Search

Objective: This study aimed to explore the clinical value of ultrasound radiomics analysis in the diagnosis of cervical lymph node metastasis (CLNM) in patients with nasopharyngeal carcinoma (NPC). Methods: A total of 205 cases of NPC CLNM and 284 cases of benign lymphadenopathy with pathologic diagnosis were retrospectively included. Grayscale ultrasound (US) images of the largest section of every lymph node underwent feature extraction. Feature selection was done by maximum relevance minimum redundancy (mRMR) algorithm and multivariate logistic least absolute shrinkage and selection operator (LASSO) regression. Logistic regression models were developed based on clinical features, radiomics features, and the combination of those features. The AUCs of models were analyzed by DeLong's test. Results: In the clinical model, lymph nodes in the upper neck, larger long axis, and unclear hilus were significant factors for CLNM (p < 0.001). MRMR and LASSO regression selected 7 significant features for the radiomics model from the 386 radiomics features extracted. In the validation dataset, the AUC value was 0.838 (0.776–0.901) in the clinical model, 0.810 (0.739–0.881) in the radiomics model, and 0.880 (0.826–0.933) in the combined model. There was not a significant difference between the AUCs of clinical models and radiomics models in both datasets. DeLong's test revealed a significantly larger AUC in the combined model than in the clinical model in both training (p = 0.049) and validation datasets (p = 0.027). Conclusion: Ultrasound radiomics analysis has potential value in screening meaningful ultrasound features and improving the diagnostic efficiency of ultrasound in CLNM of patients with NPC. Key Points: • Radiomics analysis of gray-scale ultrasound images can be used to develop an effective radiomics model for the diagnosis of cervical lymph node metastasis in nasopharyngeal carcinoma patients. • Radiomics model combined with general ultrasound features performed better than the clinical model in differentiating cervical lymph node metastases from benign lymphadenopathy. [ABSTRACT FROM AUTHOR]

Quantitative three-dimensional molecular ultrasound is a promising technology for longitudinal imaging applications such as therapy monitoring; the risk profile is favorable compared to positron emission tomography and computed tomography. However, clinical translation of quantitative methods for this technology are limited in that they assume that tumor tissues are homogeneous, and often depend on contrast-destruction events that can produce unintended bioeffects. Here, we develop quantitative features (henceforth image features) that capture tumor spatial information, and that are extracted without contrast destruction. We compare these techniques with the contrast-destruction derived differential targeted enhancement parameter (dTE) in predicting response to therapy. We found thirty-three reproducible image features that predict response to antiangiogenic therapy, without the need for a contrast agent disruption pulse. Multiparametric analysis shows that several of these image features can differentiate treated versus control animals with comparable performance to post-destruction measurements, suggesting that these can potentially replace parameters such as the dTE. The highest performing pre-destruction image features showed strong linear correlations with conventional dTE parameters with less overall variance. Thus, our study suggests that image features obtained during the wash in of the molecular agent, pre-destruction, may replace conventional post-destruction image features or the dTE parameter. [ABSTRACT FROM AUTHOR]

Purpose: The Breast Imaging‐Reporting and Data System (BI‐RADS) for ultrasound imaging provides a widely used reporting schema for breast imaging. Previous studies have shown that in ultrasound imaging, 90% of BI‐RADS 4A tumors are benign lesions after biopsies. Unnecessary biopsy procedures can be avoided by accurate classification of BI‐RADS 4A tumors. However, the classification task is challenging and has not been fully investigated by existing studies. For benign and malignant tumors of BI‐RADS 4A, the appearances of intra‐class tumors are highly variable, the characteristics of inter‐class tumors is overall‐similar. Discriminative features need to be found to improve classification accuracy of BI‐RADS 4A tumors. Methods: In this study, we designed the network using the clinical features of BI‐RADS 4A tumors to improve the discrimination ability of network. The boundary information is embedded into the input of the network using the uncertainty. A fine‐grained data augmentation method is used to find discriminative features in tumor information embedded with boundary information. Two mathematical methods, voting‐based and variance‐based, are used to define the uncertainty of boundary, and the differences of these two definitions are compared in a classification network. Results: The dataset we used to evaluate our method had 1155 2D grayscale images. Each image represented a unique BI‐RADS 4A tumor. Among them, 248 tumors were proven to be malignant by biopsy, and the remaining 907 were benign. A weakly supervised data augmentation network (WS‐DAN) was used as the backbone classification network, which showed competitive performance in finding discriminative features. Using the auxiliary input of the uncertain boundaries defined by the voting method, the area under the curve (AUC) value of our method was 0.8347 (sensitivity = 0.7774, specificity = 0.7459). The AUC value of the variance‐based uncertainty was 0.7789. The voting‐based uncertainty was higher than the baseline (AUC = 0.803), which only inputs the original image. Compared with the classic classification network, our method had a significant effect improvement (p < 0.01). Conclusions: Using the uncertain boundaries defined by the voting methods as auxiliary information, we obtained a better performance in the classification of BI‐RADS 4A ultrasound images, while variance‐based uncertain boundaries had no effect on improving classification performance. Additionally, fine‐grained network helped find discriminative features comparing with the commonly used classification networks. [ABSTRACT FROM AUTHOR]

Quantitative ultrasound (QUS), which attempts to extract quantitative features from the US radiofrequency (RF) or envelope data for tissue characterization, is becoming a promising technique for noninvasive assessments of liver fibrosis. However, the number of feature variables examined and finally used in the existing QUS methods is typically small, limiting the diagnostic performance. Therefore, this paper devises a new multiparametric QUS (MP-QUS) method which enables the extraction of a large number of feature variables from US RF signals and allows for the use of feature-engineering and machine-learning based algorithms for liver fibrosis assessment. In the MP-QUS, eighty-four feature variables were extracted from multiple QUS parametric maps derived from the RF signals and the envelope data. Afterwards, feature reduction and selection were performed in turn to remove the feature redundancy and identify the best combination of features in the reduced feature set. Finally, a variety of machine-learning algorithms were tested for fibrosis classification with the selected features, based on the results of which the optimal classifier was established. The performance of the proposed MP-QUS method for staging liver fibrosis was evaluated on an animal model, with histologic examination as the reference standard. The mean accuracy, sensitivity, specificity and area under the receiver-operating-characteristic curve achieved by MP-QUS are respectively 83.38%, 86.04%, 80.82%, and 0.891 for recognizing significant liver fibrosis, and 85.50%, 88.92%, 85.24%, and 0.924 for diagnosing liver cirrhosis. The proposed MP-QUS method paves a way for its future extension to assess liver fibrosis in human subjects. [ABSTRACT FROM AUTHOR]

There is a need for noninvasive repeatable biomarkers to detect early cancer treatment response and spare non-responders unnecessary morbidities and costs. Here, we introduce three-dimensional (3D) dynamic contrast enhanced ultrasound (DCE-US) perfusion map characterization as inexpensive, bedside and longitudinal indicator of tumor perfusion for prediction of vascular changes and therapy response. More specifically, we developed computational tools to generate perfusion maps in 3D of tumor blood flow, and identified repeatable quantitative features to use in machine-learning models to capture subtle multi-parametric perfusion properties, including heterogeneity. Models were developed and trained in mice data and tested in a separate mouse cohort, as well as early validation clinical data consisting of patients receiving therapy for liver metastases. Models had excellent (ROC-AUC > 0.9) prediction of response in pre-clinical data, as well as proof-of-concept clinical data. Significant correlations with histological assessments of tumor vasculature were noted (Spearman R > 0.70) in pre-clinical data. Our approach can identify responders based on early perfusion changes, using perfusion properties correlated to gold-standard vascular properties. [ABSTRACT FROM AUTHOR]

Vesicoureteral reflux (VUR) is one of the most common urinary tract anomalies in children. It has been reported that VUR may be associated with reflux nephropathy. Ultrasound contrast-enhanced voiding urosonography (CeVUS) has become a routine diagnostic method for VUR in a number of European countries; however, it is not widely used in China. The aim of the present study was to analyze the clinical application and evaluate the safety of CeVUS as a diagnostic tool for VUR in children in order to establish a standardized operating procedure for CeVUS in pediatric VUR in China. Between August 2016 and October 2017, 90 children who were susceptible to VUR were admitted into the Pediatric Nephrology Department of Tongji Hospital and underwent CeVUS. The SonoVue second-generation USA contrast agent was administered intravesically via a transurethral bladder catheter at a dose of 1 ml. The occurrence of adverse events was monitored. Urine analysis and culture were performed. A total of 90 children (47 female, 43 male; mean age, 36.6 months) with 178 Pelvi-Ureteral Units (PUUs) underwent CeVUS to screen for VUR. VUR was detected in 44/90 pediatric patients (48.89%) and 65/178 PUUs (36.52%) by CeVUS. The grade distribution of the 65 PUUs with VUS was as follows: Grade I, 3; Grade II, 9; Grade III, 14; Grade IV, 22; and Grade V, 17. The accuracy of CeVUS in the present study were consistent with previous reports. No urethral anomalies were detected and there were no adverse events. CeVUS was demonstrated to be a safe, accurate and reliable imaging technique for detecting VUR in high-risk children, including neonates. Results of the present study indicated that CeVUS can be adopted as the primary screening and follow-up method for pediatric VUR diagnoses in China. [ABSTRACT FROM AUTHOR]

Objectives: To evaluate the feasibility and time saving of fusing CT and MR enterography with ultrasound for ultrasound molecular imaging (USMI) of inflammation in an acute small bowel inflammation of swine.Methods: Nine swine with ileitis were scanned with either CT (n = 3) or MR (n = 6) enterography. Imaging times to load CT/MR images onto a clinical ultrasound machine, fuse them to ultrasound with an anatomical landmark-based approach, and identify ileitis were compared to the imaging times without anatomical road mapping. Inflammation was then assessed by USMI using dual selectin-targeted (MBSelectin) and control (MBControl) contrast agents in diseased and healthy control bowel segments, followed by ex vivo histology.Results: Cross-sectional image fusion with ultrasound was feasible with an alignment error of 13.9 ± 9.7 mm. Anatomical road mapping significantly reduced (P < 0.001) scanning times by 40%. Localising ileitis was achieved within 1.0 min. Subsequently performed USMI demonstrated significantly (P < 0.001) higher imaging signal using MBSelectin compared to MBControl and histology confirmed a significantly higher inflammation score (P = 0.006) and P- and E-selectin expression (P ≤ 0.02) in inflamed vs. healthy bowel.Conclusions: Fusion of CT and MR enterography data sets with ultrasound in real time is feasible and allows rapid anatomical localisation of ileitis for subsequent quantification of inflammation using USMI.Key Points: • Real-time fusion of CT/MRI with ultrasound to localise ileitis is feasible. • Anatomical road mapping using CT/MRI significantly decreases the scanning time for USMI. • USMI allows quantification of inflammation in swine, verified with ex vivo histology. [ABSTRACT FROM AUTHOR]

The aim of this study was to assess the ability of the modified contrast-enhanced ultrasound (CEUS) Liver Imaging Reporting and Data System (LI-RADS) to distinguish malignancy in patients without known hepatocellular carcinoma (HCC) risk factors and compare diagnostic accuracy with that of the World Federation for Ultrasound in Medicine and Biology (WFUMB) guideline across radiologists with different levels of CEUS experience. A total of 848 individuals with no hepatitis infection presenting with 870 lesions in non-cirrhotic livers were included and divided into the Testing and Validation groups. The modified CEUS LI-RADS was proposed, including downgrading of focal nodular hyperplasia with typical features. Diagnostic performance of the modified CEUS LI-RADS was assessed in the Testing group. In the Validation group, two radiologists with more than 9 y of CEUS experience (Experts) and two radiologists with less than 6 mo of CEUS experience (Novices) used both the modified CEUS LI-RADS and the WFUMB guideline to evaluate performance in diagnosis of the lesions. LR-5 + M (combination of modified LR-5 and modified LR-M) revealed optimal performance with a sensitivity, specificity and area under the curve (AUC) of 99.3%, 81.6% and 0.904, respectively. Novices using the modified CEUS LI-RADS outperformed those using the WFUMB guideline (AUC: 0.858 vs. 0.767, p = 0.005). Additionally, the sensitivity, specificity and AUC of Novices were comparable to those of Experts using the modified CEUS LI-RADS (94.1%, 77.6% and 0.858 vs. 96.1%, 77.6% and 0.868 for experts, respectively). The modified CEUS LI-RADS is a valuable method for distinguishing hepatic malignancy in patients without HCC risk factors. This is particularly beneficial for radiologists with limited CEUS expertise. [ABSTRACT FROM AUTHOR]

Ultrasonography is the most widely used technique to diagnose echinococcosis; however, challenges in using this technique and the demand on medical resources, especially in low-income or remote areas, can delay diagnosis. We aimed to develop a deep convolutional neural network (DCNN) model based on ultrasonography to identify echinococcosis and its types, especially alveolar echinococcosis. • This study used 9631 ultrasound images from patients assessed at 84 hospitals in China, obtained between Jan 1, 2002, and Dec 31, 2021. Patients with a diagnosis of cystic echinococcosis, alveolar echinococcosis, or seven other types of focal liver lesions were included. • We tested ResNet-50, ResNext-50, and VGG-16 as the backbone network architecture for a classification DCNN model and input the perinodular information from the ultrasound images. We trained and validated the DCNN model to diagnose and classify echinococcosis using still greyscale ultrasound images of focal liver lesions. • We then tested the algorithm on internal, external, and prospective test datasets. The performance of DCNN was also compared with that of 12 radiologists in a subset of ultrasound data that were randomly chosen from the prospective test dataset. Results The DCNN model was shown to be accurate and robust, and could improve the ultrasound diagnostic ability of radiologists for Echinococcosis and its types for highly endemic and remote regions. [ABSTRACT FROM AUTHOR]

To evaluate the diagnostic performance of computed tomography combined with ultrasound (CT/US) in metastatic central lymph nodes (CLNs) compared with US in patients with papillary thyroid cancer (PTC). Six-hundred patients with surgically proven PTC who underwent both US and CT examination before CLN dissection were evaluated retrospectively. All cases were divided into four subgroups according to the tumor size and number. Diagnostic performances of US, CT and CT/US were evaluated. Among 600 patients with CLN dissection, CT/US showed higher sensitivity (89.10%) and accuracy (83.00%) than US alone (76.06%, 76.50%). In the subgroup of solitary non-microcarcinomas, the AUC of CT/US was significantly higher than that for US alone (0.827 vs. 0.722, P < 0.05). For the subgroup of solitary or multiple microcarcinomas, the performance efficiency of CT had no obvious advantage over that of US (0.698 vs. 0.740, 0.798 vs. 0.802, P > 0.05). For the subgroup of multiple non-microcarcinomas, CT, US and US/CT had high diagnostic efficacy (AUC > 0.8, Accuracy >80%. P > 0.05), and there was no significant difference between them. In the subgroup of solitary non-microcarcinoma, CT combined with US provided better diagnostic efficacy, and for those cases, complementary CT examination was recommended. In other subgroups, the diagnostic efficacy of US/CT was similar to that of US alone, and there was no significant benefit from additional CT examination. • In total, the CT combined with ultrasound showed higher sensitivity and accuracy than US alone. • For the subgroup of solitary or multiple microcarcinomas, the performance efficiency of CT had no obvious advantage than US. • In the subgroup of solitary non-microcarcinomas, the performance efficiency of CT/US was significantly higher than US alone. • For the subgroup of multiple non-microcarcinomas, CT, US and US/CT had high diagnostic efficacy. [ABSTRACT FROM AUTHOR]

To assess the diagnostic performance and inter-observer agreement of the American College of Radiology (ACR) Ovarian-Adnexal Reporting and Data System Ultrasound (O-RADS US). From January 2016 to December 2018 a total of 1054 adnexal lesions in 1035 patients with pathologic results from two hospitals were retrospectively included. Each lesion was assigned to an O-RADS US category according to the criteria. Kappa (κ) statistics were applied to assess inter-observer agreement between a less experienced and an expert radiologist. Of the 1054 adnexal lesions, 750 were benign and 304 were malignant. The malignancy rates of O-RADS 5, O-RADS 4, O-RADS 3, and O-RADS 2 lesions were 89.57%, 34.46%, 1.10%, and 0.45% respectively. Area under the receiver operating characteristic curve was 0.960 (95% CI, 0.947–0.971). The optimal cutoff value for predicting malignancy was >O-RADS 3 with a sensitivity and specificity of 98.7% (95% CI, 0.964–0.996) and 83.2% (95% CI, 0.802–0.858) respectively. When sub-classifying multilocular cysts and smooth solid lesions in O-RADS 4 lesions as O-RADS 4a lesions and the rest cystic lesions with solid components as O-RADS 4b lesions, the malignancy rate were 17.02% and 42.57% respectively, which showed better risk stratification (P < 0.001). The inter-observer agreement between a less-experienced and an expert radiologist of O-RADS categorization was good (κ = 0.714). The ACR O-RADS US provides effective malignancy risk stratification for adnexal lesions with high reliability for radiologists with different experience. Sub-grouping of O-RADS 4 lesions into two groups facilitated better stratification of the intermediate risk. • The ACR O-RADS US has a good diagnostic performance in differentiating malignant from benign adnexal lesions. • The ACR O-RADS US is applicable to radiologists with different experience even freshman. • Explicit sub-classification into two groups among O-RADS 4 lesions showed better stratification of the intermediate risk. [ABSTRACT FROM AUTHOR]

The aim of our study was to compare strain elastography (SE), acoustic radiation force impulse-inducing Virtual Touch Imaging ([VTI] Siemens Medical Solutions, Mountain View, CA, USA), Virtual Touch Imaging Quantification ([VTIQ] Siemens Medical Solutions) and combined methods in the evaluation of ultrasound (US) Breast Imaging-Reporting and Data System (BI-RADS) category 4 lesions to explore an applicable way to reduce unnecessary biopsy by reducing false positives of conventional US without yielding false-negative cases. A total of 267 patients with 278 BI-RADS category 4 lesions (151 benign and 127 malignant) were evaluated with conventional B-mode US, SE, VTI and VTIQ implemented on a Siemens Acuson S2000 US system. Diagnostic performance, including area under the receiver operating characteristic curve, sensitivity, specificity, accuracy, positive predictive value (PPV) and negative predictive value (NPV) were evaluated. Overall, VTI alone exhibited the highest NPV (91.74%), although combined elastic methods exhibited higher NPV than single methods, with the highest NPV at 100% when the VTI, SE and VTIQ methods were combined. Compared with conventional US, PPV increased from 45.7% (127 of 278) to 63.18% (127 of 201) when adding combined elastography (VTI + SE +VTIQ). In addition, 52.5% (63/120) and 50.8% (61/120) of BI-RADS 4 A lesions were downgraded when using combined methods (VTI + SE and VTI + SE + VTIQ, respectively) without missing any cancer. However, 2 intraductal papillomas and 1 phyllodes tumor were not identified. In conclusion, the combination of different elastic methods have the potential to downgrade BI-RADS 4A lesions to reduce false-positive biopsies without increasing the risk of missing cancers. [ABSTRACT FROM AUTHOR]

The aim of this study was to evaluate whether quantitative analysis of elastic heterogeneity (EH) could improve the diagnostic performance of shear wave elastography (SWE) in breast lesions. From August 2016 to August 2017, 280 patients were enrolled in this prospective study. All lesions were evaluated with the ultrasound Breast Imaging Reporting and Data System (BI-RADS) and SWE with Virtual Touch tissue imaging quantification. The shear wave velocity (SWV) of the three areas of highest stiffness and lowest stiffness within the lesions were measured to calculate the maximum SWV (SWVmax), mean SWV (SWVmean) and EH. The EH was determined as the difference between the averaged highest SWV and lowest SWV. The diagnostic performance-including the area under the receiver operating characteristic curve (AUC) and the sensitivity and specificity of BI-RADS, EH, SWVmax and SWVmean-were analyzed. The AUC of EH, SWVmax and SWVmean were 0.963, 0.949 and 0.937, respectively. The sensitivity of EH was 93.75%, which was significantly higher than that of SWVmax (84.37%) and SWVmean (84.37%) (p < 0.001); there was no significant difference in the specificity among EH, SWVmax and SWVmean (p > 0.05). For category 4A lesions, EH predicted all the malignant lesions, while two cancers were misdiagnosed by SWVmax and SWVmean, respectively. Quantitative analysis of EH can improve the sensitivity of SWE for the differential diagnosis of breast lesions without loss of specificity. [ABSTRACT FROM AUTHOR]