Ultrasound-based radiomics score: a potential biomarker for the prediction of microvascular invasion in hepatocellular carcinoma

Construction of a nomogram combining CEUS and MRI imaging for preoperative diagnosis of microvascular invasion in hepatocellular carcinoma

Purpose

To use Sonazoid contrast-enhanced ultrasound (S-CEUS) and Gadolinium-Ethoxybenzyl-Diethylenetriamine Penta-Acetic Acid magnetic-resonance imaging (EOB-MRI), exploring a non-invasive preoperative diagnostic strategy for microvascular invasion (MVI) of hepatocellular carcinoma (HCC).

Methods

111 newly developed HCC cases were retrospectively collected. Both S-CEUS and EOB-MRI examinations were performed within one month of hepatectomy. The following indicators were investigated: size; vascularity in three phases of S-CEUS; margin, signal intensity, and peritumoral wedge shape in EOB-MRI; tumoral homogeneity, presence and integrity of the tumoral capsule in S-CEUS or EOB-MRI; presence of branching enhancement in S-CEUS; baseline clinical and serological data. The least absolute shrinkage and selection operator regression and multivariate logistic regression analysis were applied to optimize feature selection for the model. A nomogram for MVI was developed and verified by bootstrap resampling.

Results

Of the 16 variables we included, wedge and margin in HBP of EOB-MRI, capsule integrity in AP or HBP/PVP images of EOB-MRI/S-CEUS, and branching enhancement in AP of S-CEUS were identified as independent risk factors for MVI and incorporated into construction of the nomogram. The nomogram achieved an excellent diagnostic efficiency with an area under the curve of 0.8434 for full data training set and 0.7925 for bootstrapping validation set for 500 repetitions. In evaluating the nomogram, Hosmer-Lemeshow test for training set exhibited a good model fit with P > 0.05. Decision curve analysis of nomogram model yielded excellent clinical net benefit with a wide range (5-80 % and 85-94 %) of risk threshold.

Conclusions

The MVI Nomogram established in this study may provide a strategy for optimizing the preoperative diagnosis of MVI, which in turn may improve the treatment and prognosis of MVI-related HCC.

A dual-center study: can ultrasound radiomics differentiate type I and type II epithelial ovarian cancer patients with normal CA125 levels?

OBJECTIVE

CA125 is recommended by many countries as the primary screening test for ovarian cancer. But there are patients with ovarian cancer having normal CA125. We hope to identify the types of EOC with normal CA125 levels better by building a refined model based on the ultrasound radiomics, thus providing precise medical treatment for patients.

METHODS

We included 58 patients with EOC with normal CA125 from 2 centres, who were confirmed by preoperative ultrasound and pathology. We extracted 1130 radiomics features based on the tumour's region of interest from the most typical ultrasound image of each patient. We selected radiomics and clinical features by LASSO and logistic regression to construct Rad-score and clinical models, respectively. Receiver operating characteristic curves judged their test efficacy. On the basis of the combined model, we developed a nomogram.

RESULTS

Area under the curves (AUCs) of 0.93 and 0.83 were achieved in both the training and test groups for the combined model. There were similar AUCs between the Rad-score and clinical models of 0.82 and 0.80, respectively. By analysing the calibration curves, it was determined that the nomogram matched actual observations in the training cohort.

CONCLUSION

Ultrasound radiomics can differentiate type I and type II EOC with normal CA125 levels.

ADVANCES IN KNOWLEDGE

This study is the first to focus on EOC cases with normal level of CA125. The subset of patients constituting 20% of the disease population may require more refined radiomics models.

The Enhanced Role of Eosinophils in Radiomics-Based Diagnosis of Microvascular Invasion and Its Association with the Immune Microenvironment in Hepatocellular Carcinoma

Objective

To investigate the role of eosinophil counts (EC) in microvascular invasion (MVI) for enhancing the radiomics based diagnostic model. Additionally, its correlation with early recurrence and tumor immune microenvironment was explored.

Methods

Propensity score matching was employed to evaluate on 462 cases whether EC was an independent risk factor for MVI. Subgroup analyses examined EC's effect on MVI across varying hypersplenism degrees. Univariate-multivariate logistic regression identified MVI's independent factors to develop a diagnostic model. Univariate-multivariate COX regression determined early recurrence factors. Co-detection by indexing (CODEX) constructed the immune score (IS), and Spearman correlation analyzed its association with peripheral immunity.

Results

EC was an independent risk factor for MVI (p=0.038, OR=1.304 (95% CI: 1.014-1.677)), and its effect on MVI disappeared with the severity of hypersplenism. The diagnostic model with EC was significantly improved (AUC=0.787 (95% CI: 0.737-0.836) vs AUC=0.748(95% CI: 0.694-0.802, p=0.005)). MVI was an independent risk factor for early recurrence (p<0.001, HR = 2.254 (95% CI: 1.557-3.263)). IS was negatively correlated with lymphocyte counts (R=-0.311, p=0.022), and positively correlated with EC (R=0.301, p=0.027) and RS (R = 0.315, p = 0.018).

Conclusion

EC was an independent risk factor for MVI and was related to the tumor immune microenvironment. EC should be included in the diagnosis of MVI to improve diagnostic efficiency.

Ultrasound-based radiomics nomogram for predicting HER2-low expression breast cancer

Purpose

Accurate preoperative identification of Human epidermal growth factor receptor 2 (HER2) low expression breast cancer (BC) is critical for clinical decision-making. Our aim was to use machine learning methods to develop and validate an ultrasound-based radiomics nomogram for predicting HER2-low expression in BC.

Methods

In this retrospective study, 222 patients (108 HER2-0 expression and 114 HER2-low expression) with BC were included. The enrolled patients were randomly divided into a training cohort and a test cohort with a ratio of 8:2. The tumor region of interest was manually delineated from ultrasound image, and radiomics features were subsequently extracted. The features underwent dimension reduction using the least absolute shrinkage and selection operator (LASSO) algorithm, and rad-score were calculated. Five machine learning algorithms were applied for training, and the algorithm demonstrating the best performance was selected to construct a radiomics (USR) model. Clinical risk factors were integrated with rad-score to construct the prediction model, and a nomogram was plotted. The performance of the nomogram was assessed using receiver operating characteristic curve and decision curve analysis.

Results

A total of 480 radiomics features were extracted, out of which 11 were screened out. The majority of the extracted features were wavelet features. Subsequently, the USR model was established, and rad-scores were computed. The nomogram, incorporating rad-score, tumor shape, border, and microcalcification, achieved the best performance in both the training cohort (AUC 0.89; 95%CI 0.836-0.936) and the test cohort (AUC 0.84; 95%CI 0.722-0.958), outperforming both the USR model and clinical model. The calibration curves showed satisfactory consistency, and DCA confirmed the clinical utility of the nomogram.

Conclusion

The nomogram model based on ultrasound radiomics exhibited high prediction value for HER2-low BC.

Dual-energy computed tomography iodine quantification combined with laboratory data for predicting microvascular invasion in hepatocellular carcinoma: a two-centre study

OBJECTIVES

Microvascular invasion (MVI) is a recognized biomarker associated with poorer prognosis in patients with hepatocellular carcinoma. Dual-energy computed tomography (DECT) is a highly sensitive technique that can determine the iodine concentration (IC) in tumour and provide an indirect evaluation of internal microcirculatory perfusion. This study aimed to assess whether the combination of DECT with laboratory data can improve preoperative MVI prediction.

METHODS

This retrospective study enrolled 119 patients who underwent DECT liver angiography at 2 medical centres preoperatively. To compare DECT parameters and laboratory findings between MVI-negative and MVI-positive groups, Mann-Whitney U test was used. Additionally, principal component analysis (PCA) was conducted to determine fundamental components. Mann-Whitney U test was applied to determine whether the principal component (PC) scores varied across MVI groups. Finally, a general linear classifier was used to assess the classification ability of each PC score.

RESULTS

Significant differences were noted (P < .05) in alpha-fetoprotein (AFP) level, normalized arterial phase IC, and normalized portal phase IC between the MVI groups in the primary and validation datasets. The PC1-PC4 accounted for 67.9% of the variance in the primary dataset, with loadings of 24.1%, 16%, 15.4%, and 12.4%, respectively. In both primary and validation datasets, PC3 and PC4 were significantly different across MVI groups, with area under the curve values of 0.8410 and 0.8373, respectively.

CONCLUSIONS

The recombination of DECT IC and laboratory features based on varying factor loadings can well predict MVI preoperatively.

ADVANCES IN KNOWLEDGE

Utilizing PCA, the amalgamation of DECT IC and laboratory features, considering diverse factor loadings, showed substantial promise in accurately classifying MVI. There have been limited endeavours to establish such a combination, offering a novel paradigm for comprehending data in related research endeavours.

Research progresses of imaging studies on preoperative prediction of microvascular invasion of hepatocellular carcinoma

Hepatocellular carcinoma (HCC) is the predominant form of primary liver cancer, accounting for approximately 90% of liver cancer cases. It currently ranks as the fifth most prevalent cancer worldwide and represents the third leading cause of cancer-related mortality. As a malignant disease with surgical resection and ablative therapy being the sole curative options available, it is disheartening that most HCC patients who undergo liver resection experience relapse within five years. Microvascular invasion (MVI), defined as the presence of micrometastatic HCC emboli within liver vessels, serves as an important histopathological feature and indicative factor for both disease-free survival and overall survival in HCC patients. Therefore, achieving accurate preoperative noninvasive prediction of MVI holds vital significance in selecting appropriate clinical treatments and improving patient prognosis. Currently, there are no universally recognized criteria for preoperative diagnosis of MVI in clinical practice. Consequently, extensive research efforts have been directed towards preoperative imaging prediction of MVI to address this problem and the relative research progresses were reviewed in this article to summarize its current limitations and future research prospects.

Multimodal Machine Learning-Based Ductal Carcinoma in situ Prediction from Breast Fibromatosis

Objective

To develop a clinical-radiomics model using a multimodal machine learning method for distinguishing ductal carcinoma in situ (DCIS) from breast fibromatosis.

Methods

The clinical factors, ultrasound features, and related ultrasound images of 306 patients (198 DCIS patients) were retrospectively collected. Patients in the development and validation cohort were 184 and 122, respectively. The independent clinical and ultrasound factors identified by the multivariable logistic regression analysis were used for the clinical-ultrasound model construction. Then, the region of interest of breast lesions was delineated and radiomics features were extracted. Six machine learning algorithms were trained to develop a radiomics model. The algorithm with higher and more stable prediction ability was chosen to convert the output of the results into the Radscore. Further, the independent clinical predictors and Radscore were enrolled into the logistic regression analysis to generate a combined clinical-radiomics model. The receiver operating characteristic curve analysis, DeLong test, and decision curve analysis were adopted to compare the prediction ability and clinical efficacy of three different models.

Results

Among the six classifiers, logistic regression model was selected as the final radiomics model. Besides, the combined clinical-radiomics model exhibited a superior ability in distinguishing DCIS from breast fibromatosis to the clinical-ultrasound model and the radiomics model.

Conclusion

The combined model by integrating clinical-ultrasound factors and radiomics features performed well in predicting DCIS, which might promote prompt interventions to improve the early diagnosis and prognosis of the patients.

Predicting hepatocellular carcinoma: A new non-invasive model based on shear wave elastography

BACKGROUND

Integrating conventional ultrasound features with 2D shear wave elastography (2D-SWE) can potentially enhance preoperative hepatocellular carcinoma (HCC) predictions.

AIM

To develop a 2D-SWE-based predictive model for preoperative identification of HCC.

METHODS

A retrospective analysis of 884 patients who underwent liver resection and pathology evaluation from February 2021 to August 2023 was conducted at the Oriental Hepatobiliary Surgery Hospital. The patients were divided into the modeling group (n = 720) and the control group (n = 164). The study included conventional ultrasound, 2D-SWE, and preoperative laboratory tests. Multiple logistic regression was used to identify independent predictive factors for malignant liver lesions, which were then depicted as nomograms.

RESULTS

In the modeling group analysis, maximal elasticity (Emax) of tumors and their peripheries, platelet count, cirrhosis, and blood flow were independent risk indicators for malignancies. These factors yielded an area under the curve of 0.77 (95% confidence interval: 0.73-0.81) with 84% sensitivity and 61% specificity. The model demonstrated good calibration in both the construction and validation cohorts, as shown by the calibration graph and Hosmer-Lemeshow test (P = 0.683 and P = 0.658, respectively). Additionally, the mean elasticity (Emean) of the tumor periphery was identified as a risk factor for microvascular invasion (MVI) in malignant liver tumors (P = 0.003). Patients receiving antiviral treatment differed significantly in platelet count (P = 0.002), Emax of tumors (P = 0.033), Emean of tumors (P = 0.042), Emax at tumor periphery (P < 0.001), and Emean at tumor periphery (P = 0.003).

CONCLUSION

2D-SWE's hardness value serves as a valuable marker for enhancing the preoperative diagnosis of malignant liver lesions, correlating significantly with MVI and antiviral treatment efficacy.

Prediction model of early recurrence of multimodal hepatocellular carcinoma with tensor fusion

Objective.In oncology, clinical decision-making relies on a multitude of data modalities, including histopathological, radiological, and clinical factors. Despite the emergence of computer-aided multimodal decision-making systems for predicting hepatocellular carcinoma (HCC) recurrence post-hepatectomy, existing models often employ simplistic feature-level concatenation, leading to redundancy and suboptimal performance. Moreover, these models frequently lack effective integration with clinically relevant data and encounter challenges in integrating diverse scales and dimensions, as well as incorporating the liver background, which holds clinical significance but has been previously overlooked.Approach.To address these limitations, we propose two approaches. Firstly, we introduce the tensor fusion method to our model, which offers distinct advantages in handling multi-scale and multi-dimensional data fusion, potentially enhancing overall performance. Secondly, we pioneer the consideration of the liver background's impact, integrating it into the feature extraction process using a deep learning segmentation-based algorithm. This innovative inclusion aligns the model more closely with real-world clinical scenarios, as the liver background may contain crucial information related to postoperative recurrence.Main results.We collected radiomics (MRI) and histopathological images from 176 cases diagnosed by experienced clinicians across two independent centers. Our proposed network underwent training and 5-fold cross-validation on this dataset before validation on an external test dataset comprising 40 cases. Ultimately, our model demonstrated outstanding performance in predicting early recurrence of HCC postoperatively, achieving an AUC of 0.883.Significance.These findings signify significant progress in addressing challenges related to multimodal data fusion and hold promise for more accurate clinical outcome predictions. In this study, we exploited global 3D liver background into modelling which is crucial to to the prognosis assessment and analyzed the whole liver background in addition to the tumor region. Both MRI images and histopathological images of HCC were fused at high-dimensional feature space using tensor techniques to solve cross-scale data integration issue.

Ultrasonography of Hepatocellular Carcinoma: From Diagnosis to Prognosis

Hepatocellular carcinoma (HCC) is a prominent contributor to cancer-related mortality worldwide. Early detection and diagnosis of liver cancer can significantly improve its prognosis and patient survival. Ultrasound technology, serving has undergone substantial advances as the primary method of HCC surveillance and has broadened its scope in recent years for effective management of HCC. This article is a comprehensive overview of ultrasound technology in the treatment of HCC, encompassing early detection, diagnosis, staging, treatment evaluation, and prognostic assessment. In addition, the authors summarized the application of contrast-enhanced ultrasound in the diagnosis of HCC and assessment of prognosis. Finally, the authors discussed further directions in this field by emphasizing overcoming existing obstacles and integrating cutting-edge technologies.

Predicting Ki-67 expression in hepatocellular carcinoma: nomogram based on clinical factors and contrast-enhanced ultrasound radiomics signatures

PURPOSE

To develop a contrast-enhanced ultrasound (CEUS) clinic-radiomics nomogram for individualized assessment of Ki-67 expression in hepatocellular carcinoma (HCC).

METHODS

A retrospective cohort comprising 310 HCC individuals who underwent preoperative CEUS (using SonoVue) at three different centers was partitioned into a training set, a validation set, and an external test set. Radiomics signatures indicating the phenotypes of the Ki-67 were extracted from multiphase CEUS images. The radiomics score (Rad-score) was calculated accordingly after feature selection and the radiomics model was constructed. A clinic-radiomics nomogram was established utilizing multiphase CEUS Rad-score and clinical risk factors. A clinical model only incorporated clinical factors was also developed for comparison. Regarding clinical utility, calibration, and discrimination, the predictive efficiency of the clinic-radiomics nomogram was evaluated.

RESULTS

Seven radiomics signatures from multiphase CEUS images were selected to calculate the Rad-score. The clinic-radiomics nomogram, comprising the Rad-score and clinical risk factors, indicated a good calibration and demonstrated a better discriminatory capacity compared to the clinical model (AUCs: 0.870 vs 0.797, 0.872 vs 0.755, 0.856 vs 0.749 in the training, validation, and external test set, respectively) and the radiomics model (AUCs: 0.870 vs 0.752, 0.872 vs 0.733, 0.856 vs 0.729 in the training, validation, and external test set, respectively). Furthermore, both the clinical impact curve and the decision curve analysis displayed good clinical application of the nomogram.

CONCLUSION

The clinic-radiomics nomogram constructed from multiphase CEUS images and clinical risk parameters can distinguish Ki-67 expression in HCC patients and offer useful insights to guide subsequent personalized treatment.

Exploring the diagnostic value of ultrasound radiomics for neonatal respiratory distress syndrome

BACKGROUND

Neonatal respiratory distress syndrome (NRDS) is a prevalent cause of respiratory failure and death among newborns, and prompt diagnosis is imperative. Historically, diagnosis of NRDS relied mostly on typical clinical manifestations, chest X-rays, and CT scans. However, recently, ultrasound has emerged as a valuable and preferred tool for aiding NRDS diagnosis. Nevertheless, evaluating lung ultrasound imagery necessitates rigorous training and may be subject to operator-dependent bias, limiting its widespread use. As a result, it is essential to investigate a new, reliable, and operator-independent diagnostic approach that does not require subjective factors or operator expertise. This article aims to explore the diagnostic potential of ultrasound-based radiomics in differentiating NRDS from other non-NRDS lung disease.

METHODS

A total of 150 neonatal lung disease cases were consecutively collected from the department of neonatal intensive care unit of the Quanzhou Maternity and Children's Hospital, Fujian Province, from September 2021 to October 2022. Of these patients, 60 were diagnosed with NRDS, whereas 30 were diagnosed with neonatal pneumonia, meconium aspiration syndrome (MAS), and transient tachypnea (TTN). Two ultrasound images with characteristic manifestations of each lung disease were acquired and divided into training (n = 120) and validation cohorts (n = 30) based on the examination date using an 8:2 ratio. The imaging texture features were extracted using PyRadiomics and, after the screening, machine learning models such as random forest (RF), logistic regression (LR), K-nearest neighbors (KNN), support vector machine (SVM), and multilayer perceptron (MLP) were developed to construct an imaging-based diagnostic model. The diagnostic efficacy of each model was analyzed. Lastly, we randomly selected 282 lung ultrasound images and evaluated the diagnostic efficacy disparities between the optimal model and doctors across differing levels of expertise.

RESULTS

Twenty-two imaging-based features with the highest weights were selected to construct a predictive model for neonatal respiratory distress syndrome. All models exhibited favorable diagnostic performances. Analysis of the Youden index demonstrated that the RF model had the highest score in both the training (0.99) and validation (0.90) cohorts. Additionally, the calibration curve indicated that the RF model had the best calibration (P = 0.98). When compared to the diagnostic performance of experienced and junior physicians, the RF model had an area under the curve (AUC) of 0.99; however, the values for experienced and junior physicians were 0.98 and 0.85, respectively. The difference in diagnostic efficacy between the RF model and experienced physicians was not statistically significant (P = 0.24), whereas that between the RF model and junior physicians was statistically significant (P < 0.0001).

CONCLUSION

The RF model exhibited excellent diagnostic performance in the analysis of texture features based on ultrasound radiomics for diagnosing NRDS.

A dual-targeted Gd-based contrast agent for magnetic resonance imaging in tumor diagnosis

Enhanced magnetic resonance imaging (MRI) has important clinical value in the diagnosis of tumors. Much effort has been made to improve the relaxivity and specificity of contrast agents (CAs) in tumor diagnosis over the past few decades. However, there is still a lack of CAs which not only enhance the signal intensity of tumors rather than surrounding tissues in MRI but also maintain a high signal intensity prolonged for a long time. Herein, we synthesized a dual-targeted CA, RGD-(DOTA-Gd)-TPP (RDP), in which RGD is used to target the αvβ3 integrin receptor overexpressed in tumor cells and TPP is used to bind to a mitochondrion further. The structure of RDP was characterized and its properties, such as relaxivity and biosafety, were measured and in vitro and in vivo MRI assays were carried out. It has been proven that RDP has higher relaxivity of aqueous solution than Magnevist used in clinics. Moreover, RDP achieved higher signal intensity and a longer signal duration in tumor imaging. Therefore, RDP can be applied as the potential dual-targeted MRI CA for clinical tumor diagnosis.

Identifying Coronary Artery Calcification Using Chest X-ray Radiographs and Machine Learning: The Role of the Radiomics Score

PURPOSE

To evaluate the ability of radiomics score (RS)-based machine learning to identify moderate to severe coronary artery calcium (CAC) on chest x-ray radiographs (CXR).

MATERIALS AND METHODS

We included 559 patients who underwent a CAC scan with CXR obtained within 6 months and divided them into training (n = 391) and validation (n = 168) cohorts. We extracted radiomic features from annotated cardiac contours in the CXR images and developed an RS through feature selection with the least absolute shrinkage and selection operator regression in the training cohort. We evaluated the incremental value of the RS in predicting CAC scores when combined with basic clinical factor in the validation cohort. To predict a CAC score ≥100, we built an RS-based machine learning model using random forest; the input variables were age, sex, body mass index, and RS.

RESULTS

The RS was the most prominent factor for the CAC score ≥100 predictions (odds ratio = 2.33; 95% confidence interval: 1.62-3.44; P < 0.001) compared with basic clinical factor. The machine learning model was tested in the validation cohort and showed an area under the receiver operating characteristic curve of 0.808 (95% confidence interval: 0.75-0.87) for a CAC score ≥100 predictions.

CONCLUSIONS

The use of an RS-based machine learning model may have the potential as an imaging marker to screen patients with moderate to severe CAC scores before diagnostic imaging tests, and it may improve the pretest probability of detecting coronary artery disease in clinical practice.

3D fusion is superior to 2D point-to-point contrast-enhanced US to evaluate the ablative margin after RFA for hepatocellular carcinoma

PURPOSE

To compare the efficiency of three-dimensional (3D) and two-dimensional (2D) contrast-enhanced ultrasound (CEUS)-derived techniques in evaluating the ablative margin (AM) after radiofrequency ablation (RFA) for hepatocellular carcinoma (HCC).

METHODS

In total, 98 patients with 98 HCCs were enrolled. The 2D CEUS point-to-point imaging (2D CEUS-PI) was conducted by comparing the pre- and post-RFA 2D CEUS images manually, and the 3D CEUS fusion imaging (3D CEUS-FI) was conducted by fusing the pre- and post-RFA 3D CEUS images automatically. These two techniques were compared in distinguishing an adequate AM ≥ 5 mm. Risk factors for local tumor progression (LTP) after RFA were analyzed by the Kaplan-Meier method with log-rank test.

RESULTS

The mean registration time of 3D CEUS-FI and 2D CEUS-PI was 5.0 and 9.3 min, respectively (p < 0.0001). The kappa coefficient was 0.680 for agreement between 2D CEUS-PI and 3D CEUS-FI in the evaluation of AM (p < 0.0001). Tumors with AM < 5 mm by 2D CEUS-PI were all identified as AM < 5 mm by 3D CEUS-FI. Nonetheless, 16 (26%) tumors identified as AM ≥ 5 mm by 2D CEUS-PI were re-classified as AM < 5 mm by 3D CEUS-FI. During a median follow-up time of 31.2 months (range, 3.2-66.0 months), LTP was identified in 8 tumors. The estimated 1-/2-/3-year cumulative incidence of LTP was 4.4%, 8.1%, and 10.3%, respectively. Higher estimated cumulative incidence of LTP was identified in tumors with AM < 5 mm by 2D CEUS-PI (at 3-year, 27.2% vs 0%; p < 0.001), and by 3D CEUS-FI (at 3-year, 20.7% vs 0%; p = 0.004).

CONCLUSION

3D CEUS-FI excelled in the evaluation of AM when compared with 2D CEUS-PI. With equivalent efficacy in the prediction of LTP, 3D CEUS-FI was superior to 2D CEUS-PI for its automatic and time-saving procedure.

CLINICAL RELEVANCE STATEMENT

3D CEUS fusion imaging may serve as an effective tool in evaluating ablative margin and predicting local tumor progression after RFA in HCC.

KEY POINTS

• Both 2D and 3D CEUS-derived techniques could evaluate ablative margin (AM) after RFA for hepatocellular carcinoma. • 3D CEUS fusion imaging was more precise in the evaluation of AM compared to 2D CEUS point-to-point imaging, with advantages of its automatic and time-saving procedure. • An inadequate AM < 5 mm evaluated by CEUS-derived techniques was the only risk factor of LTP after RFA for hepatocellular carcinoma (p < 0.001 for 2D CEUS point-to-point imaging, and p = 0.004 for 3D CEUS fusion imaging).

Enhancing preoperative diagnosis of microvascular invasion in hepatocellular carcinoma: domain-adaptation fusion of multi-phase CT images

Objectives

In patients with hepatocellular carcinoma (HCC), accurately predicting the preoperative microvascular invasion (MVI) status is crucial for improving survival rates. This study proposes a multi-modal domain-adaptive fusion model based on deep learning methods to predict the preoperative MVI status in HCC.

Materials and methods

From January 2008 to May 2022, we collected 163 cases of HCC from our institution and 42 cases from another medical facility, with each case including Computed Tomography (CT) images from the pre-contrast phase (PCP), arterial phase (AP), and portal venous phase (PVP). We divided our institution's dataset (n=163) into training (n=119) and test sets (n=44) in an approximate 7:3 ratio. Additionally, we included cases from another institution (n=42) as an external validation set (test1 set). We constructed three single-modality models, a simple concatenated multi-modal model, two current state-of-the-art image fusion model and a multi-modal domain-adaptive fusion model (M-DAFM) based on deep learning methods. We evaluated and analyzed the performance of these constructed models in predicting preoperative MVI using the area under the receiver operating characteristic curve (AUC), decision curve analysis (DCA), and net reclassification improvement (NRI) methods.

Results

In comparison with all models, M-DAFM achieved the highest AUC values across the three datasets (0.8013 for the training set, 0.7839 for the test set, and 0.7454 for the test1 set). Notably, in the test set, M-DAFM's Decision Curve Analysis (DCA) curves consistently demonstrated favorable or optimal net benefits within the 0-0.65 threshold probability range. Additionally, the Net Reclassification Improvement (NRI) values between M-DAFM and the three single-modal models, as well as the simple concatenation model, were all greater than 0 (all p < 0.05). Similarly, the NRI values between M-DAFM and the two current state-of-the-art image fusion models were also greater than 0. These findings collectively indicate that M-DAFM effectively integrates valuable information from multi-phase CT images, thereby enhancing the model's preoperative predictive performance for MVI.

Conclusion

The M-DAFM proposed in this study presents an innovative approach to improve the preoperative predictive performance of MVI.

A novel nomogram for predicting microvascular invasion in hepatocellular carcinoma

INTRODUCTION AND OBJECTIVES

In hepatocellular carcinoma (HCC), the prognosis of patients with microvascular invasion (MVI) is poor. Therefore, in this study, we established and evaluated the performance of a novel nomogram to predict MVI in patients with HCC.

MATERIALS AND METHODS

We retrospectively obtained clinical data of 497 patients with HCC who underwent hepatectomy at Liaoning Cancer Hospital from November 1, 2018, to November 4, 2021. The patients (n = 497) were randomized in a 7:3 ratio into the training cohort (TC, n = 349) and the validation cohort (VC, n = 148). We performed Least Absolute Shrinkage and Selection Operator (LASSO) and univariate as well as multivariate logistic regression analyses (ULRA, MRLA) on patients in the TC to identify factors independently predicting MVI.

RESULTS

Preoperative FIB-4, AFU, AFP levels, liver cirrhosis, and non-smooth tumor margin were independent risk factors for preoperative MVI prediction. The C-index of the TC, VC, and the entire cohort was 0.846, 0.786, and 0.829, respectively. The calibration curves demonstrated the outstanding agreement between predicted MVI incidences by our model and the actual MVI risk. Decision curve analysis (DCA) confirmed the significance of our predictive model in clinical settings. The Kaplan-Meier (KM) survival curve showed that the recurrence-free survival (RFS) and overall survival (OS) of patients in the high-MVI risk group were poor compared to those in the low-MVI risk group.

CONCLUSIONS

We constructed and evaluated the performance of the novel nomogram for predicting MVI risk. Our predictive model could adequately predict MVI risk and aid clinicians in selecting appropriate therapeutic strategies for patients.

Preoperative prediction and risk assessment of microvascular invasion in hepatocellular carcinoma

Hepatocellular carcinoma (HCC) is one of the most common and highly lethal tumors worldwide. Microvascular invasion (MVI) is a significant risk factor for recurrence and poor prognosis after surgical resection for HCC patients. Accurately predicting the status of MVI preoperatively is critical for clinicians to select treatment modalities and improve overall survival. However, MVI can only be diagnosed by pathological analysis of postoperative specimens. Currently, numerous indicators in serology (including liquid biopsies) and imaging have been identified to effective in predicting the occurrence of MVI, and the multi-indicator model based on deep learning greatly improves accuracy of prediction. Moreover, several genes and proteins have been identified as risk factors that are strictly associated with the occurrence of MVI. Therefore, this review evaluates various predictors and risk factors, and provides guidance for subsequent efforts to explore more accurate predictive methods and to facilitate the conversion of risk factors into reliable predictors.

Contrast-Enhanced Ultrasound with Deep Learning with Attention Mechanisms for Predicting Microvascular Invasion in Single Hepatocellular Carcinoma

RATIONALE AND OBJECTIVES

Prediction of microvascular invasion (MVI) status of hepatocellular carcinoma (HCC) holds clinical significance for decision-making regarding the treatment strategy and evaluation of patient prognosis. We developed a deep learning (DL) model based on contrast-enhanced ultrasound (CEUS) to predict MVI of HCC.

MATERIALS AND METHODS

We retrospectively analyzed the data for single primary HCCs that were evaluated with CEUS 1 week before surgical resection from December 2014 to February 2022. The study population was divided into training (n = 198) and test (n = 54) cohorts. In this study, three DL models (Resnet50, Resnet50+BAM, Resnet50+SE) were trained using the training cohort and tested in the test cohort. Tumor characteristics were also evaluated by radiologists, and multivariate regression analysis was performed to determine independent indicators for the development of predictive nomogram models. The performance of the three DL models was compared to that of the MVI prediction model based on radiologist evaluations.

RESULTS

The best-performing model, ResNet50+SE model achieved the ROC of 0.856, accuracy of 77.2, specificity of 93.9%, and sensitivity of 52.4% in the test group. The MVI prediction model based on a combination of three independent predictors showed a C-index of 0.729, accuracy of 69.4, specificity of 73.8%, and sensitivity of 62%.

CONCLUSION

The DL algorithm can accurately predict MVI of HCC on the basis of CEUS images, to help identify high-risk patients for the assist treatment.

Intratumoral and peritumoral radiomics model based on abdominal ultrasound for predicting Ki-67 expression in patients with hepatocellular cancer

Background

Hepatocellular cancer (HCC) is one of the most common tumors worldwide, and Ki-67 is highly important in the assessment of HCC. Our study aimed to evaluate the value of ultrasound radiomics based on intratumoral and peritumoral tissues in predicting Ki-67 expression levels in patients with HCC.

Methods

We conducted a retrospective analysis of ultrasonic and clinical data from 118 patients diagnosed with HCC through histopathological examination of surgical specimens in our hospital between September 2019 and January 2023. Radiomics features were extracted from ultrasound images of both intratumoral and peritumoral regions. To select the optimal features, we utilized the t-test and the least absolute shrinkage and selection operator (LASSO). We compared the area under the curve (AUC) values to determine the most effective modeling method. Subsequently, we developed four models: the intratumoral model, the peritumoral model, combined model #1, and combined model #2.

Results

Of the 118 patients, 64 were confirmed to have high Ki-67 expression while 54 were confirmed to have low Ki-67 expression. The AUC of the intratumoral model was 0.796 (0.649-0.942), and the AUC of the peritumoral model was 0.772 (0.619-0.926). Furthermore, combined model#1 yielded an AUC of 0.870 (0.751-0.989), and the AUC of combined model#2 was 0.762 (0.605-0.918). Among these models, combined model#1 showed the best performance in terms of AUC, accuracy, F1-score, and decision curve analysis (DCA).

Conclusion

We presented an ultrasound radiomics model that utilizes both intratumoral and peritumoral tissue information to accurately predict Ki-67 expression in HCC patients. We believe that incorporating both regions in a proper manner can enhance the diagnostic performance of the prediction model. Nevertheless, it is not sufficient to include both regions in the region of interest (ROI) without careful consideration.

Integration of Multimodal Computed Tomography Radiomic Features of Primary Tumors and the Spleen to Predict Early Recurrence in Patients with Postoperative Adjuvant Transarterial Chemoembolization

Background

Hepatocellular carcinoma (HCC) is one of the most lethal malignancies in the world. Patients with HCC choose postoperative adjuvant transarterial chemoembolization (PA-TACE) after surgical resection to reduce the risk of recurrence. However, many of them have recurrence within a short period.

Methods

In this retrospective analysis, a total of 173 patients who underwent PA-TACE between September 2016 and March 2020 were recruited. Radiomic features were derived from the arterial and venous phases of each patient. Early recurrence (ER)-related radiomics features of HCC and the spleen were selected to build two rad-scores using the least absolute shrinkage and selection operator (LASSO) Cox regression analysis. Logistic regression was applied to establish the Radiation (Rad)_score by combining the two regions. We constructed a nomogram containing clinical information and dual-region rad-scores, which was evaluated in terms of discrimination, calibration, and clinical usefulness.

Results

All three radiological scores showed good performance for ER prediction. The combined Rad_score performed the best, with an area under the curve (AUC) of 0.853 (95% confidence interval [CI], 0.783-0.908) in the training set and 0.929 (95% CI, 0.789-0.988) in the validation set. Multivariate analysis identified total bilirubin (TBIL) and the combined Rad_score as independent prognostic factors for ER. The nomogram was found to be clinically valuable, as determined by the decision curves (DCA) and clinical impact curves (CIC).

Conclusion

A multimodal dual-region radiomics model combining HCC and the spleen is an independent prognostic tool for ER. The combination of dual-region radiomics features and clinicopathological factors has a good clinical application value.

Ultrasound radiomics in the prediction of microvascular invasion in hepatocellular carcinoma: A systematic review and meta-analysis

Purpose

To systematically assess the clinical value of ultrasound radiomics in the prediction of microvascular invasion in hepatocellular carcinoma (HCC).

Methods

Relevant articles were searched in PubMed, Web of Science, Cochrane Library, Embase and Medline and screened according to the eligibility criteria. The quality of the included articles was assessed based on the Quality Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2) tool. After article assessment and data extraction, the diagnostic performance of ultrasound radiomics was evaluated based on pooled sensitivity, specificity, positive likelihood ratio (PLR), negative likelihood ratio (NLR) and diagnostic odds ratio (DOR), and the area under the curve (AUC) was calculated by generating the ROC curve. Meta-analysis was performed using Stata 15.1, and subgroup analysis was conducted to identify the sources of heterogeneity. A Fagan nomogram was generated to assess the clinical utility of ultrasound radiomics.

Results

Five studies involving 1260 patients were included. Meta-analysis showed that ultrasound radiomics had a pooled sensitivity of 79% (95% CI: 75-83%), specificity of 70% (95% CI: 59-79%), PLR of 2.6 (95% CI: 1.9-3.7), NLR of 0.30 (95% CI: 0.23-0.39), DOR of 9 (95% CI: 5-16), and AUC of 0.81 (95% CI: 0.78-0.85). Sensitivity analysis indicated that the results were statistically reliable and stable, and no significant difference was identified during subgroup analysis.

Conclusion

Ultrasound radiomics has favorable predictive performance in the microvascular invasion of HCC and may serve as an auxiliary tool for guiding clinical decision-making.

Non-invasive Assessment of Axillary Lymph Node Metastasis Risk in Early Invasive Breast Cancer Adopting Automated Breast Volume Scanning-Based Radiomics Nomogram: A Multicenter Study

OBJECTIVE

The aim of the work described here was to develop a non-invasive tool based on the radiomics and ultrasound features of automated breast volume scanning (ABVS), clinicopathological factors and serological indicators to evaluate axillary lymph node metastasis (ALNM) in patients with early invasive breast cancer (EIBC).

METHODS

We retrospectively analyzed 179 ABVS images of patients with EIBC at a single center from January 2016 to April 2022 and divided the patients into training and validation sets (ratio 8:2). Additionally, 97 ABVS images of patients with EIBC from a second center were enrolled as the test set. The radiomics signature was established with the least absolute shrinkage and selection operator. Significant ALNM predictors were screened using univariate logistic regression analysis and further combined to construct a nomogram using the multivariate logistic regression model. The receiver operating characteristic curve assessed the nomogram's predictive performance.

DISCUSSION

The constructed radiomics nomogram model, including ABVS radiomics signature, ultrasound assessment of axillary lymph node (ALN) status, convergence sign and erythrocyte distribution width (standard deviation), achieved moderate predictive performance for risk probability evaluation of ALNs in patients with EIBC. Compared with ultrasound, the nomogram model was able to provide a risk probability evaluation tool not only for the ALNs with positive ultrasound features but also for micrometastatic ALNs (generally without positive ultrasound features), which benefited from the radiomics analysis of multi-sourced data of patients with EIBC.

CONCLUSION

This ABVS-based radiomics nomogram model is a pre-operative, non-invasive and visualized tool that can help clinicians choose rational diagnostic and therapeutic protocols for ALNM.

A nomogram for preoperative prediction of microvascular invasion in ruptured hepatocellular carcinoma

BACKGROUND AND AIM

Microvascular invasion (MVI) is defined as the presence of micrometastatic cancer cell emboli in hepatic vessels, including small vessels, and at present, researchers believe that is an important factor for early postoperative recurrence and survival. Here, we developed and validated a preoperative predictive model for the presence of MVI in patients with ruptured hepatocellular carcinoma (rHCC).

METHODS

We retrospectively collected data for 210 rHCC patients who underwent staged hepatectomy at Wuhan Tongji Hospital, and 91 patients who underwent staged hepatectomy at Zhongshan People's Hospital between January 2010 and March 2021. Then, the former was used as the training cohort and the latter was used as the validation cohort. Logistic regression was used to screen for variables associated with MVI, and these variables were used to construct nomograms. We used R software to assess the discrimination, calibration ability, as well as clinical efficacy of nomograms.

RESULTS

Multivariate logistic regression analysis identified four risk factors independently associated with MVI: max tumor length [odds ratio (OR) = 1.385; 95% confidence interval (CI), 1.072-1.790], number of tumors (OR = 2.182; 95% CI, 1.129-5.546), direct bilirubin (OR = 1.515; 95% CI, 1.189-1.930), and alpha-fetoprotein (cutoff = 400 ng/mL) (OR = 2.689; 95% CI, 3.395-13.547). Nomograms were built from the four variables and they were tested for discrimination and calibration, and the results were good.

CONCLUSION

We developed and validated a preoperative predictive model for the presence of MVI in patients with ruptured HCC. This model can help clinicians identify patients at risk of MVI and make better treatment options.

A novel predictive model of microvascular invasion in hepatocellular carcinoma based on differential protein expression

BACKGROUND

This study aims to construct and verify a nomogram model for microvascular invasion (MVI) based on hepatocellular carcinoma (HCC) tumor characteristics and differential protein expressions, and explore the clinical application value of the prediction model.

METHODS

The clinicopathological data of 200 HCC patients were collected and randomly divided into training set and validation set according to the ratio of 7:3. The correlation between MVI occurrence and primary disease, age, gender, tumor size, tumor stage, and immunohistochemical characteristics of 13 proteins, including GPC3, CK19 and vimentin, were statistically analyzed. Univariate and multivariate analyzes identified risk factors and independent risk factors, respectively. A nomogram model that can be used to predict the presence of MVI was subsequently constructed. Then, receiver operating characteristic (ROC) curve, calibration curve, and decision curve analysis (DCA) were conducted to assess the performance of the model.

RESULTS

Multivariate logistic regression analysis indicated that tumor size, GPC3, P53, RRM1, BRCA1, and ARG were independent risk factors for MVI. A nomogram was constructed based on the above six predictors. ROC curve, calibration, and DCA analysis demonstrated the good performance and the clinical application potential of the nomogram model.

CONCLUSIONS

The predictive model constructed based on the clinical characteristics of HCC tumors and differential protein expression patterns could be helpful to improve the accuracy of MVI diagnosis in HCC patients.

Radiomics in the diagnosis and treatment of hepatocellular carcinoma

Hepatocellular carcinoma (HCC) is a common malignant tumor. At present, early diagnosis of HCC is difficult and therapeutic methods are limited. Radiomics can achieve accurate quantitative evaluation of the lesions without invasion, and has important value in the diagnosis and treatment of HCC. Radiomics features can predict the development of cancer in patients, serve as the basis for risk stratification of HCC patients, and help clinicians distinguish similar diseases, thus improving the diagnostic accuracy. Furthermore, the prediction of the treatment outcomes helps determine the treatment plan. Radiomics is also helpful in predicting the HCC recurrence, disease-free survival and overall survival. This review summarized the role of radiomics in the diagnosis, treatment and prognosis of HCC.

Radiomics nomograms based on R2* mapping and clinical biomarkers for staging of liver fibrosis in patients with chronic hepatitis B: a single-center retrospective study

OBJECTIVES

To investigate the value of R2* mapping-based radiomics nomograms in staging liver fibrosis in patients with chronic hepatitis B.

METHODS

Between January 2020 and December 2020, 151 patients with chronic hepatitis B were randomly divided into training (n = 103) and validation (n = 48) cohorts. From January to February 2021, 58 patients were included in a test cohort. Radiomics features were selected using the interclass correlation coefficient and least absolute shrinkage and selection operator method. Three radiomics nomograms, combining the radiomics score (Radscore) derived from R2* mapping and clinical variables, were used for staging significant and advanced fibrosis, and cirrhosis. Performance of the model was evaluated using the AUC. The utility and clinical benefits were evaluated using the continuous net reclassification index (NRI), integrated discrimination improvement (IDI), and decision curve analysis (DCA).

RESULTS

The Radscore calculated by 12 radiomics features and independent factors (laminin and platelet) of advanced fibrosis were used to construct the radiomics nomograms. In the test cohort, the AUCs of the radiomics nomograms for staging significant fibrosis, advanced fibrosis, and cirrhosis were 0.738 (95% confidence interval [CI]: 0.604-0.872), 0.879 (95% CI: 0.779-0.98), and 0.952 (95% CI: 0.878-1), respectively. NRI, IDI, and DCA confirmed that radiomics nomograms demonstrated varying degrees of clinical benefit and improvement for advanced fibrosis and cirrhosis, but not for significant fibrosis.

CONCLUSIONS

Radiomics nomograms combined with R2* mapping-based Radscore, laminin, and platelet have value in staging advanced fibrosis and cirrhosis but limited value for staging significant fibrosis.

KEY POINTS

• Laminin and platelets were independent predictors of advanced fibrosis. • Radiomics analysis based on R2* mapping was beneficial for evaluating advanced fibrosis and cirrhosis. • It was difficult to distinguish significant fibrosis using a radiomics nomogram, which is possibly due to the complex pathological microenvironment of chronic liver diseases.

Quality of radiomics for predicting microvascular invasion in hepatocellular carcinoma: a systematic review

OBJECTIVES

To comprehensively evaluate the reporting quality, risk of bias, and radiomics methodology quality of radiomics models for predicting microvascular invasion in hepatocellular carcinoma.

METHODS

A systematic search of available literature was performed in PubMed, Embase, Web of Science, Scopus, and the Cochrane Library up to January 21, 2022. Studies that developed and/or validated machine learning models based on radiomics data to predict microvascular invasion in hepatocellular carcinoma were included. These studies were reviewed by two investigators and the consensus data were used for analyzing. The reporting quality, risk of bias, and radiomics methodological quality were evaluated by Transparent Reporting of a multivariable prediction model for Individual Prognosis or Diagnosis (TRIPOD), Prediction model Risk of Bias Assessment Tool, and Radiomics Quality Score (RQS), respectively.

RESULTS

A total of 30 studies met eligibility criteria with 24 model developing studies and 6 model developing and external validation studies. The median overall TRIPOD adherence was 75.4% (range 56.7-94.3%). All studies were at high risk of bias with at least 2 of 20 sources of bias. Furthermore, 28 studies showed unclear risks of bias in up to 5 signaling questions because of the lack of specified reports. The median RQS score was 37.5% (range 25-61.1%).

CONCLUSION

Current radiomic models for MVI-status prediction have moderate to good reporting quality, moderate radiomics methodology quality, and high risk of bias in model development and validation.

KEY POINTS

• Current microvascular invasion prediction radiomics studies have moderate to good reporting quality, moderate radiomics methodology quality, and high risk of bias in model development and validation. • Data representativeness, feature robustness, events-per-variable ratio, evaluation metrics, and appropriate validation are five main aspects futures studies should focus more on to improve the quality of radiomics. • Both Radiomics Quality Score and Prediction model Risk of Bias Assessment Tool are needed to comprehensively evaluate a radiomics study.

PET-guided attention for prediction of microvascular invasion in preoperative hepatocellular carcinoma on PET/CT

PURPOSE

To achieve PET/CT-based preoperative prediction of microvascular invasion in hepatocellular carcinoma by combining the advantages of PET and CT.

METHODS

This retrospective study included a total of 100 patients from two institutions who underwent PET/CT imaging. The above patients were divided into a training cohort (n = 70) and a validation cohort (n = 30). This study was based on PET/CT images to evaluate the possibility of microvascular invasion (MVI) of patients. In this study, we proposed a two-branch PET-guided attention network to predict MVI. The model used a two-branch network to extract image features from PET and CT, respectively. The PET-guided attention module aimed to enable the model to focus on the lesion region and reduce the disturbance of irrelevant and redundant information. Model performance was evaluated by the area under the curve (AUC) of the receiver operating characteristic (ROC) curve.

RESULTS

The method outperformed the single-modality prediction model for preoperative hepatocyte microvascular invasion, achieving an AUC of 0.907. On the validation set, accuracy reached 0.846, precision reached 0.881, recall 0.793, and F1-score 0.835.

CONCLUSION

The model exploits the particularities of the molecular metabolic function of PET and the anatomical structure of CT and can strongly improve the accuracy of clinical diagnosis of MVI.

Prediction of Ki-67 of Invasive Ductal Breast Cancer Based on Ultrasound Radiomics Nomogram

PURPOSE

The objective of this research was to develop and validate an ultrasound-based radiomics nomogram for the pre-operative assessment of Ki-67 in breast cancer (BC).

MATERIALS AND METHODS

From December 2016 to December 2018, 515 patients with invasive ductal breast cancer who received two-dimensional (2D) ultrasound and Ki-67 examination were studied and analyzed retrospectively. The dataset was distributed at random into a training cohort (n = 360) and a test cohort (n = 155) in the ratio of 7:3. Each tumor region of interest was defined based on 2D ultrasound images and radiomics features were extracted. ANOVA, maximum correlation minimum redundancy (mRMR) algorithm, and minimum absolute shrinkage and selection operator (LASSO) were performed to pick features, and independent clinical predictors were integrated with radscore to construct the nomogram for predicting Ki-67 index by univariate and multivariate logistic regression analysis. The performance and utility of the models were evaluated by plotting receiver operating characteristic (ROC) curves, decision curve analysis (DCA), and calibration curves.

RESULTS

In the testing cohort, the area under the receiver characteristic curve (AUC) of the nomogram was 0.770 (95% confidence interval, 0.690-0.860). In both cohorts, the nomogram outperformed both the clinical model and the radiomics model (P < .05 according to the DeLong test). The analysis of DCA proved that the model has clinical utility.

CONCLUSIONS

The nomogram based on 2D ultrasound images offered an approach for predicting Ki-67 in BC.

Musculoskeletal Ultrasound Image-Based Radiomics for the Diagnosis of Achilles Tendinopathy in Skiers

OBJECTIVES

Our study aimed to develop and validate an efficient ultrasound image-based radiomic model for determining the Achilles tendinopathy in skiers.

METHODS

A total of 88 feet of skiers clinically diagnosed with unilateral chronic Achilles tendinopathy and 51 healthy feet were included in our study. According to the time order of enrollment, the data were divided into a training set (n = 89) and a test set (n = 50). The regions of interest (ROIs) were segmented manually, and 833 radiomic features were extracted from red, green, blue color channels and grayscale of ROIs using Pyradiomics, respectively. Three feature selection and three machine learning modeling algorithms were implemented respectively, for determining the optimal radiomics pipeline. Finally, the area under the receiver operating characteristic curve (AUC), consistency analysis, and decision analysis were used to evaluate the diagnostic performance.

RESULTS

By comparing nine radiomics analysis strategies of three color channels and grayscale, the radiomic model under the green channel obtained the best diagnostic performance, using the Random Forest selection and Support Vector Machine modeling, which was selected as the final machine learning model. All the selected radiomic features were significantly associated with the Achilles tendinopathy (P < .05). The radiomic model had a training AUC of 0.98, a test AUC of 0.99, a sensitivity of 0.90, and a specificity of 1, which could bring sufficient clinical net benefits.

CONCLUSIONS

Ultrasound image-based radiomics achieved high diagnostic performance, which could be used as an intelligent auxiliary tool for the diagnosis of Achilles tendinopathy.

Combining Preoperative Clinical and Imaging Characteristics to Predict MVI in Hepatitis B Virus-Related Combined Hepatocellular Carcinoma and Cholangiocarcinoma

BACKGROUND

Combined hepatocellular carcinoma and cholangiocarcinoma (cHCC-CCA) is a rare form of primary liver malignancy. Microvascular invasion (MVI) indicates poor postsurgical prognosis in cHCC-CCA. The objective of this study was to investigate preoperative predictors of MVI in hepatitis B virus (HBV) -related cHCC-CCA patients.

METHODS

A total of 69 HBV-infected patients with pathologically confirmed cHCC-CCA who underwent hepatectomy were included. Univariate and multivariate analyses were conducted to determine independent risk factors that were then incorporated into the predictive model associated with MVI. Receiver operating characteristic analysis was used to assess the predictive performance of the new model.

RESULTS

For the multivariate analysis, γ-glutamyl transpeptidase (OR, 3.69; p = 0.034), multiple nodules (OR, 4.41; p = 0.042) and peritumoral enhancement (OR, 6.16; p = 0.004) were independently associated with MVI. Active replication of HBV indicated by positive HBeAg showed no differences between MVI-positive and MVI-negative patients. The prediction score using the independent predictors achieved an area under the curve of 0.813 (95% CI 0.717-0.908). A significantly lower recurrence-free survival was observed in the high-risk group with a score of ≥1 (p < 0.001).

CONCLUSION

γ-glutamyl transpeptidase, peritumoral enhancement and multiple nodules were independent preoperative predictors of MVI in HBV-related cHCC-CCA patients. The established prediction score demonstrated satisfactory performance in predicting MVI pre-operatively and may facilitate prognostic stratification.

Performance of Radiomics in Microvascular Invasion Risk Stratification and Prognostic Assessment in Hepatocellular Carcinoma: A Meta-Analysis

BACKGROUND

Primary liver cancer is the sixth most commonly diagnosed cancer and the third leading cause of cancer death. Advances in phenomenal imaging are paving the way for application in diagnosis and research. The poor prognosis of advanced HCC warrants a personalized approach. The objective was to assess the value of imaging phenomics for risk stratification and prognostication of HCC.

METHODS

We performed a meta-analysis of manuscripts published to January 2023 on MEDLINE addressing the value of imaging phenomics for HCC risk stratification and prognostication. Publication information for each were collected using a standardized data extraction form.

RESULTS

Twenty-seven articles were analyzed. Our study shows the importance of imaging phenomics in HCC MVI prediction. When the training and validation datasets were analyzed separately by the random-effects model, in the training datasets, radiomics had good MVI prediction (AUC of 0.81 (95% CI 0.76-0.86)). Similar results were found in the validation datasets (AUC of 0.79 (95% CI 0.72-0.85)). Using the fixed effects model, the mean AUC of all datasets was 0.80 (95% CI 0.76-0.84).

CONCLUSIONS

Imaging phenomics is an effective solution to predict microvascular invasion risk, prognosis, and treatment response in patients with HCC.

Ensemble Learning of Multiple Models Using Deep Learning for Multiclass Classification of Ultrasound Images of Hepatic Masses

Ultrasound (US) is often used to diagnose liver masses. Ensemble learning has recently been commonly used for image classification, but its detailed methods are not fully optimized. The purpose of this study is to investigate the usefulness and comparison of some ensemble learning and ensemble pruning techniques using multiple convolutional neural network (CNN) trained models for image classification of liver masses in US images. Dataset of the US images were classified into four categories: benign liver tumor (BLT) 6320 images, liver cyst (LCY) 2320 images, metastatic liver cancer (MLC) 9720 images, primary liver cancer (PLC) 7840 images. In this study, 250 test images were randomly selected for each class, for a total of 1000 images, and the remaining images were used as the training. 16 different CNNs were used for training and testing ultrasound images. The ensemble learning used soft voting (SV), weighted average voting (WAV), weighted hard voting (WHV) and stacking (ST). All four types of ensemble learning (SV, ST, WAV, and WHV) showed higher values of accuracy than the single CNN. All four types also showed significantly higher deep learning (DL) performance than ResNeXt101 alone. For image classification of liver masses using US images, ensemble learning improved the performance of DL over a single CNN.

Preoperative prediction model of microvascular invasion in patients with hepatocellular carcinoma

BACKGROUND

Microvascular invasion (MVI) is an adverse factor for the prognosis of patients with hepatocellular carcinoma (HCC). We aimed to construct a preoperative prediction model for MVI, thereby providing a reference for clinicians in formulating treatment options for HCC.

METHODS

A total of 360 patients with non-metastatic HCC were retrospectively enrolled. We used logistic regression analysis to screen out independent risk factors for MVI and further constructed a predictive model for MVI. The performance of the model was evaluated by receiver operating characteristic (ROC) curve, calibration curve, and decision curve analysis (DCA).

RESULTS

Logistic regression analysis revealed that fibrinogen (>4 g/L) (OR: 6.529), alpha-fetoprotein (≥ 400 ng/mL) (OR: 2.676), cirrhosis (OR: 2.25), tumor size (OR: 1.239), and poor tumor border (OR: 3.126) were independent risk factors of MVI. The prediction model of MVI had C-index of 0.746 and 0.772 in the training and validation cohorts, respectively. The calibration curves showed good agreement between actual and predicted MVI risk. Finally, DCA reveals that this model has good clinical utility.

CONCLUSION

The nomogram-based model we established can predict the preoperative MVI well and provides reference for surgeons to make clinical treatment decisions.

Ultrasound-based radiomics score for pre-biopsy prediction of prostate cancer to reduce unnecessary biopsies

BACKGROUND

Patients undergoing prostate biopsies (PBs) suffer from low positive rates and potential risk for complications. This study aimed to develop and validate an ultrasound (US)-based radiomics score for pre-biopsy prediction of prostate cancer (PCa) and subsequently reduce unnecessary PBs.

METHODS

Between December 2015 and March 2018, 196 patients undergoing initial transrectal ultrasound (TRUS)-guided PBs were retrospectively enrolled and randomly assigned to the training or validation cohort at a ratio of 7:3. A total of 1044 radiomics features were extracted from grayscale US images of each prostate nodule. After feature selection through the least absolute shrinkage and selection operator (LASSO) regression model, the radiomics score was developed from the training cohort. The prediction nomograms were developed using multivariate logistic regression analysis based on the radiomics score and clinical risk factors. The performance of the nomograms was assessed and compared in terms of discrimination, calibration, and clinical usefulness.

RESULTS

The radiomics score consisted of five selected features. Multivariate logistic regression analysis demonstrated that the radiomics score, age, total prostate-specific antigen (tPSA), and prostate volume were independent factors for prediction of PCa (all p < 0.05). The integrated nomogram incorporating the radiomics score and three clinical risk factors reached an area under the curve (AUC) of 0.835 (95% confidence interval [CI], 0.729-0.941), thereby outperforming the clinical nomogram which based on only clinical factors and yielded an AUC of 0.752 (95% CI, 0.618-0.886) (p = 0.04). Both nomograms showed good calibration. Decision curve analysis indicated that using the integrated nomogram would add more benefit than using the clinical nomogram.

CONCLUSION

The radiomics score was an independent factor for pre-biopsy prediction of PCa. Addition of the radiomics score to the clinical nomogram shows incremental prognostic value and may help clinicians make precise decisions to reduce unnecessary PBs.

Preoperative and Prognostic Prediction of Microvascular Invasion in Hepatocellular Carcinoma: A Review Based on Artificial Intelligence

Microvascular invasion of hepatocellular carcinoma is an important factor affecting tumor recurrence after liver resection and liver transplantation. There are many ways to classify microvascular invasion, however, an international consensus is urgently needed. Recently, artificial intelligence has emerged as an important tool for improving the clinical management of hepatocellular carcinoma. Many studies about microvascular invasion currently focus on preoperative and prognosis prediction of microvascular invasion using artificial intelligence. In this paper, we review the definition and staging of microvascular invasion, especially the diagnosis of it by using artificial intelligence. In preoperative prediction, deep learning based on multimodal data modeling of radiomics-screened features, clinical features, and medical images is currently the most effective means. In prognostic prediction, pathology is the gold standard, and the techniques used should more effectively utilize the global features of the pathology images.

Ultrasound-based artificial intelligence in gastroenterology and hepatology

Artificial intelligence (AI), especially deep learning, is gaining extensive attention for its excellent performance in medical image analysis. It can automatically make a quantitative assessment of complex medical images and help doctors to make more accurate diagnoses. In recent years, AI based on ultrasound has been shown to be very helpful in diffuse liver diseases and focal liver lesions, such as analyzing the severity of nonalcoholic fatty liver and the stage of liver fibrosis, identifying benign and malignant liver lesions, predicting the microvascular invasion of hepatocellular carcinoma, curative transarterial chemoembolization effect, and prognoses after thermal ablation. Moreover, AI based on endoscopic ultrasonography has been applied in some gastrointestinal diseases, such as distinguishing gastric mesenchymal tumors, detection of pancreatic cancer and intraductal papillary mucinous neoplasms, and predicting the preoperative tumor deposits in rectal cancer. This review focused on the basic technical knowledge about AI and the clinical application of AI in ultrasound of liver and gastroenterology diseases. Lastly, we discuss the challenges and future perspectives of AI.

Circulating tumour cell counts and ultrasomics signature-based nomogram for preoperative prediction of early recurrence of hepatocellular carcinoma after radical treatment

METHODS

Between December 2017 and December 2018, 153 HCC patients (134 males and 19 females; mean age, 56.0 ± 10.2 years; range, 28-78 years) treated with radical therapy were enrolled in our retrospective study and were divided into a training cohort (n = 107) and a validation cohort (n = 46). All patients underwent preoperative CTC tests and CEUS examinations before treatment. The ultrasomics signature was extracted and built from CEUS images. Univariate and multivariate logistic regression analyses were used to identify the significant variables related to ER, which were then combined to build a predictive nomogram. The performance of the nomogram was evaluated by its discrimination, calibration and clinical utility. The predictive model was further evaluated in the internal validation cohort.

RESULTS

HBV DNA, serum AFP level, CTC status, tumour size and ultrasomics score were identified as independent predictors associated with ER (all p < 0.05). Multivariable logistic regression analysis showed that the CTC status (OR = 7.02 [95% CI, 2.07 to 28.38], p = 0.003) and ultrasomics score (OR = 148.65 [95% CI, 25.49 to 1741.72], p < 0.001) were independent risk factors for ER. The nomogram based on ultrasomics score, CTC status, serum AFP level and tumour size exhibited C-indexes of 0.933 (95% CI, 0.878 to 0.988) and 0.910 (95% CI, 0.765 to 1.055) in the training and validation cohorts, respectively, fitting well in calibration curves. Decision curve analysis further confirmed the clinical usefulness of the nomogram.

CONCLUSION

The nomogram incorporating CTC, ultrasomics features and independent clinical risk factors achieved satisfactory preoperative prediction of ER in HCC patients after radical treatment.

ADVANCES IN KNOWLEDGE

1. CTC status and ultrasomics score were identified as independent predictors associated with ER of HCC after radical treatment. 2. The nomogram constructed by ultrasomics score generated by 17 ultrasomics features, combined with CTCs and independent clinical risk factors such as AFP and tumour size. 3. The nomogram exhibited satisfactory discriminative power, and could be clinically useful in the preoperative prediction of ER after radical treatment in HCC patients.

Prediction of early recurrence of HCC after hepatectomy by contrast-enhanced ultrasound-based deep learning radiomics

Objective

This study aims to evaluate the predictive model based on deep learning (DL) and radiomics features from contrast-enhanced ultrasound (CEUS) to predict early recurrence (ER) in patients with hepatocellular carcinoma (HCC).

Methods

One hundred seventy-two patients with HCC who underwent hepatectomy and followed up for at least 1 year were included in this retrospective study. The data were divided according to the 7:3 ratios of training and test data. The ResNet-50 architecture, CEUS-based radiomics, and the combined model were used to predict the early recurrence of HCC after hepatectomy. The receiver operating characteristic (ROC) curve and calibration curve were drawn to evaluate its diagnostic efficiency.

Results

The CEUS-based radiomics ROCs of the “training set” and “test set” were 0.774 and 0.763, respectively. The DL model showed increased prognostic value, the ROCs of the “training set” and “test set” were 0.885 and 0.834, respectively. The combined model ROCs of the “training set” and “test set” were 0.943 and 0.882, respectively.

Conclusion

The deep learning radiomics model integrating DL and radiomics features from CEUS was used to predict ER and achieve satisfactory performance. Its diagnostic efficiency is significantly better than that of the single model.

Ultrasound-based radiomics XGBoost model to assess the risk of central cervical lymph node metastasis in patients with papillary thyroid carcinoma: Individual application of SHAP

Objectives

A radiomics-based explainable eXtreme Gradient Boosting (XGBoost) model was developed to predict central cervical lymph node metastasis (CCLNM) in patients with papillary thyroid carcinoma (PTC), including positive and negative effects.

Methods

A total of 587 PTC patients admitted at Binzhou Medical University Hospital from 2017 to 2021 were analyzed retrospectively. The patients were randomized into the training and test cohorts with an 8:2 ratio. Radiomics features were extracted from ultrasound images of the primary PTC lesions. The minimum redundancy maximum relevance algorithm and the least absolute shrinkage and selection operator regression were used to select CCLNM positively-related features and radiomics scores were constructed. Clinical features, ultrasound features, and radiomics score were screened out by the Boruta algorithm, and the XGBoost model was constructed from these characteristics. SHapley Additive exPlanations (SHAP) was used for individualized and visualized interpretation. SHAP addressed the cognitive opacity of machine learning models.

Results

Eleven radiomics features were used to calculate the radiomics score. Five critical elements were used to build the XGBoost model: capsular invasion, radiomics score, diameter, age, and calcification. The area under the curve was 91.53% and 90.88% in the training and test cohorts, respectively. SHAP plots showed the influence of each parameter on the XGBoost model, including positive (i.e., capsular invasion, radiomics score, diameter, and calcification) and negative (i.e., age) impacts. The XGBoost model outperformed the radiologist, increasing the AUC by 44%.

Conclusions

The radiomics-based XGBoost model predicted CCLNM in PTC patients. Visual interpretation using SHAP made the model an effective tool for preoperative guidance of clinical procedures, including positive and negative impacts.

Role of preoperative prediction of microvascular invasion in hepatocellular carcinoma based on the texture of FDG PET image: A comparison of quantitative metabolic parameters and MRI

Objective: To investigate the role of prediction microvascular invasion (mVI) in hepatocellular carcinoma (HCC) by ¹⁸F-FDG PET image texture analysis and hybrid criteria combining PET/CT and multi-parameter MRI.

Materials and methods: Ninety-seven patients with HCC who received the examinations of MRI and ¹⁸F-FDG PET/CT were retrospectively included in this study and were randomized into training and testing cohorts. The lesion image texture features of ¹⁸F-FDG PET were extracted using MaZda software. The optimal predictive texture features of mVI were selected, and the classification procedure was conducted. The predictive performance of mVI by radiomics classier in training and testing cohorts was respectively recorded. Next, the hybrid model was developed by integrating the ¹⁸F-FDG PET image texture, metabolic parameters, and MRI parameters to predict mVI through logistic regression. Furthermore, the diagnostic performance of each time was recorded.

Results: The ¹⁸F-FDG PET image radiomics classier showed good predicted performance in both training and testing cohorts to discriminate HCC with/without mVI, with an AUC of 0.917 (95% CI: 0.824–0.970) and 0.771 (95% CI: 0.578, 0.905). The hybrid model, which combines radiomics classier, SUVmax, ADC, hypovascular arterial phase enhancement pattern on contrast-enhanced MRI, and non-smooth tumor margin, also yielded better predictive performance with an AUC of 0.996 (95% CI: 0.939, 1.000) and 0.953 (95% CI: 0.883, 1.000). The differences in AUCs between radiomics classier and hybrid classier were significant in both training and testing cohorts (DeLong test, both p < 0.05).

Conclusion: The radiomics classier based on ¹⁸F-FDG PET image texture and the hybrid classier incorporating ¹⁸F-FDG PET/CT and MRI yielded good predictive performance, which might provide a precise prediction of HCC mVI preoperatively.

The latest application progress of radiomics in prediction and diagnosis of liver diseases

INTRODUCTION

Early detection and individualized treatment of patients with liver disease is the key to survival. Radiomics can extract high-throughput quantitative features by multimode imaging, which has good application prospects for the diagnosis, staging and prognosis of benign and malignant liver diseases. Therefore, this paper summarizes the current research status in the field of liver disease, in order to help these patients achieve personalized and precision medical care.

AREAS COVERED

This paper uses several keywords on the PubMed database to search the references, and reviews the workflow of traditional radiomics, as well as the characteristics and influencing factors of different imaging modes. At the same time, the references on the application of imaging in different benign and malignant liver diseases were also summarized.

EXPERT OPINION

For patients with liver disease, the traditional imaging evaluation can only provide limited information. Radiomics exploits the characteristics of high-throughput and high-dimensional extraction, enabling liver imaging capabilities far beyond the scope of traditional visual image analysis. Recent studies have demonstrated the prospect of this technology in personalized diagnosis and treatment decision in various fields of the liver. However, further clinical validation is needed in its application and practice.

Nomogram based on ultrasound radiomics score and clinical variables for predicting histologic subtypes of epithelial ovarian cancer

OBJECTIVE

Ovarian cancer is one of the most common causes of death in gynecological tumors, and its most common type is epithelial ovarian cancer (EOC). This study aimed to establish a radiomics signature based on ultrasound images to predict the histopathological types of EOC.

METHODS

Overall, 265 patients with EOC who underwent preoperative ultrasonography and surgery were eligible. They were randomly sorted into two cohorts (training cohort: test cohort = 7:3). We outlined the region of interest of the tumor on the ultrasound images of the lesion. Then, the radiomics features were extracted. Clinical, Rad-score and combined models were constructed based on the least absolute shrinkage, selection operator, and logistic regression analysis. The performance of the models was evaluated using receiver operating characteristic curves and decision curve analysis (DCA). A nomogram was formulated based on the combined prediction model.

RESULTS

The combined model had good performance in predicting EOC histopathological types, with an AUC of 0.83 (95% CI: 0.77-0.90) and 0.82 (95% CI: 0.71-0.93) in the training and test cohorts, respectively. The calibration curves showed that the nomogram estimation was consistent with the actual observations. DCA also verified the clinical value of the combined model.

CONCLUSIONS

The combined model containing clinical and ultrasound radiomics features showed an excellent performance in predicting type I and type II EOC.

ADVANCES IN KNOWLEDGE

This study presents the first application of ultrasound radiomics features to distinguish EOC histopathological types. The proposed clinical-radiomics nomogram could help gynecologists non-invasively identify EOC types before surgery.

Ultrasound-based radiomics technology in fetal lung texture analysis prediction of neonatal respiratory morbidity

To develop a novel method for predicting neonatal respiratory morbidity (NRM) by ultrasound-based radiomics technology. In this retrospective study, 430 high-throughput features per fetal-lung image were extracted from 295 fetal lung ultrasound images (four-chamber view) in 295 single pregnancies. Images had been obtained between 28⁺³ and 37⁺⁶ weeks of gestation within 72 h before delivery. A machine-learning model built by RUSBoost (Random under-sampling with AdaBoost) architecture was created using 20 radiomics features extracted from the images and 2 clinical features (gestational age and pregnancy complications) to predict the possibility of NRM. Of the 295 standard fetal lung ultrasound images included, 210 in the training set and 85 in the testing set. The overall performance of the neonatal respiratory morbidity prediction model achieved AUC of 0.88 (95% CI 0.83–0.92) in the training set and 0.83 (95% CI 0.79–0.97) in the testing set, sensitivity of 84.31% (95% CI 79.06–89.44%) in the training set and 77.78% (95% CI 68.30–87.43%) in the testing set, specificity of 81.13% (95% CI 78.16–84.07%) in the training set and 82.09% (95% CI 77.65–86.62%) in the testing set, and accuracy of 81.90% (95% CI 79.34–84.41%) in the training set and 81.18% (95% CI 77.33–85.12%) in the testing set. Ultrasound-based radiomics technology can be used to predict NRM. The results of this study may provide a novel method for non-invasive approaches for the prenatal prediction of NRM.

Artificial intelligence in liver ultrasound

Artificial intelligence (AI) is playing an increasingly important role in medicine, especially in the field of medical imaging. It can be used to diagnose diseases and predict certain statuses and possible events that may happen. Recently, more and more studies have confirmed the value of AI based on ultrasound in the evaluation of diffuse liver diseases and focal liver lesions. It can assess the severity of liver fibrosis and nonalcoholic fatty liver, differentially diagnose benign and malignant liver lesions, distinguish primary from secondary liver cancers, predict the curative effect of liver cancer treatment and recurrence after treatment, and predict microvascular invasion in hepatocellular carcinoma. The findings from these studies have great clinical application potential in the near future. The purpose of this review is to comprehensively introduce the current status and future perspectives of AI in liver ultrasound.

Auto-Segmentation Ultrasound-Based Radiomics Technology to Stratify Patient With Diabetic Kidney Disease: A Multi-Center Retrospective Study

Background

An increasing proportion of patients with diabetic kidney disease (DKD) has been observed among incident hemodialysis patients in large cities, which is consistent with the continuous growth of diabetes in the past 20 years.

Purpose

In this multicenter retrospective study, we developed a deep learning (DL)-based automatic segmentation and radiomics technology to stratify patients with DKD and evaluate the possibility of clinical application across centers.

Materials and Methods

The research participants were enrolled retrospectively and separated into three parts: training, validation, and independent test datasets for further analysis. DeepLabV3+ network, PyRadiomics package, and least absolute shrinkage and selection operator were used for segmentation, extraction of radiomics variables, and regression, respectively.

Results

A total of 499 patients from three centers were enrolled in this study including 246 patients with type II diabetes mellitus (T2DM) and 253 patients with DKD. The mean intersection-over-union (Miou) and mean pixel accuracy (mPA) of automatic segmentation of the data from the three medical centers were 0.812 ± 0.003, 0.781 ± 0.009, 0.805 ± 0.020 and 0.890 ± 0.004, 0.870 ± 0.002, 0.893 ± 0.007, respectively. The variables from the renal parenchyma and sinus provided different information for the diagnosis and follow-up of DKD. The area under the curve (AUC) of the radiomics model for differentiating between DKD and T2DM patients was 0.674 ± 0.074 and for differentiating between the high and low stages of DKD was 0.803 ± 0.037.

Conclusion

In this study, we developed a DL-based automatic segmentation, radiomics technology to stratify patients with DKD. The DL technology was proposed to achieve fast and accurate anatomical-level segmentation in the kidney, and an ultrasound-based radiomics model can achieve high diagnostic performance in the diagnosis and follow-up of patients with DKD.

Radiomics models for preoperative prediction of microvascular invasion in hepatocellular carcinoma: a systematic review and meta-analysis

PURPOSE

To assess the methodological quality and to evaluate the predictive performance of radiomics studies for preoperative prediction of microvascular invasion (MVI) in hepatocellular carcinoma (HCC).

METHODS

Publications between 2017 and 2021 on radiomic MVI prediction in HCC based on CT, MR, ultrasound, and PET/CT were included. The risk of bias was assessed using the prediction model risk of bias assessment tool (PROBAST). Methodological quality was assessed through the radiomics quality score (RQS). Fourteen studies classified as TRIPOD Type 2a or above were used for meta-analysis using random-effects model. Further analyses were performed to investigate the technical factors influencing the predictive performance of radiomics models.

RESULTS

Twenty-three studies including 4947 patients were included. The risk of bias was mainly related to analysis domain. The RQS reached an average of (37.7 ± 11.4)% with main methodological insufficiencies of scientific study design, external validation, and open science. The pooled areas under the receiver operating curve (AUC) were 0.85 (95% CI 0.82-0.89), 0.87 (95% CI 0.83-0.92), and 0.74 (95% CI 0.67-0.80), respectively, for CT, MR, and ultrasound radiomics models. The pooled AUC of ultrasound radiomics model was significantly lower than that of CT (p = 0.002) and MR (p < 0.001). Portal venous phase for CT and hepatobiliary phase for MR were superior to other imaging sequences for radiomic MVI prediction. Segmentation of both tumor and peritumor regions showed better performance than tumor region.

CONCLUSION

Radiomics models show promising prediction performance for predicting MVI in HCC. However, improvements in standardization of methodology are required for feasibility confirmation and clinical translation.