Comparison of clinical characteristics and outcomes in primary neuroendocrine breast carcinoma versus invasive ductal carcinoma.

Background: Neuroendocrine breast carcinoma (NECB) is a rare, special histologic type of breast cancer. There are some small sample studies on the clinical outcomes of NECB patients, which are worthy of further discussion. Methods: We conducted a retrospective case-control study of clinical characteristics and outcomes among patients with primary NECB versus invasive carcinoma of no special type (NST) between November 2004 and November 2017 in the Peking Union Medical College Hospital, Beijing. NST patients were strictly matched 1:4 during the same period based on the TNM stage. Statistical comparisons were performed to determine the differences in survival between NST and NECB patients and to identify clinical factors that correlate with prognosis. Results: A total of 121 participants affected by primary NECB were included in our analysis from November 2004 to November 2017. Elderly persons (>60 years of age) were more likely to have primary NECB than young persons (p=0.001). In addition, primary NECB patients had significantly higher odds of having tumors 2-5 cm (36.5%) and >5 cm (6.1%) in size than NST patients. Despite a significant difference in tumor size, the proportion of patients with lymph node metastases showed no difference between the two groups (p=0.021). In addition, the rate of patients with ER-negative tumors in the NECB group (4.2%) was significantly lower than that in the primary NST group (29.8%). Significant differences were noted in the PR-negative (13.3% versus 36.6%, P<0.001) and HER2-negative (90.5% versus 76.4%, P=0.001) expression statuses among these patients. Of 121 primary NECB patients, 11 (9.1%) experienced relapses during the follow-up period. We found that tumor size was an independent risk factor for relapse. For hormone receptors on tumor cells, ER-positive breast cancer patients had significantly lower odds of relapse than receptor-negative patients. Conclusions: Our data demonstrate no significant difference in mortality and relapse between the primary NECB and NST groups. The tumor size in the primary NECB group was significantly larger than that in the NST group. In addition, the absence of ER independently increased the relapse rate for breast carcinoma patients.

Ustekinumab in the Treatment of Generalized Pustular Psoriasis in a Pregnant Patient.

Although the use of biologics has led to great improvement in psoriasis patients, the treatment of psoriasis during pregnancy still faces many challenges. We herein report on a 29-year-old pregnant woman treated with ustekinumab for generalized pustular psoriasis. Upon becoming pregnant, the woman underwent continued treatment with ustekinumab in the first trimester. We also considered the need for neonatal vaccination. The patient discontinued ustekinumab therapy in the second trimester, and during the period of drug discontinuation we noted a slight rash recurrence. The patient was treated with ultraviolet B phototherapy and topical corticosteroids, and the rash was localized to the abdomen. However, in the 27th week of pregnancy, the patient was infected with COVID-19, which made the condition worse. The rash erupted rapidly and spread throughout her body, and she experienced a high fever with her blood count showing augmented numbers of white blood cells. The patients self-administered 0.3 g of acetaminophen three times per day, and after four days her core body temperature was 38.0°C; the rash, however, did not diminish. We diagnosed an outbreak of generalized pustular psoriasis and treated the patient with ustekinumab. The rash resolved quickly, and a healthy newborn was delivered by caesarean section at 39 weeks.

Head-to-head comparison of three different US-based quantitative parameters for hepatic steatosis assessment: a prospective study.

PURPOSE: To evaluate the diagnostic performance of attenuation coefficient (AC), hepato-renal index (HRI) and controlled attenuation parameter (CAP) in quantitative assessment of hepatic steatosis by employing histopathology as reference standard. METHODS: Participants with suspected metabolic-associated fatty liver disease (MAFLD) who underwent US-based parameter examinations and liver biopsy were prospectively recruited. The distributions of US parameters across different grades of steatosis were calculated, and diagnostic performance was determined based on the areas under the receiver operating characteristic curve (AUC). RESULTS: A total of 73 participants were included, with hepatic steatosis grades S0, S1, S2, and S3 distributed as follows: 13, 20, 27, and 13 respectively. The correlation coefficients for CAP, AC, and HRI ranged from 0.67 to 0.74. AC and HRI showed a strong correlation with steatosis grade. The AUC for CAP and AC in diagnosing steatosis ≥ S1 were significantly higher at 0.99 and 0.98 compared to HRI's value. For diagnosing steatosis ≥ S2, the AUC of CAP (AUC: 0.85) was lower than that of AC (AUC: 0.94), and HRI (AUC: 0.94). Similarly for diagnosing steatosis S3, the AUC of CAP (AUC: 0.68) was lower than that of AC (AUC: 0.88), and HRI (AUC: 0.88). CONCLUSION: The AC and HRI values increased with the progression of hepatic steatosis grade, while CAP increased from S0 to S2 but not from S2 to S3. For mild steatosis diagnosis, CAP and AC showed superior diagnostic performance compared to HRI, while AC and HRI were more advantageous in differentiating moderate and severe steatosis.

Signal2signal: Pushing the Spatiotemporal Resolution to the Limit by Single Chemical Hyperspectral Imaging.

There is growing interest in developing a high-performance self-supervised denoising algorithm for real-time chemical hyperspectral imaging. With a good understanding of the working function of the zero-shot Noise2Noise-based denoising algorithm, we developed a self-supervised Signal2Signal (S2S) algorithm for real-time denoising with a single chemical hyperspectral image. Owing to the accurate distinction and capture of the weak signal from the random fluctuating noise, S2S displays excellent denoising performance, even for the hyperspectral image with a spectral signal-to-noise ratio (SNR) as low as 1.12. Under this condition, both the image clarity and the spatial resolution could be significantly improved and present an almost identical pattern with a spectral SNR of 7.87. The feasibility of real-time denoising during imaging was well demonstrated, and S2S was applied to monitor the photoinduced exfoliation of transition metal dichalcogenide, which is hard to accomplish by confocal Raman spectroscopy. In general, the real-time denoising capability of S2S offers an easy way toward in situ/in vivo/operando research with much improved spatial and temporal resolution. S2S is open-source at https://github.com/3331822w/Signal2signal and will be accessible online at https://ramancloud.xmu.edu.cn/tutorial.

Predicting three or more metastatic nodes using contrast-enhanced lymphatic US findings in early breast cancer.

OBJECTIVES: To develop and validate a nomogram for predicting ≥ 3 metastatic axillary lymph nodes (ALNs) in early breast cancer with no palpable axillary adenopathy by clinicopathologic data, contrast-enhanced (CE) lymphatic ultrasound (US), and grayscale findings of sentinel lymph nodes (SLNs). MATERIALS AND METHODS: Women with T1-2N0 invasive breast cancer were consecutively recruited for the CE lymphatic US. Patients from Center 1 were grouped into development and internal validation cohorts at a ratio of 2:1. The external validation cohort was constructed from Center 2. The clinicopathologic data and US findings of SLNs were analyzed. A nomogram was developed to predict women with ≥ 3 metastatic ALNs. Nomogram performance was assessed with the area under the receiver operating characteristic curve (AUC) and calibration curve analysis. RESULTS: One hundred seventy-nine from Center 1 were considered the development cohorts. The remaining 90 participants from Center 1 were internal cohorts and 197 participants from Center 2 were external validation cohorts. The US findings of no enhancement (odds ratio (OR), 15.3; p = 0.01), diffuse (OR, 19.1; p = 0.01) or focal eccentric (OR, 27.7; p = 0.003) cortical thickening, and absent hilum (OR, 169.7; p < 0.001) were independently associated with ≥ 3 metastatic ALNs. Compared to grayscale US or CE lymphatic US alone, the nomogram showed the highest AUC of 0.88 (0.85, 0.91). The nomogram showed a calibration slope of 1.0 (p = 0.80-0.81; Brier = 0.066-0.067) in validation cohorts in predicting ≥ 3 metastatic ALNs. CONCLUSION: Patients likely to have ≥ 3 metastatic ALNs were identified by combining the lymphatic and grayscale US findings of SLNs. Our nomogram could aid in multidisciplinary treatment decision-making. TRIAL REGISTRATION: This trial is registered on www.chictr.org.cn : ChiCTR2000031231. Registered March 25, 2020. CRITICAL RELEVANCE STATEMENT: A nomogram combining lymphatic CEUS and grayscale US findings of SLNs could identify early breast cancer patients with low or high axillary tumor burden preoperatively, which is more applicable to the Z0011 era. Our nomogram could be useful in aiding multidisciplinary treatment decision-making for patients with early breast cancer. KEY POINTS: • CEUS can help identify and diagnose SLN in early breast cancer preoperatively. • Combining lymphatic and grayscale US findings can predict axillary tumor burden. • The nomogram showed a high diagnostic value in validation cohorts.

Long-term prognosis analysis of surgical therapy for bilateral synchronous multiple primary lung cancer: a follow-up of 293 cases.

Background: Bilateral synchronous multiple primary lung cancer (BSMPLC) presents significant clinical challenges due to its unique characteristics and prognosis. Understanding the risk factors that influence overall survival (OS) and recurrence-free survival (RFS) is crucial for optimizing therapeutic strategies for BSMPLC patients. Methods: We retrospectively analyzed clinical characteristics and treatment outcomes of 293 patients with BSMPLC who underwent surgical treatment between January 2010 and July 2017. Results: The 10-year OS and RFS rates were 96.1% and 92.8%, respectively. Preoperative forced expiratory volume in 1 second (FEV1) ≥70% [hazard ratio (HR), 0.214; 95% confidence interval (CI): 0.053 to 0.857], identical pathology types (HR, 9.726; 95% CI: 1.886 to 50.151), largest pT1 (HR, 7.123; 95% CI: 2.663 to 19.055), and absence of lymphovascular invasion (LVI; HR, 7.021; 95% CI: 1.448 to 34.032) emerged as independent predictors of improved OS. Moreover, the sum of tumor sizes less than or equal to 3 cm (HR, 6.229; 95% CI: 1.411 to 27.502) and absence of pleural invasion (HR, 3.442; 95% CI: 1.352 to 8.759) were identified as independent predictors of enhanced RFS. The presence or absence of residual nodules after bilateral surgery did not influence patients' OS (P=0.987) and RFS (P=0.054). Conclusions: Patients with BSMPLC who underwent surgery generally had a favorable prognosis. Whether or not to remove all nodules bilaterally does not affect the patient's long-term prognosis, suggesting the need for an individualized surgical approach.

Public Opinion Through Art: Exploring Chinese University Students' Perspectives on COVID-19 Mass Nucleic Acid Testing.

In the context of the global COVID-19 pandemic, this study focuses on Chinese university students, employing graphic elicitation as a qualitative research method to analyze their hand-drawn paintings and related descriptions. Augmented by A/r/tography and metacognitive methods, the research aims to unveil the participants' collective memory, as well as the perspectives and responses of these students to policies related to the pandemic. By specifically examining this particular demographic, the study incorporates Fairclough's ethical theory, applying deontological ethics, consequentialist ethics, and virtue ethics to establish a comprehensive framework for evaluating adjustments to pandemic response policies. This research not only enhances our understanding of how these university students perceive and adapt to COVID-19 policies but also provides valuable insights for decision-makers. The particular methodology, combining graphic elicitation and metacognitive research, contributes to policy assessment and ethical analysis, offering a nuanced perspective on the interplay between individual perceptions, policy responses, and ethical considerations amid the complexities of a public health crisis.

CircRNA and Stroke: New Insight of Potential Biomarkers and Therapeutic Targets.

Stroke, the second-largest cause of death and the leading cause of disability globally, presents significant challenges in terms of prognosis and treatment. Identifying reliable prognosis biomarkers and treatment targets is crucial to address these challenges. Circular RNA (circRNA) has emerged as a promising research biomarkers and therapeutic targets because of its tissue specificity and conservation. However, the potential role of circRNA in stroke prognosis and treatment remains largely unexplored. This review briefly elucidate the mechanism underlying circRNA's involvement in stroke pathophysiology. Additionally, this review summarizes the impact of circRNA on different forms of strokes, including ischemic stroke and hemorrhagic stroke. And, this article discusses the positive effects of circRNA on promoting cerebrovascular repair and regeneration, maintaining the integrity of the blood-brain barrier (BBB), and reducing neuronal injury and immune inflammatory response. In conclusion, the significance of circRNA as a potential prognostic biomarker and a viable therapeutic target was underscored.

Transgelin Promotes Glioblastoma Stem Cell Hypoxic Responses and Maintenance Through p53 Acetylation.

Glioblastoma (GBM) is a lethal cancer characterized by hypervascularity and necrosis associated with hypoxia. Here, it is found that hypoxia preferentially induces the actin-binding protein, Transgelin (TAGLN), in GBM stem cells (GSCs). Mechanistically, TAGLN regulates HIF1α transcription and stabilizes HDAC2 to deacetylate p53 and maintain GSC self-renewal. To translate these findings into preclinical therapeutic paradigm, it is found that sodium valproate (VPA) is a specific inhibitor of TAGLN/HDAC2 function, with augmented efficacy when combined with natural borneol (NB) in vivo. Thus, TAGLN promotes cancer stem cell survival in hypoxia and informs a novel therapeutic paradigm.

Effects of Twin-Block with an expanding device on the upper airway in growing children with skeletal class II malocclusion-a retrospective study based on the consistency of three-dimensional and two-dimensional data.

OBJECTIVES: Skeletal class II malocclusion is one of the most common malocclusions. Among the functional appliances for skeletal class II malocclusion, the Twin-Block appliance with a maxillary expander is effective in repositioning the mandible forward. In this study, we focused our efforts on investigating the effects of Twin-Block appliances with maxillary expanders on the upper airway in growing children with skeletal class II malocclusion by tracing and measuring lateral cephalograms after evaluating the consistency of three-dimensional CBCT data and two-dimensional lateral cephalogram data. MATERIALS AND METHODS: A total of 102 patients ranging from 9 to 15 years old (11.37 ± 2.80, male/female ratio = 1:1) with skeletal class II malocclusion were selected to evaluate the consistency of CBCT data and lateral cephalogram data. The strongly and moderately correlated segments were then selected to study the effects of Twin-Block with a maxillary expander on the upper airway in 66 growing children with skeletal class II malocclusion (11.31 ± 1.23 years old, male/female ratio = 1:1) by lateral cephalograms. RESULTS: The results showed a strong significant correlation in the nasopharynx (r = 0.708) and moderate significant correlations in the overall upper airway (r = 0.641), palatopharynx (r = 0.553), and glossopharynx (r = 0.575) but a weak correlation in the hypopharynx (r = 0.323). The corresponding determination coefficient (R2) was also determined by scatter plot analysis. Moreover, compared with the pretreatment data (T1), the total area of the upper airway and the areas of the nasopharynx, palatopharynx, and glossopharynx after functional treatment (T2) increased statistically and significantly. CONCLUSIONS: Lateral cephalograms can reflect the volume of the nasopharynx and oropharynx in skeletal class II children to a certain extent, while Twin-Block appliances with maxillary expanders can widen the volume of the nasopharynx and oropharynx significantly. CLINICAL RELEVANCE: The lateral cephalogram is reliable for analyzing the nasopharynx, palatopharynx, and glossopharynx in orthodontic clinical practice. Twin-Block appliances with maxillary expanders have a positive effect on skeletal class II patients with airway stenosis.

Predictive Modeling of Tensile Strength in Aluminum Alloys via Machine Learning.

Aluminum alloys are widely used due to their exceptional properties, but the systematic relationship between their grain size and their tensile strength has not been thoroughly explored in the literature. This study aims to fill this gap by compiling a comprehensive dataset and utilizing machine learning models that consider both the alloy composition and the grain size. A pivotal enhancement to this study was the integration of hardness as a feature variable, providing a more robust predictor of the tensile strength. The refined models demonstrated a marked improvement in predictive performance, with XGBoost exhibiting an R2 value of 0.914. Polynomial regression was also applied to derive a mathematical relationship between the tensile strength, alloy composition, and grain size, contributing to a more profound comprehension of these interdependencies. The improved methodology and analytical techniques, validated by the models' enhanced accuracy, are not only relevant to aluminum alloys, but also hold promise for application to other material systems, potentially revolutionizing the prediction of material properties.

Breast cancer pre-clinical screening using infrared thermography and artificial intelligence: a prospective, multicentre, diagnostic accuracy cohort study.

BACKGROUND: Given the limited access to breast cancer (BC) screening, the authors developed and validated a mobile phone-artificial intelligence-based infrared thermography (AI-IRT) system for BC screening. MATERIALS AND METHODS: This large prospective clinical trial assessed the diagnostic performance of the AI-IRT system. The authors constructed two datasets and two models, performed internal and external validation, and compared the diagnostic accuracy of the AI models and clinicians. Dataset A included 2100 patients recruited from 19 medical centres in nine regions of China. Dataset B was used for independent external validation and included 102 patients recruited from Langfang People's Hospital. RESULTS: The area under the receiver operating characteristic curve of the binary model for identifying low-risk and intermediate/high-risk patients was 0.9487 (95% CI: 0.9231-0.9744) internally and 0.9120 (95% CI: 0.8460-0.9790) externally. The accuracy of the binary model was higher than that of human readers (0.8627 vs. 0.8088, respectively). In addition, the binary model was better than the multinomial model and used different diagnostic thresholds based on BC risk to achieve specific goals. CONCLUSIONS: The accuracy of AI-IRT was high across populations with different demographic characteristics and less reliant on manual interpretations, demonstrating that this model can improve pre-clinical screening and increase screening rates.

The Prognostic Value of the Age-Adjusted Charlson Comorbidity Index Among the Elderly with Breast Cancer.

Purpose: This study aimed to assess the effect of comorbidities on prognosis using the Age-adjusted Charlson Comorbidity Index (ACCI) among the elderly with breast cancer (BC). Methods: This study included 745 patients divided into two groups following the ACCI score (≤3 vs >3). Multivariate logistic regression analysis was conducted for all kinds of outcomes, including BC-specific death (BCSD) and non-breast cancer-specific death (NBCSD). The Kaplan-Meier curves were plotted, and survival analysis was conducted for disease-free survival (DFS), overall survival (OS), BC-specific survival (BCSS), and non-BCSS (NBCSS). Results: A significantly higher NBCSD was found in the high-score (ACCI > 3) group than in the low-score (ACCI < 3) group (p = 0.032). The multivariate logistic regression analysis revealed ACCI score as an independent affecting factor for all-cause death (hazard ratio [HR] = 0.42, 95% confidence interval [CI]: 0.22-0.83, p = 0.012) and NBCSD (HR = 0.41, 95% CI: 0.20-0.87, p = 0.020). The Kaplan-Meier curves revealed statistical differences only in NBCSS between the two groups (p = 0.039). Subgroup analysis revealed a worse prognosis in the high-score group for OS and NBCSS among hormone receptor-positive participants and those who without undergoing axillary dissection or receiving chemotherapy (all p < 0.05). Multivariate Cox regression analysis revealed ACCI as an independent prognostic predictor for OS (HR = 2.18, 95% CI: 1.22-3.92, p = 0.009) and NBCSS (HR = 2.04, 95% CI: 1.02-4.08, p = 0.044). Conclusion: ACCI was indeed an effective indicator of the effects of comorbidities on survival among elderly patients with BC. However, the co-effect from age and comorbidities was not significant enough on cancer-specific prognosis, although it exerted a significant effect on treatments received.

A nomogram for individually predicting overall survival for elderly patients with early breast cancer: a consecutive cohort study.

Background: Elderly patients with breast cancer are highly heterogeneous, and tumor load and comorbidities affect patient prognosis. Prediction models can help clinicians to implement tailored treatment plans for elderly patients with breast cancer. This study aimed to establish a prediction model for breast cancer, including comorbidities and tumor characteristics, in elderly patients with breast cancer. Methods: All patients were ≥65 years old and admitted to the Peking Union Medical College Hospital. The clinical and pathological characteristics, recurrence, and death were observed. Overall survival (OS) was analyzed using the Kaplan-Meier curve and a prediction model was constructed using Cox proportional hazards model regression. The discriminative ability and calibration of the nomograms for predicting OS were tested using concordance (C)-statistics and calibration plots. Clinical utility was demonstrated using decision curve analysis (DCA). Results: Based on 2,231 patients, the 5- and 10-year OS was 91.3% and 78.4%, respectively. We constructed an OS prediction nomogram for elderly patients with early breast cancer (PEEBC). The C-index for OS in PEEBC in the training and validation cohorts was 0.798 and 0.793, respectively. Calibration of the nomogram revealed a good predictive capability, as indicated by the calibration plot. DCA demonstrated that our model is clinically useful. Conclusion: The nomogram accurately predicted the 3-year, 5-year, and 10-year OS in elderly patients with early breast cancer.

Multi-stage Unet segmentation and automatic measurement of pharyngeal airway based on lateral cephalograms.

OBJECTIVES: Orthodontic treatment profoundly impact the pharyngeal airway (PA) of patients. Airway examination is an integral part of daily orthodontic diagnosis, and lateral cephalograms (LC) are reliable to reveal PA structures. This study attempted to develop a simple method to help clinicians make a preliminary judgement of patients' PA conditions and assess the impact of orthodontic treatment on their airways. METHODS: LCs of 764 patients were used to train a multistage unit segmentation model. Another 130 images were used to validate the model and more 130 images were used to test the model. RESULTS: Unet was used as the backbone, with a mean dice value of 0.8180, precision of 0.8393, and recall of 0.8188. Furthermore, we identified seven key points and measured related indices. The length of the line separating the nasopharynx and oropharynx and the line separating the oropharynx and hypopharynx were manually measured thrice and the average values was compared. The intraclass correlation coefficient (ICC) for the two lines was 0.599 and 0.855. Then, we performed a single linear regression analysis, which indicated a strong correlation between the predictions and measurements for the two lines. CONCLUSIONS: This method is reliable for segmenting three regions (nasopharynx, oropharynx, and hypopharynx) of the PA and calculating related indices. However, the predictions obtained from this model still have errors, and it is necessary for clinical practitioners to assess and adjust the predictions. CLINICAL SIGNIFICANCE: Our model can help orthodontists formulate personalised treatment plans and evaluate the risk of airway stenosis during orthodontic treatment. This method may mark the beginning of a new and simpler approach for PA obstruction detection, specifically tailored to orthodontic patients.

Vehicle Trajectory Prediction via Urban Network Modeling.

Taxis are an important component of the transportation system, and empty taxis represent a significant waste of transportation resources. To alleviate the imbalance between supply and demand and relieve traffic congestion, real-time prediction of taxi trajectories is necessary. Most existing trajectory prediction studies focus on extracting time-series information but do not capture spatial information sufficiently. In this paper, we focus on the construction of an urban network and propose an urban topology-encoding spatiotemporal attention network (UTA) to address destination prediction problems. Firstly, this model discretizes the production and attraction units of transportation, combining them with key nodes in the road network to form an urban topological network. Secondly, GPS records are matched to the urban topological map to construct a topological trajectory, which significantly improves trajectory consistency and endpoint certainty, helping to model destination prediction problems. Thirdly, semantic information concerning surrounding space is attached to effectively mine the spatial dependencies of trajectories. Finally, after the topological encoding of city space and trajectories, this algorithm proposes a topological graph neural network to model the attention calculation with the trajectory context, comprehensively considering the spatiotemporal characteristics of the trajectories and improving prediction accuracy. We solve the prediction problems with the UTA model and also compare it with some other classical models, such as the HMM, RNN, LSTM, and transformer. The results suggest that all the models work well in combination with the proposed urban model (with a rough increase of 2%), while the UTA model is less affected by data sparsity.

Analysis of the metastatic mechanism and progress in the treatment of breast cancer liver metastasis: a narrative review.

Background and Objective: Breast cancer is the most common malignancy in women, and metastasis to other target organs is one of the main causes of death. Breast cancer liver metastasis (BCLM) has long been a research focus. Enhancing therapeutic effects, optimizing treatment plans and improving the prognosis of patients are major clinical challenges at present. Methods: We performed a comprehensive, nonsystematic review of the latest literature to define the current metastatic mechanism and related treatment advances of BCLM. Key Content and Findings: Due to the lack of research on the mechanism of BCLM, present treatment programs still have limited benefits, and the prognosis of patients is generally poor. New research directions and treatment ideas for BCLM are urgently needed. In this article, we indicated the specific procedures of the BCLM mechanism from the microenvironment to metastasis formation and progress in treatment, including drug therapies such as targeted therapy, surgery, intervention therapy and radiotherapy. Research on the molecular mechanism plays a crucial role in the development of BCLM-related therapies. Based on the metastasis process, we are able to propel new findings and further progression of antineoplastic drugs. Conclusions: The process of BCLM is multistep, and various factors are involved in it, which provides a powerful theoretical basis for the development of therapeutic methods for treatment of this disease. Further understanding of the mechanism of BCLM is essential to guide clinical management.

Risk stratification of gallbladder masses by machine learning-based ultrasound radiomics models: a prospective and multi-institutional study.

OBJECTIVE: This study aimed to evaluate the diagnostic performance of machine learning (ML)-based ultrasound (US) radiomics models for risk stratification of gallbladder (GB) masses. METHODS: We prospectively examined 640 pathologically confirmed GB masses obtained from 640 patients between August 2019 and October 2022 at four institutions. Radiomics features were extracted from grayscale US images and germane features were selected. Subsequently, 11 ML algorithms were separately used with the selected features to construct optimum US radiomics models for risk stratification of the GB masses. Furthermore, we compared the diagnostic performance of these models with the conventional US and contrast-enhanced US (CEUS) models. RESULTS: The optimal XGBoost-based US radiomics model for discriminating neoplastic from non-neoplastic GB lesions showed higher diagnostic performance in terms of areas under the curves (AUCs) than the conventional US model (0.822-0.853 vs. 0.642-0.706, p < 0.05) and potentially decreased unnecessary cholecystectomy rate in a speculative comparison with performing cholecystectomy for lesions sized over 10 mm (2.7-13.8% vs. 53.6-64.9%, p < 0.05) in the validation and test sets. The AUCs of the XGBoost-based US radiomics model for discriminating carcinomas from benign GB lesions were higher than the conventional US model (0.904-0.979 vs. 0.706-0.766, p < 0.05). The XGBoost-US radiomics model performed better than the CEUS model in discriminating GB carcinomas (AUC: 0.995 vs. 0.902, p = 0.011). CONCLUSIONS: The proposed ML-based US radiomics models possess the potential capacity for risk stratification of GB masses and may reduce the unnecessary cholecystectomy rate and use of CEUS. CLINICAL RELEVANCE STATEMENT: The machine learning-based ultrasound radiomics models have potential for risk stratification of gallbladder masses and may potentially reduce unnecessary cholecystectomies. KEY POINTS: • The XGBoost-based US radiomics models are useful for the risk stratification of GB masses. • The XGBoost-based US radiomics model is superior to the conventional US model for discriminating neoplastic from non-neoplastic GB lesions and may potentially decrease unnecessary cholecystectomy rate for lesions sized over 10 mm in comparison with the current consensus guideline. • The XGBoost-based US radiomics model could overmatch CEUS model in discriminating GB carcinomas from benign GB lesions.