ObjectiveDiscriminating atypical hepatocellular carcinoma (HCC) from other malignancies in liver nodules classified as Liver Imaging Reporting and Data System category M (LR-M) remains a significant diagnostic challenge on conventional ultrasound examination. The LR-M category, originally intended to capture non-HCC malignancies, paradoxically contains up to 63% of atypical HCCs that deviate from classic enhancement patterns, leading to potential misdiagnosis and suboptimal treatment planning. While deep learning has shown promise in HCC diagnosis, most existing models rely exclusively on single-modality ultrasound, overlooking the diagnostic benefits of integrating complementary information from multiple imaging sources. To address this gap, we propose a novel attention-weighted tri-modal ultrasound network (TUS-Net) that integrates contrast-enhanced ultrasound (CEUS), B-mode ultrasound (BUS), and time-intensity curves (TICs) to improve diagnostic accuracy for these clinically challenging lesions. MethodsOur framework incorporates a three-dimensional convolutional neural network (C3D) backbone to extract spatiotemporal features from CEUS videos, capturing dynamic vascular patterns critical for lesion characterization. To effectively fuse complementary modalities, we introduce a dual-channel feature fusion module (DCFFM) that adaptively combines features from CEUS and BUS through channel-wise attention mechanisms, allowing the model to dynamically weigh the contribution of each modality based on diagnostic relevance. Additionally, we propose a temporal intensity feature fusion module (TIFFM) that leverages quantitative hemodynamic information from TICs to guide the model’s attention toward diagnostically critical temporal phases, such as arterial wash-in and portal venous washout. The model is further enhanced by automated lesion localization using YOLOX and class activation mapping for interpretability, ensuring that predictions align with clinically meaningful imaging features. ResultsEvaluated on a tri-modal ultrasound dataset comprising 161 patients with pathologically confirmed LR-M nodules (131 atypical HCC and 30 non-HCC malignancies), our model achieved an accuracy of 86.83%, a sensitivity of 92.50%, a specificity of 75.50%, and an AUC of 89.32% in screening atypical HCC. Compared to single-modality baselines, TUS-Net demonstrated superior specificity, a clinically critical metric given the higher risk associated with misclassifying non-HCC malignancies. Ablation studies confirmed the contribution of each module, with the full model outperforming both standard C3D and 3D ResNet backbones integrated with attention mechanisms. A reader study involving junior and senior radiologists further validated the clinical utility of AI assistance, showing consistent improvements in specificity and inter-reader consistency, particularly for less experienced clinicians. ConclusionThese results surpass existing benchmark models and demonstrate the potential of our approach to enhance diagnostic precision in clinically specific cases. By intelligently fusing multi-modal ultrasound data with attention-guided mechanisms, TUS-Net offers a reliable and interpretable tool that holds promise for improving the non-invasive diagnosis of atypical HCC in challenging LR-M liver nodules.

ObjectiveDiscriminating atypical hepatocellular carcinoma (HCC) from other malignancies in liver nodules classified as Liver Imaging Reporting and Data System category M (LR-M) remains a significant diagnostic challenge on conventional ultrasound examination. The LR-M category, originally intended to capture non-HCC malignancies, paradoxically contains up to 63% of atypical HCCs that deviate from classic enhancement patterns, leading to potential misdiagnosis and suboptimal treatment planning. While deep learning has shown promise in HCC diagnosis, most existing models rely exclusively on single-modality ultrasound, overlooking the diagnostic benefits of integrating complementary information from multiple imaging sources. To address this gap, we propose a novel attention-weighted tri-modal ultrasound network (TUS-Net) that integrates contrast-enhanced ultrasound (CEUS), B-mode ultrasound (BUS), and time-intensity curves (TICs) to improve diagnostic accuracy for these clinically challenging lesions. MethodsOur framework incorporates a three-dimensional convolutional neural network (C3D) backbone to extract spatiotemporal features from CEUS videos, capturing dynamic vascular patterns critical for lesion characterization. To effectively fuse complementary modalities, we introduce a dual-channel feature fusion module (DCFFM) that adaptively combines features from CEUS and BUS through channel-wise attention mechanisms, allowing the model to dynamically weigh the contribution of each modality based on diagnostic relevance. Additionally, we propose a temporal intensity feature fusion module (TIFFM) that leverages quantitative hemodynamic information from TICs to guide the model’s attention toward diagnostically critical temporal phases, such as arterial wash-in and portal venous washout. The model is further enhanced by automated lesion localization using YOLOX and class activation mapping for interpretability, ensuring that predictions align with clinically meaningful imaging features. ResultsEvaluated on a tri-modal ultrasound dataset comprising 161 patients with pathologically confirmed LR-M nodules (131 atypical HCC and 30 non-HCC malignancies), our model achieved an accuracy of 86.83%, a sensitivity of 92.50%, a specificity of 75.50%, and an AUC of 89.32% in screening atypical HCC. Compared to single-modality baselines, TUS-Net demonstrated superior specificity, a clinically critical metric given the higher risk associated with misclassifying non-HCC malignancies. Ablation studies confirmed the contribution of each module, with the full model outperforming both standard C3D and 3D ResNet backbones integrated with attention mechanisms. A reader study involving junior and senior radiologists further validated the clinical utility of AI assistance, showing consistent improvements in specificity and inter-reader consistency, particularly for less experienced clinicians. ConclusionThese results surpass existing benchmark models and demonstrate the potential of our approach to enhance diagnostic precision in clinically specific cases. By intelligently fusing multi-modal ultrasound data with attention-guided mechanisms, TUS-Net offers a reliable and interpretable tool that holds promise for improving the non-invasive diagnosis of atypical HCC in challenging LR-M liver nodules.

ObjectivePrimary liver cancer, predominantly hepatocellular carcinoma (HCC), is a significant global health issue, ranking as the sixth most diagnosed cancer and the third leading cause of cancer-related mortality. Accurate and early diagnosis of HCC is crucial for effective treatment, as HCC and non-HCC malignancies like intrahepatic cholangiocarcinoma (ICC) exhibit different prognoses and treatment responses. Traditional diagnostic methods, including liver biopsy and contrast-enhanced ultrasound (CEUS), face limitations in applicability and objectivity. The primary objective of this study was to develop an advanced, light-weighted classification network capable of distinguishing HCC from other non-HCC malignancies by leveraging the automatic analysis of brightness changes in CEUS images. The ultimate goal was to create a user-friendly and cost-efficient computer-aided diagnostic tool that could assist radiologists in making more accurate and efficient clinical decisions. MethodsThis retrospective study encompassed a total of 161 patients, comprising 131 diagnosed with HCC and 30 with non-HCC malignancies. To achieve accurate tumor detection, the YOLOX network was employed to identify the region of interest (ROI) on both B-mode ultrasound and CEUS images. A custom-developed algorithm was then utilized to extract brightness change curves from the tumor and adjacent liver parenchyma regions within the CEUS images. These curves provided critical data for the subsequent analysis and classification process. To analyze the extracted brightness change curves and classify the malignancies, we developed and compared several models. These included one-dimensional convolutional neural networks (1D-ResNet, 1D-ConvNeXt, and 1D-CNN), as well as traditional machine-learning methods such as support vector machine (SVM), ensemble learning (EL), k-nearest neighbor (KNN), and decision tree (DT). The diagnostic performance of each method in distinguishing HCC from non-HCC malignancies was rigorously evaluated using four key metrics: area under the receiver operating characteristic (AUC), accuracy (ACC), sensitivity (SE), and specificity (SP). ResultsThe evaluation of the machine-learning methods revealed AUC values of 0.70 for SVM, 0.56 for ensemble learning, 0.63 for KNN, and 0.72 for the decision tree. These results indicated moderate to fair performance in classifying the malignancies based on the brightness change curves. In contrast, the deep learning models demonstrated significantly higher AUCs, with 1D-ResNet achieving an AUC of 0.72, 1D-ConvNeXt reaching 0.82, and 1D-CNN obtaining the highest AUC of 0.84. Moreover, under the five-fold cross-validation scheme, the 1D-CNN model outperformed other models in both accuracy and specificity. Specifically, it achieved accuracy improvements of 3.8% to 10.0% and specificity enhancements of 6.6% to 43.3% over competing approaches. The superior performance of the 1D-CNN model highlighted its potential as a powerful tool for accurate classification. ConclusionThe 1D-CNN model proved to be the most effective in differentiating HCC from non-HCC malignancies, surpassing both traditional machine-learning methods and other deep learning models. This study successfully developed a user-friendly and cost-efficient computer-aided diagnostic solution that would significantly enhances radiologists’ diagnostic capabilities. By improving the accuracy and efficiency of clinical decision-making, this tool has the potential to positively impact patient care and outcomes. Future work may focus on further refining the model and exploring its integration with multimodal ultrasound data to maximize its accuracy and applicability.

Objective To investigate the role of the neutrophil-to-lymphocyte ratio(NLR)in predicting the in-hospital mortality risk of patients with acute aortic dissection(AAD)in multicenter hospitals.Methods A multicenter retrospective cohort study was conducted.Clinical data were collected from 2642 AAD patients who were hospitalized in five teaching hospitals:Tongji Hospital Affiliated to Tongji Medical College of Huazhong University of Science and Technology,Henan Provincial People's Hospital,Fuwai Central China Cardiovascular Hospital,the Third Affiliated Hospital of Xinxiang Medical University,and the Second Affiliated Hospital of Chongqing Medical University between August 2010 and December 2021.According to the quartiles of serum NLRlevels,the patients were divided into four groups:first quartile(Q1,n=660),second quartile(Q2,n=661),third quartile(Q3,n=661),and fourth quartile(Q4,n=660).The clinical characteristics and biochemical indicators of each group were compared.Partial correlation analysis was used to assess the relationship between NLR and cardiovascular parameters.Restricted cubic splines,Kaplan-Meier survival analysis,and Cox regression models were employed to evaluate the association between NLR levels and in-hospital mortality risk in AAD patients.Results The median age of all patients was 54[interquartile range(IQR):46-63]years,including 2096 males and 546 females.Compared with Q1-Q3 groups,patients inQ4group had a lower incidence of smoking history and diabetes history,and were more likely to have DeBakey type Ⅰ AAD(P＜0.05).Additionally,the levels of aspartate aminotransferase,high-density lipoprotein cholesterol,creatinine,and D-dimer in Q4 group were higher,while the levels of triglycerides and C-reactive protein(CRP)were lower(P＜0.01).The results of partial correlation analysis showed that the plasma NLR level was positively correlated with D-dimer(r=0.43,P＜0.01)and creatinine(r=0.16,P＜0.01).The restricted cubic spline function in the Cox model revealed a significant non-linear relationship between the plasma NLR level and clinical outcomes in AAD patients(P＜0.01).Kaplan-Meier survival analysis indicated that patients in Q4 group had the highest in-hospital mortality rate compared with Q1-Q3 groups(P＜0.0001).Furthermore,multivariate Cox regression analysis demonstrated that compared with Q1 group,the hazard ratio(HR)of NLR in Q4 group was 1.77(95％CI 1.33-2.37,P＜0.001),which was an independent risk factor for the primary endpoint events.Conclusion A higher plasma NLR level is significantly associated with the occurrence of cardiovascular events in AAD patients,and this association remains significant even after adjusting for potential confounding factors such as the multicenter visiting hospitals.

Objective To explore the effect of Xiongcan Yishen Formula on ferroptosis in testicular tissue of male rats with late-onset hypogonadism(LOH).Methods A total of 48 male Sprague-Dawley rats aged 6 months were randomly divided into model group,low-dose and high-dose Xiongcan Yishen Formula groups and propionate testosterone group,with 12 rats in each group.An additional 6 male Sprague-Dawley rats aged 8 weeks were set as the normal group.Except for the normal group,the rats were intraperitoneally injected with Cyclophosphamide at a dose of 20 mg/(kg·d)for 5 consecutive days to establish the LOH model,while the normal group received an equal volume of physiological saline for 5 days.The normal group and model group were given equal volumes of distilled water by gavage,while the low-dose and high-dose Xiongcan Yishen Formula groups were administered concentrated decoction at doses of 10.4 g and 41.6 g/kg/d respectively,and the propionate testosterone group received intramuscular injections of 5.21 mg/kg/d propionate testosterone,all for 28 consecutive days.ELISA was used to detect serum testosterone levels in rats,HE staining and transmission electron microscopy were used to observe the gross morphology of testicular tissue and the ultrastructure of Leydig cells,and RT-qPCR and Western blotting were used to detect the expression of ferroptosis-related genes and proteins in testicular tissue.Results The LOH model rats exhibited pathological changes such as atrophy of seminiferous tubules,structural disorder,and reduced spermatocytes in the lumen,as well as ultrastructural changes in Leydig cells including altered nuclear morphology,increased mitochondrial density,and reduced cristae.After intervention with Xiongcan Yishen Formula and propionate testosterone,the pathological changes in testis and the ultrastructure of Leydig cells were improved.Compared with the normal group,serum testosterone levels in the model group were significantly decreased(P＜0.05),the expression of ROS,ACSL4 mRNA and protein in testicular tissue was significantly increased,while the expression of FTH1,GPX4,and SLC7A11 mRNA and protein was significantly decreased(P＜0.05).Compared with the model group,ser-um testosterone levels in the low-dose and high-dose Xiongcan Yishen Formula groups and the propionate testosterone group were significantly increased(P＜0.05),and the expression of ROS,ACSL4 mRNA and protein was significantly decreased(P＜0.05);the high-dose Xiongcan Yishen Formula group showed significantly increased expression of FTH1,GPX4,and SLC7A11 mRNA and protein(P＜0.05).Conclusion Ferroptosis in testicular tissue is increased in LOH rats,and Xiongcan Yishen For-mula can elevate serum testosterone levels and improve pathological changes and ultrastructure in testicular tissue of LOH rats,possibly related to the inhibition of ferroptosis in testicular tissue of LOH rats.

The development of organ transplantation has brought new hope to many patients with organ failure and their families， but it has also raised numerous ethical issues. How to balance the rights and interests between organ donors and recipients， as well as ensure the fairness and transparency of the transplantation process has become an urgent problem to be solved. Based on the latest Regulations on Organ Donation and Transplantation and the Working Rules of the Ethics Committee for Human Organ Transplantation， the current difficulties and challenges in organ transplantation ethics were deeply analyzed. Taking the ethical review practice of Shenzhen Third People’s Hospital as an example， this paper explored issues such as full informed consent of both donors and recipients， risk assessment of marginal donors， and the review of relationships between donors and recipients. It also explored and constructed a set of complete ethical review models for organ transplantation through refined management. This model improved the efficiency and quality of ethical review as well as enriched the related knowledge system. It is expected that the implementation of this model can provide a reference for promoting effective ethical review nationwide， advancing the improvement and development of ethical review work in organ transplantation. Meanwhile， more medical ethics experts and practitioners are called upon to focus on and engage in the research and practice of ethical review in organ transplantation， jointly promoting progress in this field.

Turning to critical illness is a common stage of various diseases and injuries before death. Patients usually have complex health conditions, while the treatment process involves a wide range of content, along with high requirements for doctor′s professionalism and multi-specialty teamwork, as well as a great demand for time-sensitive treatments. However, this is not matched with critical care professionals and the current state of medical care in China. Telemedicine, which shortens the distance of medical professionals and the gap of disease diagnosis and treatments in various regions through electronic information, can effectively solve the current problem. Therefore, there is an urgent need to develop a standardized, high-quality visualization telemedicine round system .Therefore, experts have been organized to search domestic and foreign literature on telemedicine round for critically ill patients and to form this consensus based on clinical experiences so as to further improve the level of critical care treatments in regions.

Objective:To explore the longitudinal effect of learning stress on learning burnout in vocational college students, and the mediating role of academic procrastination.Methods:A total of 1 212 vocational college students were selected, and two follow-up surveys were conducted at 12-week intervals in September (T1) and November (T2) of 2024 using the basic situation questionnaire, the burnout inventory-student survey, the learning pressure questionnaire and the brief academic procrastination scale. SPSS 26.0 software was used to compare the demographic characteristics of students' depersonalization using t test and single factor analysis of variance. Bootstrap was used to analyze the relationship among learning stress, academic procrastination and learning burnout. Results:The scores of learning stress at T1 and T2 for vocational college students were 14.47±3.52 and 14.52 ±3.50, the scores of academic procrastination at T1 and T2 were 27.14±9.07 and 27.21±9.04, and the scores of learning burnout T1 and T2 were 39.38±8.76 and 39.69±8.79.The t-test showed that the score of learning burnout at T1 of students aged 18 and below (36.70±8.72) was lower than students aged 18 above (40.15±8.63, t=-5.81, P<0.01). The score of learning burnout for liberal arts students at T1(40.82±8.54) was higher than that of science students (37.68±8.72, t=6.31, P<0.01). Single factor analysis of variance showed that the score of learning burnout for grade 1 students at T1(35.19±8.45) was lower than that of grade 2 students (41.33±7.98) and grade 3 students (38.92±9.88), and learning burnout score of grade 2 students at T1 was higher than that of grades 3 students ( F=61.59, P<0.01). The score of learning burnout for high-achieving students at T1(36.23±8.34) was lower than that of middle-achieving students (39.82±8.52) and low-achieving students (45.42±9.14), and the score of learning burnout for middle-achieving students at T1 were lower than that of low-achieving students ( F=36.53, P<0.01). Bootstrap test showed that academic procrastination T2 played a partial mediating role in the relationship between learning stress T1 and learning burnout T2 (effect size=0.04, 95% CI=0.03-0.07). Academic procrastination T1 played a partial mediating role in the relationship between learning stress T1 and learning burnout T2 (effect size=0.05, 95% CI=0.04-0.07). Conclusion:Learning stress can directly affect learning burnout in vocational college students, and also can indirectly affect learning burnout through the mediating effect of academic procrastination.

Objective To understand the current situation of healthcare-associated infection(HAI)in China,pro-vide data support and decision-making basis for formulating scientific and effective strategies for HAI prevention and control.Methods A nationwide cross-sectional survey on HAI was conducted among various types and levels of medical institutions in China according to a unified protocol of bedside surveys and case investigations.Results In 2024,a total of 5 736 medical institutions and 2 751 765 patients were surveyed.Among them,34 889 HAI cases were identified,with a prevalence rate of 1.27％.The number of HAI episodes was 38 032,and case prevalence rate was 1.38％.The prevalence rate of HAI in medical institutions in different regions of China ranged from 0.66％to 2.35％.Among medical institutions of different scales,those with a bed capacity of ≥900 had the high-est incidence of HAI,reaching 1.65％.The most common infection site was the lower respiratory tract(44.66％),followed by the urinary tract(12.94％),surgical site(9.32％),upper respiratory tract(7.02％),and bloodstream infection(5.78％).The top 3 departments with the highest HAI rates were the general intensive care unit(10.02％),department of neurosurgery(5.51％),and department(group)of hematology(5.34％).A total of 23 238 strains of HAI pathogens were detected,with 10 714 strains(46.10％)from lower respiratory tract speci-mens.The top 5 detected strains were Klebsiella pneumoniae(14.76％),Pseudomonas aeruginosa(13.33％),Escherichia coli(12.79％),Acinetobacter baumannii(9.23％),and Staphylococcus aureus(7.88％).231 944 pa-tients underwent class Ⅰ incision surgery were monitored,with 1 647 cases experienced surgical site infection,and the prevalence rate of surgical site infection was 0.71％.The number of patients who should undergo pathogen de-tection(patients receiving therapeutic and therapeutic combined prophylactic antimicrobial agents)was 715 179,while the actual number was 480 492,with a pathogen detection rate of 67.18％.425 225 patients received patho-genic detection before treatment,with a detection rate of 59.46％.Conclusion The overall HAI prevalence in Chi-na is lower,showing disparities among medical institutions of different regions and scales.Therefore,precise imple-mentation of measures is necessary for HAI prevention and control,with a focus on high-risk institutions and high-risk departments,key areas,and critical procedures.All levels of medical institutions should continuously reduce the incidence of HAI by strengthening monitoring,standardizing the use of antimicrobial agents,and reinforcing basic HAI prevention and control measures.