Background: Evaluation of renal blood perfusion is important for patients with hypertension. Contrast-enhanced ultrasound (CEUS) is a safe and non-invasive technique that allows semi-quantitative assessment of renal cortical perfusion. Using CEUS quantitative analysis, we aimed to employ the characteristics of renal cortical blood perfusion (RCBP) parameters in hypertension patients, providing objective evidence for quantifying the hemodynamic features of renal microcirculation. Methods: It was a single-center retrospective study that included data from 83 hypertensive patients without renal artery stenosis who underwent renal CEUS at the Department of Ultrasound Medicine, Beijing Hospital between October 2020 and June 2023. The study cohort comprised 45 males and 38 females. Data collected included patient height, weight, systolic and diastolic blood pressure, estimated glomerular filtration rate (eGFR), RCBP parameters, and other historical records. Through stratified analysis, the characteristics of RCBP parameters were analyzed for the left/right sides, different sexes, and elderly versus non-elderly individuals. Results: There were no statistically significant differences in all cortical blood flow perfusion parameters [peak intensity (PI); rise time (RT); mean transit time (MTT); area under the curve (AUC); wash-in area under the curve (iAUC); wash-out area under the curve (oAUC); time to peak (TTP); PI/MTT] between the right and left kidneys of the patients (P>0.05). The iAUC and TTP were greater in males than in females, with marginal statistical differences (both P<0.05). The MTT and TTP were greater in the elderly group than in the non-elderly group, whereas PI/MTT was greater in the non-elderly group, with marginal statistical differences (both P<0.05). The intraclass correlation coefficients (ICCs) and Bland-Altman plots demonstrated good inter-observer agreement for all cortical blood flow perfusion parameters, especially for time-related parameters. Conclusions: This study preliminarily investigated the characteristics of RCBP parameters in hypertensive patients, further validating good reproducibility of the quantitative analysis technique of contrast-enhanced ultrasound.

Peripheral artery disease (PAD) confers elevated risk for major adverse cardiovascular events (MACE), yet accurate risk stratification remains a challenge, particularly among patients with advanced disease necessitating endovascular revascularization. This study aimed to improve the prediction of MACE in a clearly defined high-risk PAD population (hospitalized patients undergoing endovascular intervention) by identifying novel protein biomarkers and developing a robust risk model. We prospectively analyzed blood samples from 164 hospitalized PAD patients scheduled for endovascular revascularization, employing untargeted plasma proteomics and metabolomics. Differential protein and metabolite profiles were compared between patients with and without subsequent MACE. Several proteins, including MMP3, MMP19, and PRB2, were markedly elevated in patients who developed MACE. A proteomics-based risk model incorporating these biomarkers achieved high discriminative accuracy (area under the curve > 0.80) for identifying individuals at increased risk. Metabolomic profiling revealed additional pathway alterations, notably involving tryptophan and glycogen metabolism, which provided mechanistic insights into cardiovascular complications but were not directly incorporated into the prediction model. This study demonstrates that integrating protein biomarkers markedly improves risk stratification in advanced PAD patients undergoing surgical intervention. The findings offer promising tools for early detection and enable more personalized management for this high-risk subgroup, while also deepening understanding of disease pathophysiology. However, further validation in larger and more diverse prospective cohorts is warranted before these findings can be broadly applied in clinical practice.

As a chronic disease characterized by progressive inflammation and fibrosis, IgG4-related disease (immunoglobulin-G4 related disease, IgG4-RD) rarely involves the retroperitoneum. Early diagnosis and evaluation are vital for IgG4-RD patients. Herein, we report on an elderly male patient diagnosed with IgG4-RD, with superior mesenteric artery involvement. Combining the patient's clinical history, serology, imaging features, and excluding other suspected diseases, this patient was provisionally diagnosed as IgG4-related disease involving the superior mesenteric artery. Before and after medication, contrast-enhanced ultrasound was used to adjunctively assess whether fibrotic lesions around the superior mesenteric artery were in active phase to aid in drug efficacy assessment.

In middle-aged and older atherosclerotic renal artery stenosis (ARAS), the anatomical severity of stenosis is a poor surrogate for microvascular competence, and the renal benefit of revascularization is unpredictable. We developed Renal-Video-AI, a self-supervised deep learning framework (Video Swin Transformer with VideoMAE pretraining) that extracts spatiotemporal hemodynamic features from contrast-enhanced ultrasound, and applied it to a multi-center Discovery Cohort (N = 1,226), an independent External Validation Cohort (N = 122), a prospective Multimodal Cohort with paired 10x Visium spatial transcriptomics (N = 57), and an aged two-kidney-one-clip (2K1C) murine model. Unsupervised phenomapping identified 3 intrinsic hemodynamic phenotypes—Preserved, Delayed, and Rarefied. The Rarefied phenotype predicted major adverse renal events (MAREs) independently of anatomical stenosis [hazard ratio (HR) 4.82, 95% confidence interval (CI) 3.10 to 6.50; Fine–Gray subdistribution HR (sHR) 5.1], and adding the phenotype to a standard clinical model improved the C-statistic from 0.72 to 0.88. A significant phenotype-by-treatment interaction (P < 0.01) showed that stenting reduced events only in the Delayed phenotype (HR 0.52, 95% CI 0.35 to 0.78), not in the Preserved (HR 0.98) or Rarefied (HR 1.05) phenotypes. In absolute terms, stenting reduced the 3-year cumulative incidence of MARE in the Delayed phenotype from 25.4% to 13.2% (absolute risk reduction 12.2%; number needed to treat = 8, 95% CI 6 to 13), with no benefit in the Preserved (8.4% versus 8.0%) or Rarefied (38.6% versus 39.4%) phenotypes. Spatial transcriptomics localized a hypoxia and pyroptosis signature to rarefied tissue, and the aged 2K1C model revealed a mitochondrial reactive oxygen species (ROS)–NLRP3–pyroptosis axis whose pharmacological inhibition (MCC950) restored microvascular perfusion. AI video-phenomapping thus reframes the revascularization decision around microvascular competence rather than anatomy, identifying both therapeutic futility (Rarefied) and a treatable window (Delayed), and nominates NLRP3-driven pyroptosis as a therapeutic target.

Background:Artificial intelligence-enhanced imaging techniques have demonstrated promising diagnostic potential for carotid plaques, a key cardiovascular and cerebrovascular risk factor. However, previous studies did not systematically synthesize their diagnostic accuracy. Objective:This study aimed to quantitatively explore the diagnostic efficacy of deep learning (DL) and radiomics for extracranial carotid plaques and establish a standardized framework for improving plaque detection. Methods:We searched the PubMed, Embase, Cochrane, Web of Science, and Institute of Electrical and Electronics Engineers databases to identify studies involving the use of radiomics or DL models to diagnose extracranial carotid artery plaques from inception up to September 24, 2025. The quality of the studies was determined using Quality Assessment of Diagnostic Accuracy Studies for Artificial Intelligence (QUADAS-AI). A meta-analysis was conducted using StataMP (version 17.0; StataCorp) with a bivariate mixed-effects model to calculate pooled sensitivity and specificity, generate summary receiver operating characteristic (SROC) curves, assess Cochran Q statistic and I²-based heterogeneity, and conduct subgroup analyses and regression analysis. Results:Among 40 studies comprising 17,246 patients, 34 integrated independent test sets or validation sets in the quantitative statistical analysis. Among them, 24 focused on DL models, 10 on machine learning models based on radiomics. The combined sensitivity, specificity, and area under the SROC curve were 0.88 (95% CI 0.85-0.91; P<.001; I2=93.58%), 0.89 (95% CI 0.85-0.92; P<.001; I2=91.38%), and 0.95 (95% CI 0.92-0.96), respectively. Compared with the machine learning models based on radiomics algorithms, DL models achieved comparable improvements in specificity and area under the SROC curve. It was observed that transfer learning and a large sample size enhanced the diagnostic performance of models. Models used to identify plaque stability and presence had similar diagnostic performances, both of which were more effective in identifying symptomatic plaque models. A total of 7 studies demonstrated that the models that combined clinical features exhibited comparable diagnostic capability to pure DL and radiomics models. Additionally, 7 studies performed external validation, obtaining lower diagnostic performance than in testing groups. Limited regression analysis failed to identify significant sources of heterogeneity, and the limited number of eligible studies restricted more comprehensive subgroup analyses. The high heterogeneity in the study results may be due to different scanning parameters, model architecture, image segmentation, and algorithms. Conclusions:Radiomics algorithms and DL models can effectively diagnose extracranial carotid plaque. However, there are concerns regarding irregularities in research design and the absence of multicenter studies and external validation. Future research should aim to reduce bias risk and enhance the generalizability and clinical orientation of the models.

Parathyroid ultrasound is widely used in clinical practice and plays a crucial role in the diagnosis and treatment of parathyroid diseases. Nevertheless, ultrasound physicians frequently encounter a number of challenges and doubts in their professional practice. For this reason, Superficial Organs and Peripheral Vessels Committee of Chinese Association of Ultrasound in Medicine and Engineering has formulated the expert consensus on certain common clinical problems of parathyroid ultrasound based on the current research progress and clinical experience, in order to guide the clinical practice. This consensus describes in detail the diagnostic and interventional common problems of parathyroid ultrasound and provides in-depth discussion on related contents.

Purpose To develop and test a machine learning (ML)-based model that integrates preoperative variables for prediction of advanced-stage progression (ASP) after transarterial chemoembolization (TACE). Materials and Methods This multicenter retrospective study (ResearchRegistry.com identifier no. researchregistry9425) included patients with intermediate-stage hepatocellular carcinoma (HCC) who underwent TACE at seven hospitals from June 2008 to December 2022. Thirty-four preoperative clinical and CT imaging variables were input into six ML-based models for prediction of ASP, and model performances were compared. Furthermore, the best-performing ML model was compared with the major staging systems, and its utility in performing post-TACE therapies was assessed. The performances of the models were compared by using area under the receiver operating characteristic curve (AUC) with DeLong test. Kaplan-Meier survival curves were compared using the log-rank test. Results A total of 2333 eligible patients (mean age, 54 years ± 12 [SD]; 2051 male patients) were categorized into the training set (n = 1026), the internal test set (n = 257), and the external test set (n = 1050). ASP was found in 8.4% (86 of 1026), 8.2% (21 of 257), and 6.7% (70 of 1050) of patients in the three datasets, respectively. Among all ML models, the Categorical Gradient Boosting (CatBoost) model yielded the highest AUC: 0.97 (95% CI: 0.95, >0.99) for the training set, 0.94 (95% CI: 0.92, 0.97) for the internal test set, and 0.93 (95% CI: 0.90, 0.95) for the external test set. Furthermore, it yielded better discriminatory ability with higher concordance indexes than the five staging systems (all P < .001). The time-dependent AUC of the CatBoost model was also higher than that of the clinical staging systems at various time points (all P < .001). Moreover, post-TACE systemic therapy improved progression-free survival and overall survival for patients in the high-risk group (both P < .001) but not in the low-risk group. Conclusion The CatBoost model demonstrated higher predictive performance compared with existing staging systems in predicting ASP after TACE in patients with intermediate-stage HCC. This model effectively stratified patients by risk level and identified those who benefited from post-TACE systemic therapy. Keywords: Liver, Oncology, Transarterial Chemoembolization, Hepatocellular Carcinoma, Advanced-stage Progression, Machine Learning, Risk Differentiation ResearchRegistry.com identifier no. researchregistry9425 Supplemental material is available for this article. © RSNA, 2025 See also commentary by Rouzbahani in this issue.

Background:The evaluation of the accessory renal artery (ARA) holds clinical significance in the effective intervention of resistant hypertension and renal vascular-related surgical procedures. Multi-modal ultrasound is a non-invasive, secure, and real-time imaging modality, especially useful in patients with renal impairment. Nevertheless, few studies have focused on the value of multi-modal ultrasound in the assessment of the ARA. This study aimed to explore the diagnostic performances of multi-modal ultrasound in the assessment of the ARA. Methods:A retrospective data collection (clinical and imaging information) was conducted on patients who underwent renal artery conventional ultrasound and contrast-enhanced ultrasound (CEUS) examinations between August 2019 and November 2023 in Beijing Hospital. A total of 73 patients with a unilateral or bilateral ARA based on their computed tomography angiography (CTA) results were included. Compared with CTA results, the accuracy of multi-modal ultrasound for the assessment of the ARA was evaluated, and underlying reasons for misdiagnosis and missed diagnosis were analyzed. Results:Among the 73 patients (144 kidneys), CTA identified 85 ARAs, whereas multi-modal ultrasound detected 70 ARAs. Although multi-modal ultrasound failed to detect 15 ARAs, it did not result in any false-positive diagnoses. When CTA did not detect any ARAs in a kidney, multi-modal ultrasound also did not find any ARA. The sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and accuracy of multi-modal ultrasound in diagnosing ARA were calculated as 82.4%, 100%, 100%, 81.0%, and 90.9%, respectively. The receiver operating characteristic (ROC) analysis demonstrated an area under the curve (AUC) of 0.906 (P<0.001). The consistency analysis yielded a kappa value of 0.806 (P<0.01). Comparisons were conducted between patients with detected ARAs and those with missed ARAs. The age and body mass index (BMI) between the two groups were found to be statistically significant (P<0.05). Conclusions:Multi-modal ultrasound, characterized by its non-invasive, safe, and reproducible nature, demonstrates a high level of diagnostic accuracy in detecting the ARA. Thus, multi-modal ultrasound holds promise as a valuable tool for evaluating the ARA.

Background:Endoleaks are common complications after endovascular aneurysm repair (EVAR) for abdominal aortic aneurysm (AAA). Computed tomographic angiography (CTA)/digital subtraction angiography (DSA) is considered the gold standard for evaluating contrast-enhanced ultrasound (CEUS) accuracy in the detection and classification of endoleaks. In recent years, CEUS has been widely used in this field. This study aimed to analyze the accuracy of CEUS in the detection and classification of endoleaks after EVAR. Methods:The data of 98 patients who underwent abdominal aorta CEUS from November 2017 to September 2023 in the ultrasound (US) department of Beijing Hospital were retrospectively analyzed. All the patients underwent EVAR of AAA before CEUS and CTA/DSA, and had complete clinical data. The CEUS and CTA/DSA results were compared to detect endoleaks and categorize the specific types of endoleaks. Results:Among the 98 patients, 74 were male and 24 were female. The patients had an average age of 74.8±9.8 years (range, 43-90 years). Among the 98 patients, 37 (37.8%) endoleaks were detected by CEUS, of which 8 were type Ia, 2 were type Ib, 15 were type II, 7 were type III, 2 were type IV, 2 were type Ia combined with type III, and 1 was type II combined with type III. In addition, among these 98 patients, 39 (39.8%) endoleaks were detected by CTA/DSA, of which 8 were type Ia, 3 were type Ib, 18 were type II, 6 were type III, 2 were type Ia combined with type III, 1 was type II combined with type III, and 1 was type Ib combined with type II. The sensitivity and specificity of CEUS in the detection of endoleaks were 92.3% and 98.3%, respectively. CEUS and CTA/DSA had similar diagnostic efficacy and good consistency in the detection and classification of endoleaks (Kappa value: 0.914, P<0.01). Conclusions:CEUS has high sensitivity and specificity in the detection and classification of endoleaks following EVAR, and its diagnostic efficacy is similar to that of CTA/DSA. In addition, US is safe, non-invasive and repeatable, and thus is worthy of extensive clinical application.

As a sensitive and non-invasive method for assessing changes in renal cortical blood perfusion in the elderly, contrast-enhanced ultrasound (CEUS) can indirectly reflect changes in kidney filtration and reabsorption function, thus providing feasibility for early evaluation of renal function changes. However, significant differences exist among researchers in terms of operational methods, contrast agent selection, post-data analysis, and many other aspects, leading to substantial heterogeneity in results. This hinders horizontal comparisons and greatly limits the clinical application of contrast-enhanced ultrasound for evaluating renal cortical blood flow perfusion. Based on the latest domestic and overseas literature and discussions with clinical experts, this consensus provides recommended guidelines for the evaluation of renal cortical blood flow perfusion using contrast-enhanced ultrasound. It is hoped that this consensus will promote a better understanding of CEUS among medical practitioners at all levels and standardize the examination of renal cortical blood flow perfusion with CEUS.

PurposeUltrasound is the imaging modality of choice for preoperative diagnosis of lymph node metastasis (LNM) in thyroid cancer (TC), yet its efficacy remains suboptimal. As radiomics gains traction in tumor diagnosis, its integration with ultrasound for LNM differentiation in TC has emerged, but its diagnostic merit is debated. This study assesses the accuracy of ultrasound-integrated radiomics in preoperatively diagnosing LNM in TC.MethodsLiteratures were searched in PubMed, Embase, Cochrane, and Web of Science until July 11, 2023. Quality of the studies was assessed by the radiomics quality score (RQS). A meta-analysis was executed using a bivariate mixed effects model, with a subgroup analysis based on modeling variables (clinical features, radiomics features, or their combination).ResultsAmong 27 articles (16,410 TC patients, 6356 with LNM), the average RQS was 16.5 (SD:5.47). Sensitivity of the models based on clinical features, radiomics features, and radiomics features plus clinical features were 0.64, 0.76 and 0.69. Specificities were 0.77, 0.78 and 0.82. SROC values were 0.76, 0.84 and 0.81.ConclusionUltrasound-based radiomics effectively evaluates LNM in TC preoperatively. Adding clinical features does not notably enhance the model's performance. Some radiomics studies showed high bias, possibly due to the absence of standard application guidelines.

Background: Frailty and clonal hematopoiesis of indeterminate potential (CHIP) have emerged as crucial predictors of adverse cardiovascular outcomes in older adults. However, their combined impact on major adverse cardiovascular events (MACE) in patients with severe atherosclerotic renal artery stenosis (ARAS) remains unclear. Methods: We conducted a prospective cohort study involving 175 patients aged 60 years and older with severe ARAS (luminal stenosis ≥ 70%) who underwent renal artery stenting at Beijing Hospital between January 2019 and December 2022. Frailty was assessed using the Fried phenotype, categorizing patients into robust, prefrail, and frail subgroups. CHIP status was determined through targeted gene sequencing of peripheral blood, stratifying patients into No CHIP (VAF < 2%), Small CHIP (VAF 2%-<10%), and Large CHIP (VAF ≥ 10%) subgroups. All patients were systematically followed up until June 30, 2024. The primary outcome was the incidence of MACE, which was a composite of renal function deterioration (RFD), initiation of renal replacement therapy, renal artery revascularization, nonfatal myocardial infarction, hospitalization for heart failure, nonfatal stroke, and cardiorenovascular death. We employed Cox proportional hazards models, Kaplan-Meier survival analysis, and heatmaps to explore the combined impact of frailty and CHIP on MACE risk. Results: The mean age of the patients was 68.3 years. Of the cohort, 64.6% had no CHIP, 26.8% had Small CHIP, and 8.6% had Large CHIP. Frail patients showed a higher prevalence of CHIP, particularly in the Small (34.7%) and Large (10.2%) CHIP categories. During a median follow-up of 32 months, 54 MACE occurred. Kaplan-Meier survival curve revealed that frailty was associated with a higher incidence of MACE (35.7% in frail vs. 29.5% in prefrail vs. 24.6% in robust, P = 0.045) and RFD (16.3% in frail vs. 11.5% in prefrail vs. 7.7% in robust, P = 0.034). Patients with Large CHIP experienced significantly higher rates of MACE (60.0% vs. 36.2% in Small CHIP vs. 24.8% in No CHIP, P = 0.004) and RFD (26.7% vs. 14.9% in prefrail vs. 8.0% in robust, P = 0.019). Findings for RFD appeared to be consistent with those for MACE. Frailty and CHIP status showed independent contribution to overall risk. The greatest spread for MACE and RFD risk was obtained in models that incorporated frail and Large CHIP. Conclusion: Frailty and CHIP, independently and jointly, contribute to a significantly higher risk of MACE and RFD in elderly patients with severe ARAS undergoing stenting. These findings highlight the necessity for integrated risk stratification and targeted management strategies in this high-risk population.

Background:Doppler ultrasound (DUS) is recommended in first-line imaging for the diagnosis of renal artery stenosis (RAS). However, the correct selection of Doppler direct or indirect parameters and their optimal thresholds remain controversial. This study explored simple ultrasound Doppler parameters to diagnose severe RAS (RAS ≥70%) in routine clinical practice. Methods:In this retrospective study, patients with clinically suspected renovascular hypertension who first underwent renal artery DUS and contrast-enhanced ultrasound (CEUS) and subsequent digital subtraction angiography (DSA) or computed tomography angiography (CTA) were consecutively included. Clinical characteristics and ultrasound Doppler hemodynamic parameters were collected, including peak systolic velocity (PSV), the ratio of the peak velocities in the renal artery and the aorta (RAR), the ratio of the peak velocities in the renal artery and the segmental artery (RSR), and the ratio of the peak velocities in the renal artery and the interlobar artery (RIR). All enrolled patients were divided into two groups based on the degree of diameter reduction: a severe stenosis group (diameter reduction ≥70%) and a non-severe stenosis group (diameter reduction <70%). Logistic regression analysis was performed to determine the independent predictors for severe stenosis. Receiver operating characteristic curves and areas under the curve were used to evaluate the diagnostic performance of the ultrasound Doppler parameters. Results:A total of 85 patients (106 renal arteries) with RAS were included in this study. The optimal thresholds of PSV in the main renal artery and the PSV ratios for diagnosing severe RAS obtained via receiver operating characteristic curves were 249.5 cm/s for PSV, 2.94 for RAR, 5.1 for RSR, and 7.5 for RIR. The areas under the curve of PSV and the ratios all exhibited good diagnostic efficiency (all >0.8). The combination of these four Doppler variables demonstrated a significant benefit to the overall diagnostic value compared with any factor alone [area under the curve (AUC) =0.962; 95% confidence interval (CI): 0.906-0.989; P<0.05]. The combination of PSV and RSR (AUC =0.925; 95% CI: 0.858-0.967) exhibited comparable diagnostic efficiency to the combination of four ultrasonographic variables (z statistic =1.882; P=0.06). Conclusions:This simple and accurate method to evaluate severe RAS based on the velocity obtained via basic DUS may facilitate the detection of severe RAS in the majority of medical institutions and provide a reliable basis for the selection of proper candidates for further angiography or revascularization.

Background:Renal hemodynamic changes in early diabetes occur before the onset of significant structural abnormalities or clinical manifestations, and timely detection of these changes has clinical significance. This study aimed to evaluate renal elasticity and perfusion changes in an early-stage diabetic rat model by shear wave elastography (SWE) and contrast-enhanced ultrasound (CEUS), and to explore the potential correlations between renal elasticity and perfusion parameters.Methods:A total of 18 male Sprague-Dawley rats were randomly divided into three groups: a control group (group 1, n=6), a diabetic group (group 2, n=6), and a diabetic group receiving drug therapy (group 3, n=6). An intraperitoneal injection of streptozotocin (STZ) for 2 days combined with a high-fat diet (HFD) was used as the early-stage diabetic rat model. The diabetic rats in group 3 were treated with canagliflozin and losartan for 6 weeks, whereas the rats in groups 1 and 2 were given equal amounts of purified water. Renal stiffness on SWE and perfusion parameters on CEUS were measured and compared among the three groups, then the rats were sacrificed, and serum, urine, and renal histopathology were evaluated to confirm the development of early diabetes.Results:The early-stage diabetic rats without significant pathological changes exhibited bigger kidneys and higher blood glucose (all P<0.05). Among the CEUS parameters, peak enhancement (PE), wash-in area under the curve (WiAUC), wash-in perfusion index (WiPI), wash-out AUC (WoAUC), wash-in and wash-out AUC (WiWoAUC), rise time (RT), and time to peak (TTP) of diabetic rats in group 2 were significantly increased (all P<0.05), and the hyperperfusion ameliorated significantly after drug treatment. The renal elasticity measured by SWE varied in accordance with certain perfusion parameters, and was strongly positively correlated with WiAUC (r=0.701, P<0.001), WoAUC (r=0.647, P<0.001), and WiWoAUC (r=0.655, P<0.001), and moderately positively correlated with PE (r=0.539, P=0.001), WiPI (r=0.555, P<0.001), RT (r=0.425, P=0.010), and TTP (r=0.439, P=0.007).Conclusions:Renal elasticity and perfusion changes in the early stage of diabetes, and renal elasticity was positively associated with delayed and increased perfusion.

OBJECTIVES:To evaluate the diagnostic performance of renal artery contrast-enhanced ultrasound (CEUS) with modified inspection section and summarize subsequent changes in imaging assessment of renal artery disease.METHODS:A total of 1015 patients underwent renal artery CEUS were included in the study. Among them, 79 patients (156 renal arteries) suspected with renal artery stenosis (RAS) underwent digital subtraction angiography (DSA) subsequently. DSA was used as the gold standard to evaluate the diagnostic performance of CEUS in detecting RAS (≥30%) and severe stenosis (≥70%), as well as the diagnostic accuracy of classification of stenosis degree. Besides, 127 of the 1015 patients underwent other imaging examinations such as computed tomography angiography (CTA) or magnetic resonance angiography (MRA) after CEUS and annual proportion of these imaging examinations was assessed.RESULTS:The sensitivity, specificity, accuracy, positive predictive value (PPV) and negative predictive value (NPV) of CEUS for detecting RAS (≥30%) was 96.4%, 88.6%, 94.2%, 95.6% and 90.7%, respectively and the kappa value was .857 (P < .01). CEUS had a good performance in distinguishing severe stenosis (≥70%) with a sensitivity of 91.1%, specificity of 95.5%, accuracy of 92.9%, PPV of 96.5%, NPV of 88.7% and the kappa value was 0.857(P < .01). There was no significant difference between CEUS and DSA in detecting stenosis (P = 1.0) and severe stenosis (P = .227). The diagnostic accuracy of CEUS in grading RAS was 85.3% and the kappa value was 0.753 (P < .01). Besides, the annual proportion of other imaging examinations decreased for 4 consecutive years.CONCLUSIONS:CEUS is a non-invasive, safe and valuable technique for the assessment of renal artery disease and worthy of promotion.

Vulnerable carotid atherosclerotic plaque (CAP) significantly contributes to ischemic stroke. Neovascularization within plaques is an emerging biomarker linked to plaque vulnerability that can be detected using contrast-enhanced ultrasound (CEUS). Computed tomography angiography (CTA) is a common method used in clinical cerebrovascular assessments that can be employed to evaluate the vulnerability of CAPs. Radiomics is a technique that automatically extracts radiomic features from images. This study aimed to identify radiomic features associated with the neovascularization of CAP and construct a prediction model for CAP vulnerability based on radiomic features. CTA data and clinical data of patients with CAPs who underwent CTA and CEUS between January 2018 and December 2021 in Beijing Hospital were retrospectively collected. The data were divided into a training cohort and a testing cohort using a 7:3 split. According to the examination of CEUS, CAPs were dichotomized into vulnerable and stable groups. 3D Slicer software was used to delineate the region of interest in CTA images, and the Pyradiomics package was used to extract radiomic features in Python. Machine learning algorithms containing logistic regression (LR), support vector machine (SVM), random forest (RF), light gradient boosting machine (LGBM), adaptive boosting (AdaBoost), extreme gradient boosting (XGBoost), and multi-layer perception (MLP) were used to construct the models. The confusion matrix, receiver operating characteristic (ROC) curve, accuracy, precision, recall, and f-1 score were used to evaluate the performance of the models. A total of 74 patients with 110 CAPs were included. In all, 1,316 radiomic features were extracted, and 10 radiomic features were selected for machine-learning model construction. After evaluating several models on the testing cohorts, it was discovered that model_RF outperformed the others, achieving an AUC value of 0.93 (95% CI: 0.88-0.99). The accuracy, precision, recall, and f-1 score of model_RF in the testing cohort were 0.85, 0.87, 0.85, and 0.85, respectively. Radiomic features associated with the neovascularization of CAP were obtained. Our study highlights the potential of radiomics-based models for improving the accuracy and efficiency of diagnosing vulnerable CAP. In particular, the model_RF, utilizing radiomic features extracted from CTA, provides a noninvasive and efficient method for accurately predicting the vulnerability status of CAP. This model shows great potential for offering clinical guidance for early detection and improving patient outcomes.