Interpretable artificial intelligence-based app assists inexperienced radiologists in diagnosing biliary atresia from sonographic gallbladder images.

BACKGROUND: A previously trained deep learning-based smartphone app provides an artificial intelligence solution to help diagnose biliary atresia from sonographic gallbladder images, but it might be impractical to launch it in real clinical settings. This study aimed to redevelop a new model using original sonographic images and their derived smartphone photos and then test the new model's performance in assisting radiologists with different experiences to detect biliary atresia in real-world mimic settings. METHODS: A new model was first trained retrospectively using 3659 original sonographic gallbladder images and their derived 51,226 smartphone photos and tested on 11,410 external validation smartphone photos. Afterward, the new model was tested in 333 prospectively collected sonographic gallbladder videos from 207 infants by 14 inexperienced radiologists (9 juniors and 5 seniors) and 4 experienced pediatric radiologists in real-world mimic settings. Diagnostic performance was expressed as the area under the receiver operating characteristic curve (AUC). RESULTS: The new model outperformed the previously published model in diagnosing BA on the external validation set (AUC 0.924 vs 0.908, P = 0.004) with higher consistency (kappa value 0.708 vs 0.609). When tested in real-world mimic settings using 333 sonographic gallbladder videos, the new model performed comparable to experienced pediatric radiologists (average AUC 0.860 vs 0.876) and outperformed junior radiologists (average AUC 0.838 vs 0.773) and senior radiologists (average AUC 0.829 vs 0.749). Furthermore, the new model could aid both junior and senior radiologists to improve their diagnostic performances, with the average AUC increasing from 0.773 to 0.835 for junior radiologists and from 0.749 to 0.805 for senior radiologists. CONCLUSIONS: The interpretable app-based model showed robust and satisfactory performance in diagnosing biliary atresia, and it could aid radiologists with limited experiences to improve their diagnostic performances in real-world mimic settings.

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).

Three-dimensional ultrasound fusion imaging in precise needle placement for thermal ablation of hepatocellular carcinoma.

PURPOSE: To investigate the value of three-dimensional ultrasound fusion imaging (3DUS FI) technique for guiding needle placement in hepatocellular carcinoma (HCC) thermal ablation. METHODS: A total of 57 patients with 60 HCCs with 3DUS FI-guided thermal ablation were retrospectively included in the study. 3DUS volume data of liver were acquired preoperatively by freehand scanning with the tumor and predetermined 5 mm ablative margin automatically segmented. Plan of needle placement was made through a predetermined simulated ablation zone to ensure a 5 mm ablative margin with the coverage rate toward tumor and ablative margin. With real-time ultrasound and 3DUS fusion imaging, ablation needles were placed according to the plan. After ablation, the ablative margin was immediately evaluated by contrast-enhanced ultrasound and 3DUS fusion imaging. The rate of adequate ablative margin, complete response (CR), local tumor progression (LTP), disease-free survival (DFS), and overall survival (OS) was evaluated. RESULTS: According to postoperative contrast-enhanced CT or MR imaging, the complete response rate was 100% (60/60), and 83% of tumors (30/36) achieved adequate ablative margin (>5 mm) three-dimensionally. During the follow-up period of 6.0-42.6 months, LTP occurred in 5 lesions, with 1- and 2-year LTP rates being 7.0% and 9.4%. The 1- and 2-year DFS rates were 76.1% and 65.6%, and 1- and 2-year OS rates were 98.1% and 94.0%. No major complications or ablation-related deaths were observed in any patients. CONCLUSIONS: Three-dimensional ultrasound fusion imaging technique may improve the needle placement of thermal ablation for HCC and reduce the rate of LTP.

Comparison of non-radiomics imaging features and radiomics models based on contrast-enhanced ultrasound and Gd-EOB-DTPA-enhanced MRI for predicting microvascular invasion in hepatocellular carcinoma within 5 cm.

OBJECTIVES: The purpose of this study is to establish microvascular invasion (MVI) prediction models based on preoperative contrast-enhanced ultrasound (CEUS) and ethoxybenzyl-enhanced magnetic resonance imaging (EOB-MRI) in patients with a single hepatocellular carcinoma (HCC) ≤ 5 cm. METHODS: Patients with a single HCC ≤ 5 cm and accepting CEUS and EOB-MRI before surgery were enrolled in this study. Totally, 85 patients were randomly divided into the training and validation cohorts in a ratio of 7:3. Non-radiomics imaging features, the CEUS and EOB-MRI radiomics scores were extracted from the arterial phase, portal phase and delayed phase images of CEUS and the hepatobiliary phase images of EOB-MRI. Different MVI predicting models based on CEUS and EOB-MRI were constructed and their predictive values were evaluated. RESULTS: Since univariate analysis revealed that arterial peritumoral enhancement on the CEUS image, CEUS radiomics score, and EOB-MRI radiomics score were significantly associated with MVI, three prediction models, namely the CEUS model, the EOB-MRI model, and the CEUS-EOB model, were developed. In the validation cohort, the areas under the receiver operating characteristic curve of the CEUS model, the EOB-MRI model, and the CEUS-EOB model were 0.73, 0.79, and 0.86, respectively. CONCLUSIONS: Radiomics scores based on CEUS and EOB-MRI, combined with arterial peritumoral enhancement on CEUS, show a satisfying performance of MVI predicting. There was no significant difference in the efficacy of MVI risk evaluation between radiomics models based on CEUS and EOB-MRI in patients with a single HCC ≤ 5 cm. CLINICAL RELEVANCE STATEMENT: Radiomics models based on CEUS and EOB-MRI are effective for MVI predicting and conducive to pretreatment decision-making in patients with a single HCC within 5 cm. KEY POINTS: • Radiomics scores based on CEUS and EOB-MRI, combined with arterial peritumoral enhancement on CEUS, show a satisfying performance of MVI predicting. • There was no significant difference in the efficacy of MVI risk evaluation between radiomics models based on CEUS and EOB-MRI in patients with a single HCC ≤ 5 cm.

Development and validation of a combined nomogram model based on deep learning contrast-enhanced ultrasound and clinical factors to predict preoperative aggressiveness in pancreatic neuroendocrine neoplasms.

OBJECTIVES: This study aimed to develop and validate a combined nomogram model based on deep learning (DL) contrast-enhanced ultrasound (CEUS) and clinical factors to preoperatively predict the aggressiveness of pancreatic neuroendocrine neoplasms (PNENs). METHODS: In this retrospective study, consecutive patients with histologically proven PNENs underwent CEUS examination at the initial work-up between January 2010 and October 2020. Patients were randomly allocated to the training and test sets. Typical sonographic and enhanced images of PNENs were selected to fine-tune the SE-ResNeXt-50 network. A combined nomogram model was developed by incorporating the DL predictive probability with clinical factors using multivariate logistic regression analysis. The utility of the proposed model was evaluated using receiver operator characteristic, calibration, and decision curve analysis. RESULTS: A total of 104 patients were evaluated, including 80 (mean age ± standard deviation, 47 years ± 12; 56 males) in the training set and 24 (50 years ± 12; 14 males) in the test set. The DL model displayed effective image recognition with an AUC of 0.81 (95%CI: 0.62-1.00) in the test set. The combined nomogram model that incorporated independent clinical risk factors, such as tumor size, arterial enhancement level, and DL predictive probability, showed strong discrimination, with an AUC of 0.85 (95%CI: 0.69-1.00) in the test set with good calibration. Decision curve analysis verified the clinical usefulness of the combined nomogram. CONCLUSIONS: The combined nomogram model could serve as a preoperative, noninvasive, and precise evaluation tool to differentiate aggressive and non-aggressive PNENs. KEY POINTS: • Tumor size (odds ratio [OR], 1.58; p = 0.02), arterial enhancement level (OR, 0.04; p = 0.008), and deep learning predictive probability (OR, 288.46; p < 0.001) independently predicted aggressiveness of pancreatic neuroendocrine neoplasms preoperatively. • The combined model predicted aggressiveness better than the clinical model (AUC: 0.97 vs. 0.87, p = 0.009), achieving AUC values of 0.97 and 0.85 in the training set and the test set, respectively.

Ultrasound-guided percutaneous radiofrequency ablation in treatment of neuroendocrine tumor liver metastases：a single-center experience.

PURPOSE: To investigate the local tumor control of ultrasound-guided percutaneous radiofrequency ablation (RFA) in treating neuroendocrine tumor liver metastases (NETLM). MATERIALS AND METHODS: From March 2011 to December 2020, 23 patients with 39 NETLM treated with percutaneous RFA were studied. The study assessed the therapeutic outcomes after RFA, including the rates of technical success, technical efficacy, major complications, local tumor progression (LTP) and overall survival. Cumulative LTP rates were estimated with the Kaplan-Meier method. RESULTS: The technical efficacy rate was 91.3% (21/23) at one month after RFA. No major complication occurred in the treatment. LTP occurred in 50% (11/22) of patients who had complete ablation, with a median progression-free survival time of 15 months (1-61 months). Patients with Ki-67 < 5% had a longer progression-free survival than those with Ki-67 ≥ 5% (22.0 vs. 3.5 months, p<.001). Four patients, who had sporadic recurrent liver metastases, received 1-6 times of additional RFA after the initial treatment. Twenty patients were alive at the end of this study, except three patients withdrawn. The overall survival was at a median of 48 months (9-182 months). CONCLUSION: RFA provides effective local tumor control with low morbidity and it can be applied repeatedly over years to control recurrence of NETLM. Patients with Ki-67 < 5% have better local tumor control with percutaneous RFA of NETLM.

The usefulness of three-dimensional ultrasound fusion imaging for precise needle placement in liver thermal ablation: a phantom and an in vivo simulation study.

PURPOSE: To investigate the value of three-dimensional ultrasound fusion imaging (3DUS-FI) in real-time guiding needle placement by phantom models and in vivo simulations. MATERIALS AND METHODS: Two radiologists (beginner and expert) performed needle placement using two-dimensional ultrasound (2DUS) and 3DUS-FI, respectively. In the phantom study, single-needle placement was performed by puncturing the center point of each ball and assessed based on the specimen length. Multiple-needles placement was performed by placing three needles in each ball, and their locations were confirmed by computed tomography, and assessed based on the distance deviation between needles. In the in vivo simulation study, simulated-needle placement was performed by placing a virtual ablation needle in each liver tumor and assessed by the simulated ablative cover rate and margin. RESULTS: Specimen length was significantly longer with 3DUS-FI in the beginner, whereas no significant difference was observed in the expert (2DUS vs. 3DUS-FI: beginner, 14.60 ± 2.60 mm vs. 16.25 ± 1.38 mm, p = .017; expert, 16.78 ± 1.40 mm vs. 16.95 ± 1.15 mm, p = .668). Distance deviation between needles was significantly smaller with 3DUS-FI (2DUS vs. 3DUS-FI: beginner, 25.06 ± 16.07 mm vs. 3.72 ± 1.99 mm, p < .001; expert, 11.70 ± 7.79 mm vs. 2.89 ± 1.52 mm, p < .001). The simulated ablative cover rate and margin were significantly larger with 3DUS-FI for the beginner, whereas only the latter was significantly larger for the expert (2DUS vs. 3DUS-FI: beginner, 73.55 ± 8.73% vs. 81.38 ± 11.84%, p = .001, 0.82 ± 0.97 mm vs. 2.65 ± 1.23 mm, p < .001; expert, 78.60 ± 9.91% vs. 83.24 ± 11.69%, p = .059; 1.65 ± 1.15 mm vs. 2.95 ± 1.13 mm, p < .001). CONCLUSIONS: 3DUS-FI is useful for real-time guiding precise needle placement and may be further use to improve the efficacy of liver thermal ablation.

The value of clinical-ultrasonographic feature model to predict the severity of secondary hyperparathyroidism.

OBJECTIVES: To analyze conventional ultrasound (CUS) and contrast-enhanced ultrasound (CEUS) features in patients with secondary hyperparathyroidism (SHPT) and to evaluate the clinical-ultrasonographic feature based model for predicting the severity of SHPT. METHODS: From February 2016 to March 2021, a total of 59 patients (age 51.3 ± 11.7 years, seCr 797.8 ± 431.7 µmol/L, iPTH 1535.1 ± 1063.9 ng/L) with SHPT (including 181 parathyroid glands (PTGs)) without the history of intact parathyroid hormone (iPTH)-reducing drugs using were enrolled. The patients were divided into the mild SHPT group (mSHPT, iPTH <800 ng/L) and the severe SHPT group (sSHPT, iPTH ≥ 800 ng/L) according to the serum iPTH level. The clinical test data of patients were collected and CUS and CEUS examinations were performed for every patient. Multivariable logistic regression model according to clinical-ultrasonographic features was adopted to establish a nomogram. We performed K-fold cross-validation on this nomogram model and nomogram performance was determined by its discrimination, calibration, and clinical usefulness. RESULTS: There were 19 patients in the mSHPT group and 40 patients in the sSHPT group. Multivariable logistic regression indicated serum calcium, serum phosphorus and total volume of PTGs were independent predictors related with serum iPTH level. Even though CEUS score of wash-in and wash-out were showed related to severity of SHPT in univariate logistic regression analysis, they were not predictors of SHPT severity (p = 0.539, 0.474 respectively). The nomogram developed by clinical and ultrasonographic features showed good calibration and discrimination. The accuracy and the area under the curve (AUC), positive predictive value (PPV), negative predictive value (NPV) and accuracy of this model were 0.888, 92.5%, 63.2% and 83.1%, respectively. When applied to internal validation, the score revealed good discrimination with stratified fivefold cross-validation in the cohort (mean AUC = 0.833). CONCLUSIONS: The clinical-ultrasonographic features model has good performance for predicting the severity of SHPT.

Ultrasound characteristics combined with gamma-glutamyl transpeptidase for diagnosis of biliary atresia in infants less than 30 days.

PURPOSE: To retrospectively assess the diagnostic performance of grey-scale ultrasound (US) characteristics and gamma-glutamyl transpeptidase (GGT) alone or combined in distinguishing biliary atresia (BA) from other cholestasis diseases in infants younger than 30 days. MATERIALS AND METHODS: Between January 2012 and October 2020, the demographic characteristics, laboratory results and US characteristics of 35 BA and 52 non-BA infants younger than 30 days were retrospectively evaluated. Areas under the receiver operating characteristic curves (AUCs) were used to estimate the probability of predicting BA, which were compared by DeLong test. RESULTS: The diagnostic performance of gallbladder classification in identifying BA was higher than that of fibrotic cord thickness (AUC 0.900 vs. 0.771, P = 0.03). With the cutoff level of 188 IU/L, serum GGT had a sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of 77.1%, 69.2%, 62.8%, and 81.8%, respectively. Combined with gallbladder classification and GGT, the sensitivity, specificity, PPV, NPV and accuracy were 100.0%, 63.5%, 64.8%, 100.0% and 78.2%. CONCLUSIONS: Gallbladder classification was more valuable than fibrotic cord thickness in the diagnosis of BA among infants less than 30 days. Combined with gallbladder classification and GGT, the sensitivity for the diagnosis of BA can reach 100.0%.

The combination of conventional ultrasound and shear-wave elastography in evaluating the segmental heterogeneity of liver fibrosis in biliary atresia patients after Kasai portoenterostomy.

OBJECTIVE: To retrospectively assess the value of the combination of conventional ultrasound and shear-wave elastography (SWE) in evaluating the segmental heterogeneity of liver fibrosis in biliary atresia (BA) patients after Kasai portoenterostomy. METHODS: A total of 35 BA patients with liver segmental deformation were enrolled. The segmental deformation was assessed by conventional ultrasound followed with SWE examinations for evaluating the liver stiffness. Liver biopsy was performed in 11 patients in the region of SWE measurement and liver fibrosis was assessed using the Metavir classification. Aminotransferase to platelet ratio index (APRI) was calculated for comparison. The correlations between serum biochemical tests with SWE values were evaluated. Spearman's rank coefficient test was performed to evaluate the correlation between variables. RESULTS: The SWE values of the biopsy segments had significant positive correlations with liver fibrosis severity (r = 0.828, p = 0.001), which was better than APRI (r = 0.366, p = 0.242). The levels of bilirubin and transaminase showed significant correlations with the SWE values at hypertrophic segments in all patients (r from 0.336 to 0.576, all p < 0.05). CONCLUSIONS: Awareness of the segmental heterogeneity of liver fibrosis evaluated by conventional ultrasound and SWE may assist in selecting an appropriate biopsy location and predicting postoperative surveillance for patients with BA.

Transarterial Chemoembolization Followed by Radiofrequency Ablation for Hepatocellular Carcinoma: Impact of the Time Interval between the Two Treatments on Outcome.

PURPOSE: To compare the efficacy of radiofrequency (RF) ablation after transarterial chemoembolization within or beyond 30 days for medium-large or multiple recurrent hepatocellular carcinomas (HCCs). MATERIALS AND METHODS: In this single-center retrospective study conducted from 2007 through 2015, 135 patients with a single recurrent HCC (>3 cm) or multiple (2-5 tumors) recurrent HCCs underwent transarterial chemoembolization plus RF ablation. A total of 62 patients underwent RF ablation after transarterial chemoembolization within 30 days (sequential group) and 73 patients underwent RF ablation after transarterial chemoembolization beyond 30 days (delayed group). Outcomes of interests included overall survival (OS), progression-free survival (PFS), and complete response (CR) rate. RESULTS: The median OS and PFS were 49.8 and 38.0 months for sequential group, and 31.0 and 11.6 months for the delayed group. The sequential group experienced significantly better OS (hazard ratio [HR]: 0.517; P = .002) and PFS (HR, 0.621; P = .021). Among patients with multiple tumors or a single tumor >5 cm, the sequential group still had significantly longer OS (P = .022; P = .018, respectively) and PFS (P = 0.042; P = .036, respectively) than the delayed group, although no significant differences were observed among patients with solitary 3- to 5-cm tumors (P = .138; P = .803, respectively). The sequential group had a significantly better CR rate than the delayed group (85.4% vs. 68.5%, respectively; P = .035). Significant predictors of OS and PFS included maximum tumor size, number of tumors, and time interval between transarterial chemoembolization and RF ablation. CONCLUSIONS: Transarterial chemoembolization plus sequential RF ablation within 30 days was more effective for recurrent HCCs than transarterial chemoembolization plus delayed RF ablation. The time interval within 30 days is required for treating large or multiple HCCs but may not be necessary for solitary medium-sized HCC.

Three-dimensional contrast-enhanced ultrasound fusion imaging predicts local tumor progression by evaluating ablative margin of radiofrequency ablation for hepatocellular carcinoma: a preliminary report.

PURPOSE: To investigate the feasibility of three-dimensional contrast-enhanced ultrasound (3DCEUS) fusion in evaluating ablative margin (AM) after radiofrequency ablation (RFA) for hepatocellular carcinoma (HCC) and compare with those of computed tomography (CT) fusion. PATIENTS AND METHODS: In total, 55 patients (47 men, 8 women; mean age, 56.0 ± 10.6 years) with 55 HCCs were included. Pre- and post-RFA 3DCEUS images were fused to assess AM immediately after RFA. The AM measured on fusion imaging was categorized into two groups: (A) AM <5 mm, and (B) AM ≥5 mm. The agreement of AM evaluation was compared between 3DCEUS and CT fusion. The influence of AMs evaluated by 3DCEUS fusion on the risk of local tumor progression (LTP) was assessed. RESULTS: For 3DCEUS fusion, registration success rate was 96.4% (53/55). AM evaluation resulted in 28 tumors in group A and 25 tumors in group B. For CT fusion, AM evaluation resulted in 31 tumors with AM <5 mm and 23 tumors with AM ≥5 mm. AM was successful evaluated by both technique in 53 (96.4%) tumors. The agreement between 3DCEUS and CT fusion was excellent (kappa coefficient = 0.924, p < .001). LTP was identified in 5 tumors (9.4%, 5/53). LTPs all occurred in the area where unachieved 5 mm AM was evaluated by 3DCEUS fusion. Cumulative rate of LTP was significantly higher in group A than in group B (28% versus 0%; p = .033, log-rank test). CONCLUSIONS: 3DCEUS fusion enables AM evaluation for HCC immediately after RFA with comparable accuracy to CT fusion. AM <5 mm evaluated by 3DCEUS fusion has high predictive value for LTP.

Ultrasound Evaluation of Biliary Atresia Based on Gallbladder Classification: Is 4 Hours of Fasting Necessary?

OBJECTIVES: The aim of this study was to evaluate the influence of fasting time on the ultrasound identification and exclusion of biliary atresia in jaundiced infants through the use of the gallbladder classification scheme and to test the value of the gallbladder classification scheme in the diagnosis of biliary atresia in inexperienced individuals. METHODS: A total of 188 jaundiced infants were enrolled in this study. All patients received detailed abdominal sonograms. Diagnoses were confirmed via liver biopsy, surgical findings, or follow-up. Infants were grouped into either the fasting group (fasting time ≥ 4 h) or the nonfasting group (fasting time < 4 h). According to the gallbladder classification scheme, the gallbladders were classified into 4 types. The accuracy of diagnosing biliary atresia based on the gallbladder classification scheme was compared between the 2 groups. The gallbladders of 177 infants that were detectable by ultrasonography were used to compare the diagnostic performances between a junior radiologist and an experienced radiologist. RESULTS: The accuracies in the diagnosis of biliary atresia with the gallbladder classification scheme were 86.3% (82 of 95 patients) for the fasting group and 93.5% (87 of 93 patients) for the nonfasting group (P = .100). The diagnostic accuracies of biliary atresia were 88.7% (157 of 177 patients) for the junior radiologist and 90.4% (160 of 177 patients) for the experienced radiologist (P = .250). The κ value for the agreement between the 2 radiologists was 0.859. CONCLUSIONS: The performance of gallbladder diagnoses was not influenced by the fasting time with the use of the gallbladder classification scheme. Additionally, the gallbladder classification scheme may help junior radiologists to more effectively identify biliary atresia and nonbiliary atresia.

Application of Contrast-Enhanced Ultrasound in the Diagnosis of Solid Pseudopapillary Tumors of the Pancreas: Imaging Findings Compared With Contrast-Enhanced Computed Tomography.

OBJECTIVES: The purpose of this study was to assess the characterization and usefulness of contrast-enhanced ultrasound (CEUS) for diagnosing solid pseudopapillary tumors of the pancreas (SPTP) and compare the enhancement patterns with contrast-enhanced computed tomography (CECT). METHODS: Forty-three SPTP lesions proved by pathologic findings in 42 patients examined with CEUS and CECT were included in this study. The enhancement characteristics and typical CEUS features of the tumors were investigated. These characteristics were compared according to lesion sizes. The enhancement patterns of CEUS were compared with CECT. RESULTS: The most common enhancement levels of SPTP in the early phase and late phase for CEUS were isoenhancement (19 of 43 [44.2%]) and hypoenhancement (32 of 43 [74.4%]), respectively. The 4 most common enhancement patterns were hypo-hypo (16 of 43 [37.2%]), iso-iso (11 of 43 [25.6%]), hyper-hypo (8 of 43 [18.6%]), and iso-hypo (8 of 43 [18.6%]) enhancement. For the 43 SPTP lesions, typical CEUS features such as lesion membrane, intralesional vessel, and intralesional compartmentalization enhancements were detected in 30 (69.8%), 27 (62.8%), and 10 (23.2%) cases. Compared with CECT, isoenhancement during the early phase and hypoenhancement during the late phase were the most common imaging characteristics of CEUS. CONCLUSIONS: Lesion membrane, intralesional vessel, and intralesional compartmentalization enhancements are typical CEUS features of SPTP, especially for large lesions (lesion size ≥3.0 cm). Isoenhancement during the early phase and hypoenhancement during the late phase are the most common imaging characteristics of CEUS and CECT, making CEUS a viable alternative diagnostic method that is noninvasive.

Feasibility and Measurement Value of Pancreatic 2-D Shear Wave Elastography in Healthy Adults: Evaluation, Influencing Factors, Reference Range, Measurement Stability and Reproducibility.

OBJECTIVE: The present study was aimed at assessing the success rate and measurement value, determining the influencing factors and reference range and examining the intra-operator stability and inter-operator reproducibility of pancreatic 2-D shear wave elastography (SWE) measurement in healthy adults. METHODS: In 2022, 387 healthy adults were prospectively recruited. Logistic regression and linear regression analyses were used to explore the factors influencing the success rate and the measurement value of pancreatic 2-D SWE measurement, respectively. A two-sided 95% reference range was estimated accordingly. The intraclass correlation coefficient was calculated to evaluate the intra-operator stability and inter-operator reproducibility of the pancreatic 2-D SWE measurement. RESULTS: The pancreatic body (89.6%) bore the highest while the tail (72.8%) bore the lowest success rate of pancreatic 2-D SWE measurement. Sex and body mass index (BMI) were the independent factors influencing measurement success rate in all three parts of the pancreas. Mean measurement values (Emean) were not the same in the three parts of the pancreas of the same participant. BMI and image depth were the independent factors influencing Emean in the pancreatic body, while region of interest depth and BMI were the only independent factors influencing Emean in the pancreatic head and tail, respectively. The intra-operator stability of pancreatic 2-D SWE measurement was found to be excellent, whereas its inter-operator reproducibility was poor to good. CONCLUSION: Pancreatic 2-D SWE is a reliable technique for evaluating pancreatic stiffness in healthy adults, but its success rate and measurement value are influenced by multiple factors.

Quantitative Parameters of Contrast-Enhanced Ultrasound Predicting the Response to Combined Immune Checkpoint Inhibitor and Anti-angiogenesis Therapies for Unresectable Hepatocellular Carcinoma.

OBJECTIVE: The aim of the work described here was to explore the value of contrast-enhanced ultrasound (CEUS) quantitative parameters in predicting the response of combined immune checkpoint inhibitor (ICI) and anti-angiogenesis therapies for unresectable hepatocellular carcinoma (HCC). METHODS: Sixty-six HCC patients who underwent combined ICI and anti-angiogenesis therapies were prospectively enrolled. A CEUS examination was performed at baseline, and tumor perfusion parameters were obtained with perfusion quantification software. The differences in CEUS quantitative parameters between the responder and non-responder groups were compared, and the correlations between CEUS parameters and progression-free survival (PFS) was evaluated. RESULTS: The objective response rate (ORR) was 21.2%. The values of rising time (RT) ratio, time to peak ratio, fall time ratio, peak enhancement ratio, wash-in rate ratio, wash-in perfusion index ratio and wash-out rate ratio differed significantly differed between the responder and non-responder groups (all p values < 0.05). Multivariable logistic regression analysis revealed that the RT ratio was the only independent factor associated with the ORR (odds ratio = 0.007, 95% confidence interval: 0.000-0.307, p = 0.010). The median RT ratios of the responder and non-responder groups were 36.9 and 58.9, respectively (p = 0.006). The appropriate cutoff point of the RT ratio was 80.1, determined with the X-tile program. Survival analysis indicated high PFS for the patients with a lower RT ratio (high RT ratio vs. low RT ratio = 4.4 mo vs. not reached, p = 0.001). CONCLUSION: CEUS quantitative parameters may predict the efficacy of ICI and anti-angiogenesis combined therapies for HCC.

Development and Validation of an Explainable Machine Learning Model for Identification of Hyper-Functioning Parathyroid Glands from High-Frequency Ultrasonographic Images.

OBJECTIVE: To develop and validate a machine learning (ML) model based on high-frequency ultrasound (HFUS) images with the aim to identify the functional status of parathyroid glands (PTGs) in secondary hyper-parathyroidism (SHPT) patients. METHODS: This retrospective study enrolled 60 SHPT patients (27 female, 33 male; mean age: 51.2 years) with 184 PTGs detected from February 2016 to June 2022. All enrollments underwent single-photon emission computed tomography/computed tomography and contrast-enhanced ultrasound examinations. The PTGs were randomly divided into training (n = 147) and testing datasets (n = 37). Four effective ML classifiers were used and combined models incorporating multi-modal HFUS visual signs and radiomics features was constructed based on the optimal classifier. Model performance was compared in terms of discrimination, calibration and clinical utility. The Shapley additive explanation method was used to explain and visualize the main predictors of the optimal model. RESULTS: This model, using a random forest classifier algorithm, outperformed other classifiers. Based on optimal classifier features, the model constructed from ultrasound visual and ML features achieved a favorable performance in the prediction of hyper-functioning PTGs. Compared with the traditional visual model, the ultrasound-based ML model achieved significant (p = 0.03) improvement (area under the curve: 0.859 vs. 0.629) and higher sensitivity (100.0% vs. 94.1%) and accuracy (86.5% vs. 67.6%). Among the predictors attributed to model development, large size and high echogenic heterogeneity of PTGs in ultrasonographic images were more often associated with high risk of hyper-functioning PTGs. CONCLUSION: The ultrasound-based ML model for identifying hyper-functioning PTGs in SHPT patients showed good performance and interpretability using high-frequency ultrasonographic images, which may facilitate clinical management.

Percutaneous Ultrasound Cholangiography With Microbubbles in Children With Biliary Diseases.

ABSTRACT: The application of intracavity contrast-enhanced ultrasound in the evaluation of biliary disease has been confirmed valuable among pediatric population. This pictorial essay aims to demonstrate the role of percutaneous ultrasound cholangiography (PUSC) with microbubbles in the diagnosis of different pediatric biliary diseases in our center. The biliary system's morphologic characteristics in PUSC mode of neonatal hepatitis, biliary atresia, choledochal cysts, and biliary complications of hepatobiliary surgery are presented.

Development and validation of a nomogram model for predicting residue of partially cystic thyroid nodules after ultrasound-guided ethanol and thermal ablation.

Objectives: To develop and validate a nomogram model for predicting residue of partially cystic thyroid nodules (PCTNs) after ethanol and thermal ablation. Materials and Methods: From July 2015 to August 2022, a total of 97 patients (age 40.78 ± 12.61 years) with 107 treated benign PCTNs receiving ethanol and thermal ablation were enrolled. Pre-ablative laboratory test results and the ultrasound (US) and contrast-enhanced ultrasound (CEUS) features of lesions were collected. They were categorized into non-residue group and residue group according to the CEUS examination assessment after ablation. Univariate and multivariate logistic regression analysis were adopted to build a nomogram. The nomogram was validated by internal stratified fivefold cross-validation. The calibration, discrimination and clinical utility of the nomogram were investigated to assess the performance of the model. Results: Residue was reported in 30 out of 107nodules (28.0%). Multivariate logistic regression analysis revealed initial volume (OR=1.12, 95%CI 1.06-1.19) and presence of septum (OR=3.19, 95%CI 1.09-9.36) were predictors of residue of PCTNs. The nomogram developed by the above factors showed good calibration and discrimination. The area under the curve (AUC), sensitivity and specificity of this model were 0.832, 86.7% and 68.8%, respectively. When applied to internal validation, the model revealed good generalizability with stratified fivefold cross-validation in the cohort (mean AUC = 0.821). Conclusions: The nomogram model has good performance for predicting the residue of PCTNs undergoing ethanol and thermal ablation. This could play a role in the decision of treatment and follow-up in clinical practice.

Long-term outcomes of combined radiofrequency ablation and multipronged ethanol ablation for the treatment of unfavorable hepatocellular carcinoma

PURPOSE: To evaluate the local efficacy, safety, and long-term outcomes of combined radiofrequency ablation (RFA) and multipronged ethanol ablation (EA) in the treatment of unfavorable hepatocellular carcinoma (HCC) and to determine the prognostic factors for survival. METHODS: Between August 2009 and December 2017, 98 patients with 110 unfavorable HCC nodules who underwent combined RFA and multipronged EA were retrospectively enrolled in the study. Unfavorable HCC was defined as a medium (3.1-5.0 cm) or large (5.1-7.0 cm) HCC nodule, a tumor located at a high-risk site, or a perivascular tumor. The treatment response, overall survival (OS), and recurrence-free survival (RFS) were analyzed. The Kaplan-Meier method and Cox proportional hazards regression model were used to evaluate the prognostic factors. RESULTS: Complete ablation was obtained in 80.9% (89/110) of the tumors after initial treatment. Major complications were observed in 3 (3.1%) patients. The cumulative incidence of local tumor progression (LTP) was 23.5% at five years, and no variable was found to be an independent predictive factor for LTP. The five-year OS and RFS rates were 41.9% and 34.0%, respectively. Multivariate analysis showed that the serum alpha-fetoprotein level, tumor size, presence of residual tumor after ablation, and extrahepatic metastases were significant prognostic factors for OS (P = 0.023, P = 0.030, P = 0.001, and P = 0.010, respectively). Tumor type and the number of tumors were predictive factors for RFS (P = 0.029 and P = 0.001, respectively). A perivascular tumor was not an independent predictive factor for OS or RFS. CONCLUSION: Combined RFA and multipronged EA is a safe and effective treatment for unfavorable HCC, especially for perivascular tumors.