Deep learning using contrast-enhanced ultrasound images to predict the nuclear grade of clear cell renal cell carcinoma.

PURPOSE: To assess the effectiveness of a deep learning model using contrastenhanced ultrasound (CEUS) images in distinguishing between low-grade (grade I and II) and high-grade (grade III and IV) clear cell renal cell carcinoma (ccRCC). METHODS: A retrospective study was conducted using CEUS images of 177 Fuhrmangraded ccRCCs (93 low-grade and 84 high-grade) from May 2017 to December 2020. A total of 6412 CEUS images were captured from the videos and normalized for subsequent analysis. A deep learning model using the RepVGG architecture was proposed to differentiate between low-grade and high-grade ccRCC. The model's performance was evaluated based on sensitivity, specificity, positive predictive value, negative predictive value and area under the receiver operating characteristic curve (AUC). Class activation mapping (CAM) was used to visualize the specific areas that contribute to the model's predictions. RESULTS: For discriminating high-grade ccRCC from low-grade, the deep learning model achieved a sensitivity of 74.8%, specificity of 79.1%, accuracy of 77.0%, and an AUC of 0.852 in the test set. CONCLUSION: The deep learning model based on CEUS images can accurately differentiate between low-grade and high-grade ccRCC in a non-invasive manner.

The kinetic quantitative characteristics of non-mass breast lesions with contrast-enhanced ultrasound: a prospective study.

OBJECTIVE: To characterize non-mass breast lesions (NML) quantitatively by contrast-enhanced ultrasound (CEUS) and to evaluate its additional diagnostic value based on the Breast Imaging Reporting and Data System (BI-RADS) categories. METHODS: A prospective study was performed among consecutive patients with NMLs. All lesions were examined by grayscale ultrasound and CEUS and diagnosed on pathology. Standard mammograms were obtained in the patients over 30 years old. Three independent radiologists assessed the features on grayscale ultrasound and mammograms and classified NMLs according to BI-RADS categories. Combined with the quantitative analysis in CEUS, the BI-RADS categories were reassessed, and the sensitivity, specificity, positive-predictive value, negative-predictive value and area under the receiver operating characteristic curve (AUC) were calculated for the evaluation of the diagnostic performance. RESULTS: 30 benign and 24 malignant NMLs were finally enrolled in this study, with ductal carcinoma in situ being the majority of malignant (15/24). Average contrast signal intensity (AI), wash-in rate (WiR) and enhancement intensity at 40 s (I40) were found to be the most efficient kinetic parameters to diagnose malignant NMLs. Combined with the cut-off values of 205.2 for AI, 127.8 for WiR and 136.4 for I40, the diagnostic accuracy was improved (AUC = 0.904), with the sensitivity of 95.8% and the specificity of 70.0%. CONCLUSION: The results suggested that hyperenhancement and rapid wash-in and wash-out are the characteristics of malignant NMLs. The kinetic analysis using CEUS can reflect hypervascular nature of malignant NMLs, thus improving the diagnostic performance combined with grayscale ultrasound. ADVANCES IN KNOWLEDGE: In this study, we quantified the enhancement characteristics of non-mass breast lesions with CEUS. We revealed that the combination of CEUS and conventional ultrasound provided higher sensitivity for diagnosing malignant NMLs.

CNN-based automatic segmentations and radiomics feature reliability on contrast-enhanced ultrasound images for renal tumors.

Objective: To investigate the feasibility and efficiency of automatic segmentation of contrast-enhanced ultrasound (CEUS) images in renal tumors by convolutional neural network (CNN) based models and their further application in radiomic analysis. Materials and methods: From 94 pathologically confirmed renal tumor cases, 3355 CEUS images were extracted and randomly divided into training set (3020 images) and test set (335 images). According to the histological subtypes of renal cell carcinoma, the test set was further split into clear cell renal cell carcinoma (ccRCC) set (225 images), renal angiomyolipoma (AML) set (77 images) and set of other subtypes (33 images). Manual segmentation was the gold standard and serves as ground truth. Seven CNN-based models including DeepLabV3+, UNet, UNet++, UNet3+, SegNet, MultilResUNet and Attention UNet were used for automatic segmentation. Python 3.7.0 and Pyradiomics package 3.0.1 were used for radiomic feature extraction. Performance of all approaches was evaluated by the metrics of mean intersection over union (mIOU), dice similarity coefficient (DSC), precision, and recall. Reliability and reproducibility of radiomics features were evaluated by the Pearson coefficient and the intraclass correlation coefficient (ICC). Results: All seven CNN-based models achieved good performance with the mIOU, DSC, precision and recall ranging between 81.97%-93.04%, 78.67%-92.70%, 93.92%-97.56%, and 85.29%-95.17%, respectively. The average Pearson coefficients ranged from 0.81 to 0.95, and the average ICCs ranged from 0.77 to 0.92. The UNet++ model showed the best performance with the mIOU, DSC, precision and recall of 93.04%, 92.70%, 97.43% and 95.17%, respectively. For ccRCC, AML and other subtypes, the reliability and reproducibility of radiomic analysis derived from automatically segmented CEUS images were excellent, with the average Pearson coefficients of 0.95, 0.96 and 0.96, and the average ICCs for different subtypes were 0.91, 0.93 and 0.94, respectively. Conclusion: This retrospective single-center study showed that the CNN-based models had good performance on automatic segmentation of CEUS images for renal tumors, especially the UNet++ model. The radiomics features extracted from automatically segmented CEUS images were feasible and reliable, and further validation by multi-center research is necessary.

Contrast-enhanced ultrasonography promotes differential diagnosis of ureteral neoplasms.

OBJECTIVE: To compare the diagnostic value of contrast-enhanced ultrasonography (CEUS) with baseline ultrasound (B-US) in ureteral neoplasms. METHODS: Retrospective analysis, comprising clinical presentation, image appearances, and diagnostic results on B-ultrasound and CEUS, considering pathological result as a gold-standard, was conducted on the clinical information of 39 patients with ureteral neoplasms. CT urography was used to detect and confirm the presence of ureteral neoplasms. Both B-ultrasound and CEUS investigations of those 39 patients under study were performed by a senior radiologist. RESULTS: Pathological outcomes established 27 ureteral malignancies and 12 ureteral benignancies. Ureteral malignancies were observed to occur in older patients than benignancies (p = 0.002). Only the morphological indicator of the ureteric wall on B-ultrasound was different in ureteral malignancies and benignancies (p = 0.030). Tumors with hyperenhancement, larger width, and hyperenhanced ureteric wall were easily diagnosed as malignant on CEUS, whereas iso-/hypoenhanced, narrower, and iso-/hypoenhanced ureteric wall indicated benign tumors. Moreover, the lesion widths, enhanced morphologies of the ureteric wall, and the ureteral wall's linear boundaries on CEUS were different between high- and low-stage ureteral urothelial carcinomas (p = 0.012, 0.002, 0.001, respectively). CONCLUSION: The display of microvessels in ureteral neoplasms was significantly enhanced by CEUS, thus contributing to the differential diagnosis of ureteral neoplasms while assisting the staging of ureteral urothelial carcinoma. ADVANCES IN KNOWLEDGE: The imaging features of different ureteral neoplasms on CEUS were analyzed in this study. The diagnostic performances of CEUS and B-ultrasound in ureteral urothelial carcinomas were also explored.

Predictive value of contrast-enhanced ultrasound combined with conventional ultrasound in solid renal parenchymal lesions.

OBJECTIVE: This study aimed to develop a model to predict the risk of malignancy in solid renal parenchymal lesions based on the imaging features of combined conventional and contrast-enhanced ultrasound (CEUS). METHODS: A retrospective review was performed among patients with focal solid renal parenchymal lesions on ultrasound images. Ultrasound features were characterized by two experienced radiologists independently. A multiple logistic regression analysis was performed to determine the most relevant features and to estimate the risk of malignancy. Scoring and counting methods were developed based on the most relevant features. The diagnostic performance was evaluated by the sensitivity, specificity, positive predictive value, negative predictive value and area under the receiver operating characteristic curve (AUC). RESULTS: A total of 519 renal lesions were included in this study. The conventional ultrasound features of diameter, echogenicity, hypoechoic rim and the CEUS feature of heterogeneity were identified as the most relevant features for prediction of malignancy. The sensitivity and specificity for the logistic regression model, the scoring method and the counting method were 95.3 and 93.4%, 93.8 and 87.8%, 88.8 and 93.9%, respectively. The logistic model had the best performance for diagnosing malignant renal lesions with AUC of 0.978, compared with the scoring method and the counting method with AUCs of 0.958 and 0.965. CONCLUSION: The combination of contrast-enhanced ultrasound with conventional ultrasound improved the diagnostic performance of solid renal lesions based on the logistic regression model. ADVANCES IN KNOWLEDGE: In this study, we revealed that the combination of CEUS and conventional ultrasound provided higher accuracy for diagnosing malignant renal tumors.

Diagnostic Performance of Ultrasound Shear Wave Elastography in Solid Small (≤4 cm) Renal Parenchymal Masses.

The aim of the study was to analyze the diagnostic performance of shear wave elastography (SWE) in differentiating between malignant and benign solid renal parenchymal masses ≤4 cm, compared with conventional ultrasound. A total of 20 healthy volunteers and 117 patients had been included in this study. Conventional ultrasound and SWE were performed in all volunteers and patients. The elasticity of healthy cortex and the elastic parameters of tumors such as mean elasticity (Emean), minimum elasticity (Emin), maximum elasticity (Emax), standard deviation and elasticity ratio of the lesion to surrounding cortex (Eratio) were measured on SWE images. Diagnostic performance of SWE was compared with that of conventional ultrasound. The cortical elasticity values of healthy right and left kidneys were 4.7 ± 1.7 and 4.5 ± 1.5 kPa, respectively. Of the 117 renal tumors, 68 were renal cell carcinomas (RCCs) and 49 were benign. Emean, Emin and Eratio were significantly lower in RCCs compared with benign lesions: Emean 7.2 ± 2.5 kPa versus 10.0 ± 2.4 kPa, Emin 2.5 ± 2.4 kPa versus 5.6 ± 2.3 kPa, Eratio 1.6 ± 0.5 versus 2.2 ± 0.6 (all p values < 0.001). The cutoff values of 9.15 kPa for Emean, 3.55 kPa for Emin and 1.99 for Eratio had the highest areas under the receiver operating characteristics curve (0.801 for Emean, 0.832 for Emin and 0.806 for Eratio). Combining Emean, Emin and Eratio with conventional ultrasound improved the specificity for predicting RCCs to 87.8%, but the sensitivity was not increased.

A combination of ultrasound-targeted microbubble destruction with transplantation of bone marrow mesenchymal stem cells promotes recovery of acute liver injury.

BACKGROUND: Bone marrow mesenchymal stem cells (BMSCs) can provide an additional source of therapeutic stem cells for regeneration of liver cells during acute liver injury (ALI). However, the insufficient hepatic homing by the transplanted BMSCs limits their applications. Ultrasound-targeted microbubble destruction (UTMD) has been reported to promote the homing of transplanted stem cells into the ischemic myocardium. In this study, we investigated whether UTMD promotes the hepatic homing of BMSCs in ALI rats and evaluated the therapeutic effect. METHODS: BMSCs were isolated from the femurs and tibias of Sprague-Dawley (SD) rats. The isolated BMSCs were stably transfected with a lentivirus expressing enhanced green fluorescent protein (EGFP) that can be visualized and quantified in vivo after transplantation. Both tumor necrosis factor α (TNF-α) and stromal cell-derived factor 1 (SDF-1) were used to verify the appropriate ultrasound parameters. The ALI rats were divided into four groups: control, BMSCs, UTMD, and UTMD + BMSCs. The protein and mRNA expression levels of SDF-1, intercellular cell adhesion molecule (ICAM-1), vascular cell adhesion molecule 1 (VCAM-1), hepatocyte growth factor (HGF), and monocyte chemotactic protein 1 (MCP-1) in the exposed livers were analyzed at 48 h after treatment. ALI recovery was determined by serum biochemical parameters and histology. RESULTS: The isolated rat BMSCs demonstrated a good proliferation potential that was both osteogenic and adipogenic in differentiation and expressed cluster of differentiation (CD) 29 and CD90, but not CD45 or CD11b/c. After BMSC and/or UTMD treatment, the number of GFP-labeled BMSCs in the UTMD + BMSCs group was significantly higher than that of the BMSCs group (9.8 ± 2.3 vs. 5.2 ± 1.1/per high-power field). Furthermore, the expression of GFP mRNA was performed for evaluation of the homing rate of BMSCs in injury sites as well. In addition, the expression levels of SDF-1, ICAM-1, VCAM-1, HGF, and MCP-1 were higher (p < 0.01) in UTMD+BMSCs group. The serum levels of biomarkers were significantly lower in the UTMD + BMSCs group, and the apoptotic rate of hepatocytes in the UTMD + BMSCs group was markedly lower than that of the BMSCs group (all p < 0.05). The hepatic pathology was significantly alleviated in the UTMD + BMSCs group. CONCLUSIONS: UTMD treatment efficiently induced a favorable microenvironment for cell engraftment, resulting in improvement of hepatic homing of BMSCs, which was probably mediated through upregulation of the expression of adhesion molecules and cytokines. UTMD treatment appeared to be an effective and noninvasive approach to achieve better efficacy of BMSC-based therapy for repairing a severely injured liver.

Ultrasound Irradiation Combined with Hepatocyte Growth Factor Accelerate the Hepatic Differentiation of Human Bone Marrow Mesenchymal Stem Cells.

This study investigated the impact of ultrasound (US) irradiation on the hepatic differentiation of human bone marrow mesenchymal stem cells (hBMSCs) induced by hepatocyte growth factor (HGF) and the possible mechanisms. We treated hBMSCs, using HGF with and without US irradiation. Cell viability and stem cell surface markers were analyzed. Hepatocyte-like cell markers and functional markers including α-fetoprotein (αFP/AFP), cytokeratin 18 (CK18), albumin (ALB) and glycogen content were analyzed at the time point of day 1, 3 and 5 after treatment. The involvement of Wnt/ß-catenin signaling pathway was evaluated as well. The results showed that the US treatment at 1.0 W/cm2 or 1.5 W/cm2 for 30 s or 60 s conditions yielded favorable cell viability and engendered stem cell differentiation. At day 5, the expressions of AFP, CK18, ALB and the glycogen content were significantly elevated in the US-treated group at both messenger ribonucleic acid and protein levels (all p <0.05), in comparison with HGF and control groups. Among all the US treated groups, the expression levels of specific hepatic markers in the (1.5 W/cm2 for 60 s) group were the highest. Furthermore, Wnt1, ß-Catenin, c-Myc and Cyclin D1 were significantly increased after US irradiation (all p <0.05), and the enhancements of c-Myc and Cyclin D1 could be obviously impaired by the inhibitor ICG-001 (p <0.05, p <0.05), in accordance with decreased ALB and CK18 expression and glycogen content (all p <0.05). In conclusion, US irradiation was able to promote the hBMSCs' differentiation mediated by HGF in vitro safely, easily and controllably. The activation of Wnt/ß-catenin signaling pathway was involved in this process. US irradiation could serve as a potentially beneficial tool for the research and application of stem cell differentiation.

Quantification of enhancement of renal parenchymal masses with contrast-enhanced ultrasound.

The purpose of this study was to investigate the value of quantitative assessment of enhancement in diagnosing renal cell carcinoma (RCC) with contrast-enhanced ultrasound (CEUS). A total of 73 solid renal parenchymal masses underwent both conventional ultrasound and CEUS. We compared the difference in maximum diameters on conventional ultrasound and CEUS between the benign and malignant groups. Enhancement features derived from a time-intensity curve were also analyzed. The diameters of renal cancer were found to be larger on CEUS than on conventional ultrasound (p < 0.05). When cutoff values of 4.74 s for washout time and 8.52% for enhancement intensity at 60 s for diagnosing RCCs were applied, the sensitivity, specificity and area under the receiver operating characteristic curve were 67.3%, 95.2%, 86.5% and 65.4%, 81.0%, 68.4%, respectively. The sensitivity and specificity for these two enhancement characteristics combined as a criterion for differentiating RCCs from benign lesions were 44.0% and 99.1%, respectively. Early washout in the area of maximal intensity in the interior of the lesion and prolonged washout in the whole area of the lesion are specific CEUS manifestations suggestive of RCC.

Evaluation of cervical cancer detection with acoustic radiation force impulse ultrasound imaging.

The aim of this study was to evaluate the application of acoustic radiation force impulse (ARFI) ultrasound imaging and its potential value in the characterization of cervical cancer. ARFI ultrasound imaging of the uterine cervix was performed in 58 patients with cervical cancer prior to surgery. The diagnosis of cervical cancer was confirmed by pathological results in each case. eSie Touch elastography imaging (EI), Virtual Touch tissue imaging (VTI) and Virtual Touch tissue quantification (VTQ; Siemens Medical Solutions, Mountain View, CA, USA) were used to qualitatively and quantitatively analyze the elasticity and hardness of lesions. For statistical analysis, the non-parametric Mann-Whitney U test and the Student's u test were used to compare the elastic parameters and the results. EI showed that, compared with the surrounding cervical tissue, 72.41% (42 of 58) of the malignant lesions showed 4th or 5th grade images and 27.59% (16 of 58) had 3rd grade images. The EI images showed a significant difference between the malignant lesions and the surrounding normal tissues (P<0.001). VTI showed that compared with the surrounding cervical tissue, 84.48% (49 of 58) of the malignant lesions were stiffer than the surrounding tissues and 15.52% (9 of 58) had black and white honeycomb-like images. The VTI images showed a significant difference between the malignant lesions and normal cervical tissues (P<0.001). The surrounding normal tissues had lower VTQ values, with a mean of 2.11±1.19 m/sec, while the VTQ values in malignant lesions were higher than the surrounding normal tissues (3.41±1.59 m/sec, P<0.001). ARFI ultrasound imaging of the uterine cervix may be an objective method for the assessment of soft tissues. It has high sensitivity and specificity in the evaluation of cervical cancer and therefore has good diagnostic value in clinical applications.

Renal oncocytoma: contrast-enhanced sonographic features.

OBJECTIVES: The purpose of this study was to illustrate the features of renal oncocytoma on contrast-enhanced sonography. METHODS: Five cases of surgical pathologically proven renal oncocytoma were retrospectively reviewed and analyzed in this study. The 5 patients studied included 3 men and 2 women with a mean age of 52.3 years (range, 32-66 years). All patients underwent conventional and contrast-enhanced sonographic examinations before surgery. The sonographic features, enhancement pattern, and dynamic perfusion change of the tumors and renal parenchyma in vascular phases were evaluated and compared to computed tomography. RESULTS: On grayscale sonography, renal oncocytomas appeared as solitary hypoechoic or isoechoic masses originating from the renal cortex measuring 2 to 6 cm with a well-defined margin. Color Doppler flow imaging showed rich blood flow signals in the periphery and striplike blood flow signals within the masses. On contrast-enhanced sonography, the features of these tumors included early enhancement, hyperenhancement, and fast wash-out compared to the adjacent renal cortex. There were irregular nonenhanced areas seen in the center of the masses (n = 3) and nonenhanced thin rims seen in the periphery (n = 5). Macroscopically, the well-marginated tumors were yellowish brown with a tender texture and were located in the renal cortex. Central irregularly shaped fibrous scars were seen in 3 tumors, consistent with nonenhanced areas on contrast-enhanced imaging. CONCLUSIONS: Contrast-enhanced sonography can provide valuable hemodynamic information on renal oncocytoma, which may help in making a differential diagnosis of this benign renal tumor in clinical practice.