Abstract Background Pancreatic cystic lesions (PCLs) have variable malignant potential, and distinguishing benign from malignant/premalignant cysts is challenging for optimal management. Trans-abdominal ultrasound (TAUS), a widely available, non-invasive, low-cost first-line pancreatic imaging tool, is underexplored for predicting PCL malignant potential. This study aimed to develop a TAUS feature-based nomogram for this purpose. Methods This retrospective study included 161 patients with pathology-confirmed PCLs from December 2012 to July 2021, divided into benign (59 cases) and non-benign (premalignant/malignant, 102 cases) groups. Relevant clinical characteristics and TAUS features were collected. Least absolute shrinkage and selection operator (LASSO) logistic regression analysis was used to optimize feature selection. Multivariate logistic regression analysis was applied to construct the nomogram. The performance of the nomogram was assessed via receiver operating characteristic curves, calibration curves and decision curve analysis (DCA). Results Among the 26 features collected, 11 features were chosen via LASSO analysis. Multivariate analysis identified echogenicity, the configuration of cysts, solid content and septation/wall thickening as independent predictors. The prediction nomogram model developed with these four variables showed moderate discriminative performance in differentiating non-benign from benign PCLs, with an area under the curve (AUC) of 0.781. Regarding internal verification, tenfold cross-validation yielded a C-index of 0.749. The Hosmer–Lemeshow test yielded a P = 0.945, suggesting that the model had a good fit. Additionally, DCA demonstrated good net clinical benefit. Conclusions This study explored the value of TAUS features for predicting the malignant potential of PCLs. The incorporation of TAUS features into a nomogram may offer a potential non-invasive tool to help clinicians in risk stratification during the initial evaluation and follow-up assessment of PCLs patients.

OBJECTIVE:To evaluate the feasibility of using large language model (LLM) for automated quality control (QC) of breast ultrasound (US) reports. METHODS:In this retrospective multicenter study, we collected breast US reports from 735 patients who had pathological confirmation of mass-type lesions on breast US across 60 hospitals in China. Each hospital's QC personnel converted free-text reports into standardized structured outputs. A gold standard was established through a multilevel expert review. The Qwen2.5-VL-7B LLM was applied to the same free-text reports, generating structured reports based on BI-RADS prompts. Accuracy was compared both between the LLM and QC personnel as well as the times required to produce the outputs for analysis of efficiency. RESULTS:The LLM demonstrated higher accuracy in conducting QC for key breast lesion US features, such as margin (80.0% versus 64.1%, P < .0001) and echo pattern (74.0% versus 56.1%, P < .0001). Subgroup analysis further confirmed its robustness: in the complex reports of multiple lesions, it maintained its advantage in margin QC (79.8% versus 64.4%). The study also revealed a significant positive correlation between the LLM's QC accuracy and BI-RADS categories (from 3 to 5) (Spearman ρ = 0.264, P = .028), a trend not observed in manual QC. In terms of efficiency, the LLM completed QC for 50 reports in an average of only 13 min, faster than 212.5 min required by manual reviewers. CONCLUSIONS:The proposed LLM-based system provides a reliable, accurate, and efficient solution for breast US reports. It achieves human-comparable performance with markedly higher efficiency, particularly in reports with high suspicion levels, offering a reliable tool to enhance report quality.

Identifying TR5 nodules smaller than 10 mm that have increased in size is an important issue in clinical practice. Our study investigated whether superb microvascular imaging (SMI) and shear wave elastography (SWE) could help identify previous enlargement in ACR TI-RADS 5 (TR5) thyroid nodules measuring ≤ 10 mm. Each nodule underwent ultrasound examination for SMI and SWE. VImax is the largest vascular index between the transverse and longitudinal sections. The SWV ratio is the ratio of thyroid nodules to peripheral glands. The enlargement of thyroid nodules was defined as a growth rate > 2 mm/y or an increase in size of > 3 mm in the largest dimension. Of all 51 TR5 TNs, 25 (49.0

To characterize the transabdominal ultrasound (US) features of pancreatic serous cystic neoplasm (SCN) morphological subtypes, evaluate US-MRI agreement, and explore reducing overtreatment. This retrospective study included 125 patients with pathologically confirmed pancreatic SCNs who underwent preoperative US. Lesions were morphologically classified into macrocystic, microcystic, mixed, and solid subtypes. US-MRI concordance (71 patients, 56.8

The standard treatment for advanced ovarian cancer follows a comprehensive ‘surgery-chemotherapy-maintenance therapy’ mode, typically involving initial cytoreductive surgery aiming for R0 resection, six cycles of platinum-based chemotherapy, followed by maintenance therapy for those who have responded well to the treatment. However, frailty and high incidence of comorbidities in elderly patients often compromise surgical outcomes, necessitate chemotherapy dose reductions, and limit maintenance therapy continuation, resulting in a poor prognosis. Poly (Adenosine diphosphate (ADP)-ribose) polymerase inhibitors (PARPi) have revolutionized the management strategy of homologous recombination deficiency (HRD)-positive patients as a groundbreaking advancement in first-line maintenance therapy. Fluzoparib, the domestically developed PARPi in China, has demonstrated significant efficacy in BRCA-mutated ovarian cancer. In the field of supportive care, megestrol acetate (MA) is recommended as the first-line preferred therapeutic agent for cancer-related anorexia by major guidelines, though its role in first-line ovarian cancer therapy remains unexplored, and evidence for its combination with PARPi is lacking. This article reported a case of an 89-year-old female patient with high-grade serous ovarian carcinoma. Due to intolerance to surgery and chemotherapy, an innovative first-line primary treatment regimen combining fluzoparib with MA was initiated based on BRCA2 mutation and HRD-positive status. Imaging assessments revealed significant tumor reduction without disease progression or grade ≥3 adverse events observed throughout follow-up. This case highlights the potential of combining PARPi and hormone therapy as a ‘chemotherapy-free’ precision treatment model for elderly and HRD-positive ovarian cancer patients, offering a promising strategy to balance efficacy and tolerability in a population traditionally underserved by conventional regimens.

Background:Pancreatic cystic lesions (PCLs) require precise imaging characterization to guide clinical management. Contrast-enhanced ultrasound (CEUS) reports contain operator-dependent narratives that challenge clinicians. Large language models (LLMs) show potential in medical text analysis but lack validation for pancreatic CEUS interpretation. This study primarily aimed to evaluate the diagnostic accuracy of LLMs in interpreting Chinese CEUS reports of PCLs. Methods:We retrospectively analyzed 80 pathologically confirmed PCLs (21 benign, 25 borderline malignant, 34 malignant). Four LLMs (GPT-4o, Claude 3.7 Sonnet, Gemini 2.0, DeepSeek-R1) and eight radiologists (senior/junior =4:4) independently interpreted reports under three input modalities: grayscale-only (IM1), grayscale + CEUS (IM2), and demographics + grayscale + CEUS (IM3). A weighted scoring system (0-100 points per case, yielding a maximum total of 8,000 points) quantified alignment with pathology-defined categories (benign/borderline/malignant). LLM errors were categorized into four reasons. Junior radiologists reinterpreted cases with LLM assistance. Results:Under CEUS input conditions (IM2/IM3), LLMs showed no statistically significant difference from senior radiologists and significantly outperformed junior radiologists. The median score per case (out of 100), after averaging across the four LLMs, increased with input complexity (IM1: 47.50; IM2: 53.75; IM3: 73.75). Diagnostic accuracy varied by pathology: malignant lesions scored highest, while benign serous cystic neoplasms scored lowest due to suboptimal CEUS visualization and LLMs' knowledge gaps. LLM guidance elevated junior radiologists' accuracy to senior levels. Conclusions:LLMs show promising capability in interpreting CEUS reports of PCLs, with no statistically significant difference from senior radiologists in this dataset. Their integration significantly improves junior clinicians' interpretation of CEUS reports. These findings support further investigation of LLMs as potential auxiliary tools for ultrasound text analysis, though external validation is needed.

Reliable methods for the non-invasive monitoring of placental status are essential for the management of preeclampsia. This study investigates the feasibility of multimodal photoacoustic-ultrasound (PA-US) imaging for the longitudinal assessment of placental oxygenation (sO2) and microvascular perfusion in a rat model. Our findings demonstrate that multimodal PA-US imaging parameters illustrate alterations in placental status non-invasively, which correlate strongly with histopathological markers, including hypoxia-inducible factor 1α (HIF-1α) and microvascular density (MVD). Besides, a characteristic inverted trajectory of imaging parameters during early gestation day 10–12 is observed. By enabling the simultaneous evaluation of oxygen and blood supply dynamics, multimodal PA-US imaging yields a more comprehensive view of placental function. Ultimately, these results suggest that multimodal PA-US imaging serves as a non-invasive and viable tool for detecting placental dysfunction and monitoring therapeutic responses in real time, laying a solid experimental foundation for advanced maternal-fetal care.

Axillary lymph node (ALN) metastasis is a critical factor influencing prognosis and treatment strategies in breast cancer patients. However, traditional methodsu2014ranging from physical examination to ultrasoundu2014often lack the precision required for clinical decision-making. In recent years, ultrasound radiomics and deep learning have emerged as promising solutions, leveraging high-throughput quantitative features from ultrasound images to enhance detection accuracy. This review explores the development and application of radiomics and deep learning across multiple ultrasound modalities (grayscale, elastography, and contrast-enhanced ultrasound), as well as in multimodal imaging approaches that integrate ultrasound with MRI and PET/CT, underscoring the benefits of incorporating clinicopathological variables to boost predictive performance. These studies provide a vital foundation for personalized treatment and precision medicine in breast cancer management.

To determine a direct method for diagnosing axillary lymph node (ALN) tumor burden preoperatively in cT1-2N0 breast cancer patients, we developed and validated a deep learning (DL) model based on ultrasound (US) images of sentinel lymph nodes (SLNs) detected by contrast-enhanced lymphatic ultrasound (CEUS). Women with cT1-T2N0 breast cancer who received CEUS were enrolled prospectively from Peking Union Medical College Hospital between April 2020 and July 2021 and from Sichuan Cancer Hospital between April 2022 and July 2022. Heavy ALN tumor burden was defined as > 2 metastatic lymph nodes according to the Z0011 criteria. We developed a DL model, the modality-adaptive network with clinicopathological information (MAN + C), using grayscale or color Doppler US images and radioclinicopathological information to predict heavy tumor burden. A total of 595 SLNs from 374 patients met the inclusion criteria. The areas under the receiver operating characteristic curve (AUCs) were calculated to evaluate the predictive performance of the model, yielding values of 0.91[95

Background: Preoperative diagnosis of medullary thyroid carcinoma (MTC) is clinically challenging due to sonographic overlap with other thyroid tumors. To address this, we aimed to develop a multi-vendor, multimodal radiomic framework for accurate MTC identification, comparing its diagnostic performance with that of experienced radiologists. Methods: This retrospective study included 467 pathologically confirmed thyroid nodules (94 MTCs, 373 non-MTCs) acquired across multiple ultrasound platforms. The dataset was randomly partitioned into training (80%) and internal testing (20%) sets. In total, 2250 radiomic features were extracted from grayscale and color Doppler images, followed by Z-score normalization to mitigate batch effects. A robust feature selection strategy (LASSO and recursive feature elimination) identified optimal signatures for developing machine learning classifiers (SVM, LR, RF). The optimal model was further validated on an independent, balanced cohort (n = 60; comprising 12 cases each of MTC, papillary carcinoma, follicular carcinoma, follicular adenoma, and nodular goiter) and compared with experienced radiologists across seven classification tasks. Results: The RF model achieved an AUC of 0.993 in distinguishing MTC from papillary carcinoma. The LR model showed an AUC of 0.991 for identifying MTC from all other nodules. In the independent validation cohort, the models maintained superior discriminatory ability, showing better diagnostic performance compared to the image interpretation by radiologists (AUC 0.993 vs. 0.488, p < 0.001). Conclusions: The proposed multi-vendor, multimodal radiomic system demonstrated good discriminative ability in the diagnosis and stratification of MTC. By integrating grayscale and Doppler ultrasound features while overcoming scanner variability, this model shows potential as a non-invasive adjunctive tool.

ABSTRACT Objectives The objective of this study is to investigate the value of conventional ultrasound, contrast‐enhanced ultrasound (CEUS) and percutaneous ultrasound‐guided biopsy in the diagnosis of pancreatic metastases from clear cell renal cell carcinoma (ccRCC). Methods A retrospective database search was conducted to identify all pancreatic lesions performed CEUS and final diagnosed as metastasis from ccRCC at our institution from September 2017 to June 2024. All enrolled patients were considered unsuitable for endoscopic ultrasound (EUS)‐guided tissue acquisition, had failed EUS‐guided biopsy, or required an alternative diagnostic approach. A total of 23 patients were enrolled in this study. Demographics, clinical features, ultrasound and CEUS findings, biopsy procedures, and results were collected and reviewed. Results Of the enrolled patients, metachronous metastasis was 100%. The median interval time from tumor‐nephrectomy to the emergence of the pancreatic metastasis was 121.0 months (range: 46.0–276.0 months). Ultrasonography revealed rounded, well‐defined, homogeneous hypoechogenic lesions in all the patients. Color Doppler flow image findings displayed plenty of blood flow signals around and inside the pancreatic lesions in 87.0% of patients. 22 patients (95.7%) exhibited rapid hyper‐enhancement in the arterial phase of CEUS and 1 patient (4.3%) showed iso‐enhancement. In the venous phase, 15 cases showed sustained hyper‐enhancement and 8 cases showed iso‐enhancement. In the biopsy procedure, ultrasound‐guided core needle biopsy demonstrated higher diagnostic yield than fine needle aspiration (75.0% vs. 46.7%). The biopsy diagnosis was positive in 13 patients (65.0%) in general. Conclusions Metastatic tumors of the pancreas can occur after a prolonged period from ccRCC diagnoses, thus long‐term follow‐up is necessary. The features of the lesion by conventional ultrasound and CEUS were specific, which are valuable for diagnosis. In patients who are unsuitable for EUS‐guided tissue acquisition or have experienced unsuccessful EUS‐guided sampling, CEUS combined with percutaneous ultrasound‐guided biopsy may serve as a valuable complementary second‐line diagnostic approach and facilitate accurate diagnosis.

Jugular vein aneurysms are extremely rare, and cases concurrent with papillary thyroid carcinoma (PTC) and multi-site venous aneurysms are even rarer. This report describes a 59-year-old female with left internal jugular vein aneurysm and left superficial femoral vein aneurysm complicated by thrombosis after total thyroidectomy for PTC. The thrombus within the internal jugular vein was similar to a metastatic lymph node; however, a correct diagnosis was ultimately reached under the physician's examination. We present this case to enhance junior physicians' skills in the differential diagnosis of neck masses and propose a preliminary hypothesis regarding the etiology of venous aneurysms.

Hypervascular pancreatic ductal adenocarcinoma (PDAC) and mass-forming pancreatitis (MFP) represent a classic diagnostic mimicry on contrast-enhanced ultrasound, as both exhibit similar arterial-phase hyperenhancement, precluding reliable visual distinction. Crucially, the hypervascular PDAC subtype is associated with a more favorable prognosis, rendering its accurate identification from its benign inflammatory mimic (MFP) a clinically significant priority for early and appropriate intervention. This study aimed to develop and validate a small-sample–oriented machine learning framework leveraging quantitative time-intensity curve (TIC) features to achieve this precise differentiation. We retrospectively included 152 patients with pathologically confirmed lesions who underwent CEUS between September 2017 and April 2024 (85 hypervascular PDAC, 67 MFP). A temporally separated split was used: 122 patients (2017–2022) formed the training cohort and 30 patients (2023–2024) served as the internal test cohort. Paired TICs were generated from the lesion and adjacent normal pancreatic parenchyma, and 22 quantitative difference/ratio features describing enhancement amplitude, temporal kinetics and curve morphology were extracted. Based on independent-samples t-tests and clinical interpretability, five representative TIC features were selected to train six classical classifiers (logistic regression, support vector machine, k-nearest neighbors, random forest, naïve Bayes, and decision tree). Model performance was assessed by stratified 10-fold cross-validation on the training cohort and by testing on the temporally separated cohort. In nested cross-validation, all models achieved AUCs of approximately 0.89–0.92. On the test cohort, AUCs ranged from about 0.83 to 0.92, with logistic regression performing best (AUC 0.915). Overall, the six classifiers showed broadly comparable discrimination. A machine-learning model built on a small set of physiologically interpretable CEUS-TIC features can provide stable and explainable quantitative support for differentiating hypervascular PDAC from MFP, even under limited-sample conditions.

The 70-gene signature (70-GS; MammaPrint) assay is useful for prognosis assessment in HR+/HER2- early breast cancer, but limited accessibility motivates development of noninvasive alternatives. We retrospectively enrolled 219 women with preoperative grayscale ultrasound and 70-GS results, including a development cohort (n = 125), an internal validation cohort (n = 53), and a temporally independent validation cohort (n = 41). Radiomic features were extracted from manually delineated ROIs using PyRadiomics, and a radiomics score was derived after LASSO selection. Candidate radiomics-only, clinicopathologic-only, and full clinicoradiomic models were explored. To reduce overfitting, we selected a parsimonious model combining the radiomics score and Ki67 as the primary model. The simplified model achieved AUCs of 0.878, 0.816, and 0.831 in the development, internal validation, and temporally independent validation cohorts, respectively. In 1000 bootstrap resamples, the optimism-corrected AUC was 0.872 and the corrected calibration slope was 0.953. Adding the radiomics score to a Ki67-only model significantly improved model fit (likelihood-ratio chi-square = 17.14, df = 1, p < 0.001). An ultrasound radiomics and Ki67 model may provide a noninvasive reference for estimating MammaPrint risk categorization, but it should be considered only as a supportive adjunct and not as a replacement for genomic testing.

Background:Histotripsy, an ablation method relying on acoustic cavitation, is characterized by its non-thermal, non-ionizing, and non-invasive properties. It has the advantages of tissue selectivity, real-time visualization, and precise ablation. Among the clinical applications of histotripsy, its use in the treatment of liver cancer is the most promising. This review aimed to synthesize preclinical evidence and provide a comprehensive evaluation of histotripsy in the treatment of liver disease. Methods:PubMed, Embase, and Cochrane Library were searched from database inception to October 15, 2024 to retrieve relevant articles. The Systematic Review Centre for Laboratory Animal Experimentation risk of bias tool was used to assess study quality, and RevMan 5.4 was used to conduct the statistical analyses. Results:In total, 30 animal studies, comprising 25 in vivo and five ex vivo liver studies were included in this review on the application of histotripsy in the treatment of liver disease. In the feasibility research, 97% (96/99) of cases achieved precise ablation with minimal adverse effects (AEs), such as venous thrombosis and local tissue damage. In the rat liver cancer studies, all cases (9/9) who underwent complete ablation achieved 12-week survival. In the partial ablation group, 80% (12/15) achieved effective tumor burden reduction, while 20% (3/15) exhibited local tumor progression. Additionally, while no liver metastases were observed in the partial ablation group, liver metastases were observed in 64.71% (11/17) of the control group. In murine studies, histotripsy not only induced local tumor growth arrest but also triggered abscopal anti-tumor immune effects, manifested by the inhibition of contralateral tumor growth. Boiling histotripsy (BH) treatment of liver fibrosis reduced fibrosis scores and promoted hepatocyte regeneration. In cell therapy, the transplanted hepatocytes proliferated and integrated into the recipient livers. Conclusions:Histotripsy, a promising localized liver ablation method, is safe, targeted, and effective. It can precisely target and ablate tumor or fibrotic tissue through mechanical effects while preserving vital anatomical structures. Additionally, it can stimulate the immune system, inhibiting tumor growth and metastases.

Objective To investigate the association between Varicella-Zoster Virus (VZV) and Giant Cell Arteritis (GCA). Methods One publicly available plasma proteomics dataset and three transcriptomics datasets were used. From the GCA samples, the corresponding values of VZV-human interacting proteins were extracted, and spectral clustering was performed on the GCA samples to calculate the silhouette coefficient. Subsequently, from the GCA samples, proteins or genes of the same quantity were randomly sampled 10,000 times; the identical spectral clustering method was applied to these randomly sampled sets, and the silhouette coefficient was calculated for each. Differences in spectral clustering performance and silhouette coefficients were compared between the VZV-human interacting proteins and the 10,000 randomly sampled protein/gene sets in the GCA samples. The same entire analytical process was repeated for the control group. Results In the GCA samples, the VZV-human interacting proteins exhibited a distinct clustering effect. Their silhouette coefficient was greater than that of most randomly sampled protein or gene sets, and this difference was statistically significant. No similar phenomenon was observed in the control group. Conclusion Varicella-Zoster Virus is associated with a subset of Giant Cell Arteritis cases. ### Competing Interest Statement The authors have declared no competing interest.

Conventional ultrasound (US) evaluation of enthesitis in psoriatic arthritis (PsA) is limited by its inability to quantify metabolic alterations such as hypoxia, a key driver of disease activity. We introduce an oxygenation-integrated multimodal photoacoustic/ultrasound (PA/US) imaging framework designed to quantify entheseal oxygen saturation (SO2) for assessing entheseal disease activity in PsA. In this cross-sectional study, 25 PsA patients underwent bilateral PA/US imaging of 12 entheses, where ultrasound lesions were scored using the Outcome Measures in Rheumatology scoring system, and PA-derived SO2 levels, quantified via dual-wavelength PA imaging, were classified into hyperoxia or hypoxia groups using k-means clustering. This approach provides metabolic insights complementary to conventional ultrasonic assessment. A composite score integrating hypoxia with US parameters was validated against clinical disease activity indices (Disease Activity Score 28-C-reactive protein, DAS28-CRP; Disease Activity Index for Psoriatic Arthritis, DAPSA). Among 300 entheses, 103 (34.3%) exhibited PA positivity, with 40 (38.8%) classified as hypoxia. Hypoxia scores independently predicted DAS28-CRP ([Formula: see text] = 0.618, p = 0.001) and DAPSA ([Formula: see text] = 0.612, [Formula: see text]). The hypoxia-optimized PAUS score demonstrated superior correlation with disease activity indices compared to conventional US (DAS28-CRP: r = 0.615, p = 0.001 versus r = 0.474, p = 0.017; DAPSA: r = 0.743, [Formula: see text] versus r = 0.567, p = 0.003), alongside superior diagnostic accuracy for minimal disease activity (area under the curve, AUC 0.776 versus 0.614, p = 0.008) and low disease activity (AUC 0.853 versus 0.772, p = 0.009). This multimodal scoring system enhances the stratification of PsA disease activity by providing unique metabolic insights, offering a potential tool for therapeutic monitoring and guiding treat-to-target strategies.