorter survival time than low-risk group in both the training [Hazard Ratio (HR) = 7.74, P < 0.001] and test sets (HR = 4.84, P < 0.001).

CONCLUSION

The Rad_Combined model might serve as an effective tool to help risk stratification and clinical decision-making for patients with DCM.

TRIAL REGISTRATION

Chinese Clinical Trial Registry, ChiCTR1800017058 by the ethics committee of West China hospital,Sichuan University.

Vessels encapsulating tumor clusters contribute to the intratumor heterogeneity of HCC on Gd-EOB-DTPA-enhanced MRI

BACKGROUND

Vessels encapsulating tumor clusters (VETC) pattern is tumor vasculature of HCC and is a predictor of prognosis and therapeutic efficacy. Recent radiological studies have demonstrated the predictability of VETC from preoperative images, but the mechanisms of image formation are not elucidated. This study aims to determine the relationship between VETC and intratumor heterogeneity in Gd-EOB-DTPA-enhanced magnetic resonance imaging (EOB-MRI) and to provide its pathological evidence.

METHODS

Radiologists visually classified preoperative arterial- and hepatobiliary-phase EOB-MRI images of 204 surgically resected HCCs into patterns based on heterogeneity and signal intensity; these classifications were validated using texture analysis. Single and multiplex immunohistochemistry for CD34, h-caldesmon, and OATP1B3 were performed to evaluate VETC, arterial vessel density (AVD), and OATP1B3 expression. Recurrence-free survival was assessed using the generalized Wilcoxon test. The contribution of clinicoradiological factors to the prediction of VETC was evaluated by random forest and least absolute shrinkage and selection operator regression.

RESULTS

VETC was frequently found in tumors with arterial-phase heterogeneous hyper-enhancement patterns and in tumors with hepatobiliary-phase heterogeneous hyperintense/isointense patterns (HBP-Hetero). AVD and OATP1B3 expression positively correlated with signal intensity in the arterial and hepatobiliary phases, respectively. Intratumor spatial analysis revealed that AVD and OATP1B3 expression were lower in VETC regions than in tumor regions without VETC. Patients with HBP-Hetero tumors had shorter recurrence-free survival. Machine learning models highlighted the importance of serum PIVKA-II, tumor size, and enhancement pattern of arterial and hepatobiliary phase for VETC prediction.

CONCLUSIONS

VETC is associated with local reductions of both AVD and OATP1B3 expression, likely contributing to heterogeneous enhancement patterns in EOB-MRI. Evaluation of the arterial and hepatobiliary phases of EOB-MRI would enhance the predictability of VETC.

Clinical-Radiomics Nomogram Model Based on CT Angiography for Prediction of Intracranial Aneurysm Rupture: A Multicenter Study

Objective

Risk estimation of intracranial aneurysm rupture is critical in determining treatment strategy. There is a scarcity of multicenter studies on the predictive power of clinical-radiomics models for aneurysm rupture. This study aims to develop a clinical-radiomics model and explore its additional value in the discrimination of aneurysm rupture.

Methods

A total of 516 aneurysms, including 273 (52.9%) with ruptured aneurysms, were retrospectively enrolled from four hospitals between January 2019 and August 2020. Relevant clinical features were collected, and radiomic characteristics associated with aneurysm were extracted. Subsequently, three models, including a clinical model, a radiomics model, and a clinical-radiomics model were constructed using multivariate logistic regression analysis to effectively classify aneurysm rupture. The performance of models was analyzed through operating characteristic curves, decision curve, and calibration curves analysis. Different models' comparison used DeLong tests. To offer an understandable and intuitive scoring system for assessing rupture risk, we developed a comprehensive nomogram based on the developed model.

Results

Three clinical risk factors and fourteen radiomics features were explored to establish three models. The area under the receiver operating curve (AUC) for the radiomics model was 0.775 (95% CI,0.719-0.830), 0.752 (95% CI,0.663-0.841), 0.747 (95% CI,0.658-0.835) in the training, internal and external test datasets, respectively. The AUC for clinical model was 0.802 (95% CI, 0.749-0.854), 0.736 (95% CI, 0.644-0.828), 0.789 (95% CI, 0.709-0.870) in these three sets, respectively. The clinical-radiomics model showed an AUC of 0.880 (95% CI,0.840-0.920), 0.807 (95% CI,0.728-0.887), 0.815 (95% CI,0.740-0.891) in three datasets respectively. Compared with the radiomics and clinical models, the clinical-radiomics model demonstrated better diagnostic performance (DeLong' test P < 0.05).

Conclusion

The clinical-radiomics model represents a promising approach for predicting rupture of intracranial aneurysms.

Evaluating microvascular invasion in hepatitis B virus-related hepatocellular carcinoma based on contrast-enhanced computed tomography radiomics and clinicoradiological factors

BACKGROUND

Microvascular invasion (MVI) is a significant indicator of the aggressive behavior of hepatocellular carcinoma (HCC). Expanding the surgical resection margin and performing anatomical liver resection may improve outcomes in patients with MVI. However, no reliable preoperative method currently exists to predict MVI status or to identify patients at high-risk group (M2).

AIM

To develop and validate models based on contrast-enhanced computed tomography (CECT) radiomics and clinicoradiological factors to predict MVI and identify M2 among patients with hepatitis B virus-related HCC (HBV-HCC). The ultimate goal of the study was to guide surgical decision-making.

METHODS

A total of 270 patients who underwent surgical resection were retrospectively analyzed. The cohort was divided into a training dataset (189 patients) and a validation dataset (81) with a 7:3 ratio. Radiomics features were selected using intra-class correlation coefficient analysis, Pearson or Spearman's correlation analysis, and the least absolute shrinkage and selection operator algorithm, leading to the construction of radscores from CECT images. Univariate and multivariate analyses identified significant clinicoradiological factors and radscores associated with MVI and M2, which were subsequently incorporated into predictive models. The models' performance was evaluated using calibration, discrimination, and clinical utility analysis.

RESULTS

Independent risk factors for MVI included non-smooth tumor margins, absence of a peritumoral hypointensity ring, and a high radscore based on delayed-phase CECT images. The MVI prediction model incorporating these factors achieved an area under the curve (AUC) of 0.841 in the training dataset and 0.768 in the validation dataset. The M2 prediction model, which integrated the radscore from the 5 mm peritumoral area in the CECT arterial phase, α-fetoprotein level, enhancing capsule, and aspartate aminotransferase level achieved an AUC of 0.865 in the training dataset and 0.798 in the validation dataset. Calibration and decision curve analyses confirmed the models' good fit and clinical utility.

CONCLUSION

Multivariable models were constructed by combining clinicoradiological risk factors and radscores to preoperatively predict MVI and identify M2 among patients with HBV-HCC. Further studies are needed to evaluate the practical application of these models in clinical settings.

Global trends in machine learning applied to clinical research in liver cancer: Bibliometric and visualization analysis (2001-2024)

This study explores the intersection of liver cancer and machine learning through bibliometric analysis. The aim is to identify highly cited papers in the field and examine the current research landscape, highlighting emerging trends and key areas of focus in liver cancer and machine learning. By analyzing citation patterns, this study sheds light on the evolving role of machine learning in liver cancer research and its potential for future advancements.

Multimodal multiphasic preoperative image-based deep-learning predicts HCC outcomes after curative surgery

BACKGROUND AND AIMS

HCC recurrence frequently occurs after curative surgery. Histological microvascular invasion (MVI) predicts recurrence but cannot provide preoperative prognostication, whereas clinical prediction scores have variable performances.

APPROACH AND RESULTS

Recurr-NET, a multimodal multiphasic residual-network random survival forest deep-learning model incorporating preoperative CT and clinical parameters, was developed to predict HCC recurrence. Preoperative triphasic CT scans were retrieved from patients with resected histology-confirmed HCC from 4 centers in Hong Kong (internal cohort). The internal cohort was randomly divided in an 8:2 ratio into training and internal validation. External testing was performed in an independent cohort from Taiwan.Among 1231 patients (age 62.4y, 83.1% male, 86.8% viral hepatitis, and median follow-up 65.1mo), cumulative HCC recurrence rates at years 2 and 5 were 41.8% and 56.4%, respectively. Recurr-NET achieved excellent accuracy in predicting recurrence from years 1 to 5 (internal cohort AUROC 0.770-0.857; external AUROC 0.758-0.798), significantly outperforming MVI (internal AUROC 0.518-0.590; external AUROC 0.557-0.615) and multiple clinical risk scores (ERASL-PRE, ERASL-POST, DFT, and Shim scores) (internal AUROC 0.523-0.587, external AUROC: 0.524-0.620), respectively (all p < 0.001). Recurr-NET was superior to MVI in stratifying recurrence risks at year 2 (internal: 72.5% vs. 50.0% in MVI; external: 65.3% vs. 46.6% in MVI) and year 5 (internal: 86.4% vs. 62.5% in MVI; external: 81.4% vs. 63.8% in MVI) (all p < 0.001). Recurr-NET was also superior to MVI in stratifying liver-related and all-cause mortality (all p < 0.001). The performance of Recurr-NET remained robust in subgroup analyses.

CONCLUSIONS

Recurr-NET accurately predicted HCC recurrence, outperforming MVI and clinical prediction scores, highlighting its potential in preoperative prognostication.

Three-Dimensional Multifrequency MR Elastography for Microvascular Invasion and Prognosis Assessment in Hepatocellular Carcinoma

BACKGROUND

Pretreatment identification of microvascular invasion (MVI) in hepatocellular carcinoma (HCC) is important when selecting treatment strategies.

PURPOSE

To improve models for predicting MVI and recurrence-free survival (RFS) by developing nomograms containing three-dimensional (3D) MR elastography (MRE).

STUDY TYPE

Prospective.

POPULATION

188 patients with HCC, divided into a training cohort (n = 150) and a validation cohort (n = 38). In the training cohort, 106/150 patients completed a 2-year follow-up.

FIELD STRENGTH/SEQUENCE

1.5T 3D multifrequency MRE with a single-shot spin-echo echo planar imaging sequence, and 3.0T multiparametric MRI (mp-MRI), consisting of diffusion-weighted echo planar imaging, T2-weighted fast spin echo, in-phase out-of-phase T1-weighted fast spoiled gradient-recalled dual-echo and dynamic contrast-enhanced gradient echo sequences.

ASSESSMENT

Multivariable analysis was used to identify the independent predictors for MVI and RFS. Nomograms were constructed for visualization. Models for predicting MVI and RFS were built using mp-MRI parameters and a combination of mp-MRI and 3D MRE predictors.

STATISTICAL TESTS

Student's t-test, Mann-Whitney U test, chi-squared or Fisher's exact tests, multivariable analysis, area under the receiver operating characteristic curve (AUC), DeLong test, Kaplan-Meier analysis and log rank tests. P < 0.05 was considered significant.

RESULTS

Tumor c and liver c were independent predictors of MVI and RFS, respectively. Adding tumor c significantly improved the diagnostic performance of mp-MRI (AUC increased from 0.70 to 0.87) for MVI detection. Of the 106 patients in the training cohort who completed the 2-year follow up, 34 experienced recurrence. RFS was shorter for patients with MVI-positive histology than MVI-negative histology (27.1 months vs. >40 months). The MVI predicted by the 3D MRE model yielded similar results (26.9 months vs. >40 months). The MVI and RFS nomograms of the histologic-MVI and model-predicted MVI-positive showed good predictive performance.

DATA CONCLUSION

Biomechanical properties of 3D MRE were biomarkers for MVI and RFS. MVI and RFS nomograms were established.

LEVEL OF EVIDENCE

2 TECHNICAL EFFICACY: Stage 2.

Correlation between 18 F-FDG PET/CT metabolic parameters and microvascular invasion before liver transplantation in patients with hepatocellular carcinoma

BACKGROUND

Microvascular infiltration (MVI) before liver transplantation (LT) in patients with hepatocellular carcinoma (HCC) is associated with postoperative tumor recurrence and survival. MVI is mainly assessed by pathological analysis of tissue samples, which is invasive and heterogeneous. PET/computed tomography (PET/CT) with 18 F-labeled fluorodeoxyglucose ( 18 F-FDG) as a tracer has been widely used in the examination of malignant tumors. This study investigated the association between 18 F-FDG PET/CT metabolic parameters and MVI before LT in HCC patients.

METHODS

About 124 HCC patients who had 18 F-FDG PET/CT examination before LT were included. The patients' clinicopathological features and 18 F-FDG PET/CT metabolic parameters were recorded. Correlations between clinicopathological features, 18 F-FDG PET/CT metabolic parameters, and MVI were analyzed. ROC curve was used to determine the optimal diagnostic cutoff value, area under the curve (AUC), sensitivity, and specificity for predictors of MVI.

RESULT

In total 72 (58.06%) patients were detected with MVI among the 124 HCC patients. Univariate analysis showed that tumor size ( P  = 0.001), T stage ( P  < 0.001), maximum standardized uptake value (SUV max ) ( P  < 0.001), minimum standardized uptake value (SUV min ) ( P  = 0.031), mean standardized uptake value (SUV mean ) ( P  = 0.001), peak standardized uptake value (SUV peak ) ( P  = 0.001), tumor-to-liver ratio (SUV ratio ) ( P  = 0.010), total lesion glycolysis (TLG) ( P  = 0.006), metabolic tumor volume (MTV) ( P  = 0.011) and MVI were significantly different. Multivariate logistic regression showed that tumor size ( P  = 0.018), T stage ( P  = 0.017), TLG ( P  = 0.023), and MTV ( P  = 0.015) were independent predictors of MVI. In the receiver operating characteristic curve, TLG predicted MVI with an AUC value of 0.645. MTV predicted MVI with an AUC value of 0.635. Patients with tumor size ≥5 cm, T3-4, TLG > 400.67, and MTV > 80.58 had a higher incidence of MVI.

CONCLUSION

18 F-FDG PET/CT metabolic parameters correlate with MVI and may be used as a noninvasive technique to predict MVI before LT in HCC patients.

18F-FDG PET/CT-based habitat radiomics combining stacking ensemble learning for predicting prognosis in hepatocellular carcinoma: a multi-center study

BACKGROUND

This study aims to develop habitat radiomic models to predict overall survival (OS) for hepatocellular carcinoma (HCC), based on the characterization of the intratumoral heterogeneity reflected in 18F-FDG PET/CT images.

METHODS

A total of 137 HCC patients from two institutions were retrospectively included. First, intratumoral habitats were achieved by a two-step unsupervised clustering process based on k-means clustering. Second, a total of 4032 radiomic features were extracted based on each habitat, including 2016 PET-based and 2016 CT-based radiomic features. Then, after feature selection, the stacking ensemble learning approach which combined six machine learning classifiers as the first-level learners with Cox proportional hazards regression as the second-level learner, was employed to build multiple radiomic models. Finally, the optimal model was selected based on the calculation of the C-index, and a combined model integrating with a clinical model was also constructed to identify the potentially complementary effect.

RESULTS

Three spatially distinct habitats were identified in the two cohorts. Among a total of 30 stacking ensemble learning models established based on different combinations of 5 types of segmented volumes of interest (VOIs) with 6 types of classifiers, the MLP-Cox-habitat-2 model was selected as the optimal radiomic model with a C-index of 0.702 in the external validation cohort. Furthermore, the combined model integrating the optimal radiomic model with the clinical model achieved an improved C-index of 0.747. Consistently, the combined model outperformed the other models for OS prediction, with a time-dependent AUC of 0.835, 0.828, and 0.800 in the 1-year, 2-year, and 3-year OS, respectively.

CONCLUSION

18F-FDG PET/CT-based habitat radiomics outperformed traditional radiomics in OS prediction for HCC, with a further improved predictive power by integrating with the clinical model. The optimal combined habitat model was potentially promising in guiding individualized treatment for HCC.

TRIAL REGISTRATION

This study was a retrospective study, so it was free from registration.

Bridging the Gap Between Imaging and Molecular Characterization: Current Understanding of Radiomics and Radiogenomics in Hepatocellular Carcinoma

Hepatocellular carcinoma (HCC) is the sixth most common malignancy worldwide and the third leading cause of cancer-related deaths. Imaging plays a crucial role in the screening, diagnosis, and monitoring of HCC; however, the potential mechanism regarding phenotypes or molecular subtyping remains underexplored. Radiomics significantly expands the selection of features available by extracting quantitative features from imaging data. Radiogenomics bridges the gap between imaging and genetic/transcriptomic information by associating imaging features with critical genes and pathways, thereby providing biological annotations to these features. Despite challenges in interpreting these connections, assessing their universality, and considering the diversity in HCC etiology and genetic information across different populations, radiomics and radiogenomics offer new perspectives for precision treatment in HCC. This article provides an up-to-date summary of the advancements in radiomics and radiogenomics throughout the HCC care continuum, focusing on the clinical applications, advantages, and limitations of current techniques and offering prospects. Future research should aim to overcome these challenges to improve the prognosis of HCC patients and leverage imaging information for patient benefit.

Preoperative Noninvasive Prediction of Recurrence-Free Survival in Hepatocellular Carcinoma Using CT-Based Radiomics Model

Purpose

This study aims to explore the value of radiomics combined with clinical parameters in predicting recurrence-free survival (RFS) after the resection of hepatocellular carcinoma (HCC).

Patients and Methods

In this retrospective study, a total of 322 patients with HCC who underwent contrast-enhanced computed tomography (CT) and radical surgical resection were enrolled and randomly divided into a training group (n = 223) and a validation group (n = 97). In the training group, Univariate and multivariate Cox regression analyses were employed to obtain clinical variables related to RFS for constructing the clinical model. The least absolute shrinkage and selection operator (LASSO) and multivariate Cox regression analyses were employed to construct the radiomics model, and the clinical-radiomics model was further constructed. Model prediction performance was subsequently assessed by the area under the time-dependent receiver operating characteristic curve (AUC) and calibration curve. Additionally, Kaplan-Meier analysis was used to evaluate the model's value in predicting RFS. Correlations between radiomics features and pathological parameters were analyzed.

Results

The clinical-radiomics model predicted RFS at 1, 2, and 3 years more accurately than the clinical or radiomics model alone (training group, AUC = 0.834, 0.765 and 0.831, respectively; validation group, AUC = 0.715, 0.710 and 0.793, respectively). The predicted high-risk subgroup based on the clinical-radiomics nomogram had shorter RFS than predicted low-risk subgroup in data sets, enabling risk stratification of various clinical subgroups. Correlation analysis revealed that the rad-score was positively related to microvascular invasion (MVI) and Edmondson-Steiner grade.

Conclusion

The clinical-radiomics model effectively predicts RFS in HCC patients and identifies high-risk individuals for recurrence.