OBJECTIVE: The objective of this study was to develop a combined model based on radiomics features of Sonazoid contrast-enhanced ultrasound (CEUS) during the Kupffer phase and to evaluate its value in differentiating well-differentiated hepatocellular carcinoma (w-HCC) from atypical benign focal liver lesions (FLLs).
METHODS: A total of 116 patients with preoperatively Sonazoid-CEUS confirmed w-HCC or benign FLL were selected from a prospective multiple study on the clinical application of Sonazoid in FLLs conducted from August 2020 to March 2021. According to the randomization principle, the patients were divided into a training cohort and a test cohort in a 7:3 ratio. Seventy-nine patients were used for establishing and training the radiomics model and combined model. In comparison, 37 patients were used for validating and comparing the performance of the models. The diagnostic efficacy of the models for w-HCC and atypical benign FLLs was evaluated using ROCs curves and decision curves. A combined model nomogram was created to assess its value in reducing unnecessary biopsies.
RESULTS: Among the patients, there were 55 cases of w-HCC and 61 cases of atypical benign FLLs, including 28 cases of early liver abscess, 16 cases of atypical hepatic hemangioma, 8 cases of hepatocellular dysplastic nodules (DN), and 9 cases of focal nodular hyperplasia (FNH). The radiomics model and combined model we established had AUCs of 0.905 and 0.951, respectively, in the training cohort, and the AUCs of the two models in the test cohort were 0.826 and 0.912, respectively. The combined model outperformed the radiomics feature model significantly. Decision curve analysis demonstrated that the combined model achieved a higher net benefit within a specific threshold probability range (0.25 to 1.00). A nomogram of the combined model was developed.
CONCLUSIONS: The combined model based on the radiomics features of Sonazoid-CEUS in the Kupffer phase showed satisfactory performance in diagnosing w-HCC and atypical benign FLLs. It can assist clinicians in timely detecting malignant FLLs and reducing unnecessary biopsies for benign diseases.

UNASSIGNED: The purpose of this study was to investigate the added value of color parameter imaging (CPI) in the differential diagnosis of focal liver lesions (FLLs) with \"homogeneous hyperenhancement but not wash out\" on contrast-enhanced ultrasound (CEUS).
UNASSIGNED: A total of 101 patients with 108 FLLs were enrolled in this study. All the FLLs received US and CEUS examinations. The stored CEUS clips of target lesions were postprocessed with CPI analysis by radiologists. The receiver operator characteristic (ROC) curve was used to evaluate the added value of CPI. The McNamara test was used to compare the diagnostic sensitivity, specificity, and accuracy between CEUS and CPI patterns. Univariate and multivariate logistic regression analyses were used to develop a CPI nomogram. The C index and calibration curve were used to evaluate the predictive ability of the nomogram. The intraclass correlation coefficient was used to test the reproducibility and reliability of CPI. Decision curve analysis (DCA) was used to evaluate the added value of applying CPI.
UNASSIGNED: The following CPI features were more frequently observed in malignant FLLs: eccentric perfusion (malignant: 70.0% vs. benign: 29.2%, p < 0.001), feeding artery (51.7% vs. 4.2%, p < 0.001), mosaic (63.3% vs. 6.3%, p < 0.001), red ingredients >1/3 (90.0% vs. 14.6%, p < 0.001). In addition, centripetal (43.8% vs. 18.3%, p = 0.004), peripheral nodular (54.2% vs. 1.7%, p < 0.001), subcapsular vessel (12.5% vs. 0.0%, p = 0.004), spoke-wheel vessels (25.0% vs. 5.0%, p = 0.003), branched vessels (22.9% vs. 5.0%, p = 0.006), blue and pink ingredients >2/3 (85.4% vs. 10.0%, p < 0.001) were more observed in benign FLLs. A nomogram incorporating peripheral nodular, spoke-wheel vessels, and red ingredients >1/3 was constructed. The model had satisfactory discrimination (AUC = 0.937), and the optimal diagnostic threshold value was 0.740 (0.983, 0.850). By the DCA, the model offered a net benefit over the treat-all-patients scheme or the treat-none scheme at a threshold probability 5%-93%.
UNASSIGNED: Using CPI can detect and render subtle information of the main features of FLLs on CEUS; it is conducive to the radiologist for imaging interpretation, and a combining read of the CEUS and CPI of the FLLs with features of \"homogenous hyperenhancement and no washout\" can improve significantly the diagnostic performance of CEUS for FLLs.

UNASSIGNED: This article aims to clarify pitfalls and find strategies for the detecting and diagnosing hyperechoic liver metastases (LMs) using Sonazoid-contrast enhanced ultrasonography (Sonazoid-CEUS).
UNASSIGNED: This study was a prospective self-controlled study. Patients with hepatic lesions suspected as LMs or benign lesions were included in the study. Baseline ultrasonography (BUS) and Sonazoid-CEUS were performed on every patient. Characteristics of LMs and benign nodules were compared by chi-square test and fisher test. Factors influenced the CEUS were demonstrated by univariate analysis and multivariate logistic regression analysis.
UNASSIGNED: 54 patients were included in this study. CEUS found additional 75 LMs from 19 patients in Kupffer phase. We found hyperechoic focal liver lesions and deep seated in liver are main confounding factors in CEUS diagnosis. Sensitivity would be improved from 16.67% to 78.57%, negative predictive value (NPV) would be improved from 28.57% to 76.92% and accuracy would be improved from 37.5% to 87.50% when using rapid \"wash-in\" and \"wash-out\" as main diagnostic criteria.
UNASSIGNED: Hyperechoic LMs especially deeply seated ones are usually not shown typical \"black hole\" sign in Kupffer phase. Quickly \"wash-in and wash out\" shows high accuracy in diagnosing malignant nodules. We highly recommend CEUS as a routing exam to detect and diagnose LMs.

OBJECTIVE: To establish shear-wave elastography (SWE) combined with contrast-enhanced ultrasound (CEUS) algorithm (SCCA) and improve the diagnostic performance in differentiating focal liver lesions (FLLs).
METHODS: We retrospectively selected patients with FLLs between January 2018 and December 2019 at the First Affiliated Hospital of Sun Yat-sen University. Histopathology was used as a standard criterion except for hemangiomas and focal nodular hyperplasia. CEUS with SonoVue (Bracco Imaging) and SCCA combining CEUS and maximum value of elastography with < 20 kPa and > 90 kPa thresholds were used for the diagnosis of FLLs. The diagnostic performance of CEUS and SCCA was calculated and compared.
RESULTS: A total of 171 FLLs were included, with 124 malignant FLLs and 47 benign FLLs. The area under curve (AUC), sensitivity, and specificity in detecting malignant FLLs were 0.83, 91.94%, and 74.47% for CEUS, respectively, and 0.89, 91.94%, and 85.11% for SCCA, respectively. The AUC of SCCA was significantly higher than that of CEUS (P = 0.019). Decision curves indicated that SCCA provided greater clinical benefits. The SCCA provided significantly improved prediction of clinical outcomes, with a net reclassification improvement index of 10.64% (P = 0.018) and integrated discrimination improvement of 0.106 (P = 0.019). For subgroup analysis, we divided the FLLs into a chronic-liver-disease group (n = 88 FLLs) and a normal-liver group (n = 83 FLLs) according to the liver background. In the chronic-liver-disease group, there were no differences between the CEUS-based and SCCA diagnoses. In the normal-liver group, the AUC of SCCA and CEUS in the characterization of FLLs were 0.89 and 0.83, respectively (P = 0.018).
CONCLUSIONS: SCCA is a feasible tool for differentiating FLLs in patients with normal liver backgrounds. Further investigations are necessary to validate the universality of this algorithm.

There are many types of focal liver lesions (FLL) presenting different lesion signs and their diagnosis and differential diagnosis are relatively difficult. It is of great clinical significance to accurately detect, classify and characterize focal liver lesions as soon as possible. Diffusion-weighted imaging (DWI) provides information on liver cell density, microstructure, and microcirculation perfusion. Gadolinium-ethoxibenzyl-diethylenetriamine pentaacetic acid (Gd-EOB-DTPA) is a hepatobiliary-specific contrast agent. Gd-EOB-DTPA-enhanced MRI examination of liver provides information on the blood perfusion of lesions and specific information on the uptake function of normal liver cells. The combined application of the two can significantly improve the sensitivity and diagnostic accuracy in the detection of FLL. Herein, we reviewed the research findings on the application of DWI and Gd-EOB-DTPA in FLL diagnosis in order to provide reference for further clinical application. Most of the existing studies only made comparison and discussion of the DWI image quality of different b values and their fitted apparent diffusion coefficient (ADC) values before and after Gd-EOB-DTPA enhancement, and the reported findings are not only varied, but also inconsistent. Whether Gd-EOB-DTPA will affect DWI images is still been debated. Future research should focus on quantitative comparison, discussion and verification of the enhancement effect after injection of Gd-EOB-DTPA, as well as the changes in the ADC value corresponding to different b values before and after enhancement, in order to provide more objective and consistent research results for clinical application.

UNASSIGNED: First-line surveillance on hepatitis B virus (HBV)-infected populations with B-mode ultrasound is relatively limited to identifying hepatocellular carcinoma (HCC) without elevated α-fetoprotein (AFP). To improve the present HCC surveillance strategy, the state of the art of artificial intelligence (AI), a deep learning (DL) approach, is proposed to assist in the diagnosis of a focal liver lesion (FLL) in HBV-infected liver background.
UNASSIGNED: Our proposed deep learning model was based on B-mode ultrasound images of surgery that proved 209 HCC and 198 focal nodular hyperplasia (FNH) cases with 413 lesions. The model cohort and test cohort were set at a ratio of 3:1, in which the test cohort was composed of AFP-negative HBV-infected cases. Four additional deep learning models (MobileNet, Resnet50, DenseNet121, and InceptionV3) were also constructed as comparative baselines. To evaluate the models in terms of diagnostic power, sensitivity, specificity, accuracy, confusion matrix, F1-score, and area under the receiver operating characteristic curve (AUC) were calculated in the test cohort.
UNASSIGNED: The AUC of our model, Xception, achieved 93.68% in the test cohort, superior to other baselines (89.06%, 85.67%, 83.94%, and 78.13% respectively for MobileNet, Resnet50, DenseNet121, and InceptionV3). In terms of diagnostic power, our model showed sensitivity, specificity, accuracy, and F1-score of 96.08%, 76.92%, 86.41%, and 87.50%, respectively, and PPV, NPV, FPR, and FNR calculated from the confusion matrix were respectively 80.33%, 95.24%, 23.08%, and 3.92% in identifying AFP-negative HCC from HBV-infected FLL cases. Satisfactory robustness of our proposed model was shown based on 5-fold cross-validation performed among the models above.
UNASSIGNED: Our DL approach has great potential to assist B-mode ultrasound in identifying AFP-negative HCC from FLL found in surveillance of HBV-infected patients.

OBJECTIVE: This study aimed to evaluate the diagnostic performance of high frame rate contrast-enhanced ultrasound (H-CEUS) of focal liver lesions (FLLs).
METHODS: From July 2017 to June 2019, conventional contrast-enhanced ultrasound (C-CEUS) and H-CEUS were performed in 78 patients with 78 nodules. The characteristics of C-CEUS and H-CEUS in malignant and benign groups and the differences between different lesion sizes (1-3 cm, 3-5 cm, or >5 cm) of C-CEUS and H-CEUS were examined. The diagnostic performance of C-CEUS and H-CEUS was analyzed. The chi-square test or Fisher\'s exact test was used to assess inter-group differences. The receiver operating characteristic curve was plotted to determine the diagnostic performance of C-CEUS and H-CEUS.
RESULTS: There were significant differences in the enhancement area, fill-in direction and vascular architecture between C-CEUS and H-CEUS for both benign and malignant lesions (all p=0.000-0.008), but there were no significant differences in washout results (p=0.566 and p=0.684, respectively). For lesions 1-3 cm in size, the enhancement area, fill-in direction, and vascular architecture on C-CEUS and H-CEUS were significantly different (all p=0.000), unlike for lesions 3-5 cm or >5 cm in size. For differentiation of malignant from benign FLLs in the 1-3 cm group, H-CEUS showed sensitivity, specificity, accuracy, and positive and negative predictive values of 92.86%, 95.0%, 96.3%, 90.48% and 93.75%, respectively, which were higher than those for C-CEUS (75.0%, 70.0%, 77.78%, 66.67% and 72.91%, respectively).
CONCLUSIONS: H-CEUS provided more vascular information which could help differentiate malignant from benign FLLs, especially for lesions 1-3 cm in size.

OBJECTIVE: The detection and characterization of focal liver lesions (FLLs) in patients with cirrhosis is challenging. Accurate information about FLLs is key to their management, which can range from conservative methods to surgical excision. We sought to develop a nomogram that incorporates clinical risk factors, blood indicators, and enhanced computed tomography (CT) imaging findings to predict the nature of FLLs in cirrhotic livers.
METHODS: A total of 348 surgically confirmed FLLs were included. CT findings and clinical data were assessed. All factors with P < 0.05 in univariate analysis were included in multivariate analysis. ROC analysis was performed, and a nomogram was constructed based on the multivariate logistic regression analysis results.
RESULTS: The FLLs were either benign (n = 79) or malignant (n = 269). Logistic regression evaluated independent factors that positively affected malignancy. AFP (OR = 10.547), arterial phase hyperenhancement (APHE) (OR = 740.876), washout (OR = 0.028), satellite lesions (OR = 15.164), ascites (OR = 156.241), and nodule-in-nodule architecture (OR =27.401) were independent predictors of malignancy. The combined predictors had excellent performance in differentiating benign and malignant lesions, with an AUC of 0.959, a sensitivity of 95.24%, and a specificity of 87.5% in the training cohort and AUC of 0.981, sensitivity of 94.74%, and specificity of 93.33% in the test cohort. The C-index was 96.80%, and calibration curves showed good agreement between the nomogram predictions and the actual data.
CONCLUSIONS: The nomogram showed excellent discrimination and calibration for malignancy risk prediction, and it may aid in making FLLs treatment decisions.

BACKGROUND: The imaging features of focal liver lesions (FLLs) are diverse and complex. Diagnosing FLLs with imaging alone remains challenging. We developed and validated an interpretable deep learning model for the classification of seven categories of FLLs on multisequence MRI and compared the differential diagnosis between the proposed model and radiologists.
METHODS: In all, 557 lesions examined by multisequence MRI were utilised in this retrospective study and divided into training-validation (n = 444) and test (n = 113) datasets. The area under the receiver operating characteristic curve (AUC) was calculated to evaluate the performance of the model. The accuracy and confusion matrix of the model and individual radiologists were compared. Saliency maps were generated to highlight the activation region based on the model perspective.
RESULTS: The AUC of the two- and seven-way classifications of the model were 0.969 (95% CI 0.944-0.994) and from 0.919 (95% CI 0.857-0.980) to 0.999 (95% CI 0.996-1.000), respectively. The model accuracy (79.6%) of the seven-way classification was higher than that of the radiology residents (66.4%, p = 0.035) and general radiologists (73.5%, p = 0.346) but lower than that of the academic radiologists (85.4%, p = 0.291). Confusion matrices showed the sources of diagnostic errors for the model and individual radiologists for each disease. Saliency maps detected the activation regions associated with each predicted class.
CONCLUSIONS: This interpretable deep learning model showed high diagnostic performance in the differentiation of FLLs on multisequence MRI. The analysis principle contributing to the predictions can be explained via saliency maps.

BACKGROUND: R2* estimation reflects the paramagnetism of the tumor tissue, which may be used to differentiate between benign and malignant liver lesions when contrast agents are contraindicated.
OBJECTIVE: To investigate whether R2* derived from multi-echo Dixon imaging can aid differentiating benign from malignant focal liver lesions (FLLs) and the impact of 2D region of interest (2D-ROI) and volume of interest (VOI) on the outcomes.
METHODS: We retrospectively enrolled 73 patients with 108 benign or malignant FLLs. All patients underwent conventional abdominal magnetic resonance imaging and multi-echo Dixon imaging. Two radiologists independently measured the mean R2* values of lesions using 2D-ROI and VOI approaches. The Bland-Altman plot was used to determine the interobserver agreement between R2* measurements. Intraclass correlation coefficient (ICC) was used to determine the reliability between the two readers. Mean R2* values were compared between benign and malignant FFLs using the nonparametric Mann-Whitney test. Receiver operating characteristic curve analysis was used to determine the diagnostic performance of R2* in differentiation between benign and malignant FFLs. We compared the diagnostic performance of R2* measured by 2D-ROI and VOI approaches.
RESULTS: This study included 30 benign and 78 malignant FLLs. The interobserver reproducibility of R2* measurements was excellent for the 2D-ROI (ICC = 0.994) and VOI (ICC = 0.998) methods. Bland-Altman analysis also demonstrated excellent agreement. Mean R2* was significantly higher for malignant than benign FFLs as measured by 2D-ROI (P < 0.001) and VOI (P < 0.001). The area under the curve (AUC) of R2* measured by 2D-ROI was 0.884 at a cut-off of 25.2/s, with a sensitivity of 84.6% and specificity of 80.0% for differentiating benign from malignant FFLs. R2* measured by VOI yielded an AUC of 0.875 at a cut-off of 26.7/s in distinguishing benign from malignant FFLs, with a sensitivity of 85.9% and specificity of 76.7%. The AUCs of R2* were not significantly different between the 2D-ROI and VOI methods.
CONCLUSIONS: R2* derived from multi-echo Dixon imaging whether by 2D-ROI or VOI can aid in differentiation between benign and malignant FLLs.