Objective: To evaluate the effect of community comprehensive management model intervention among patients with dyslipidemia, so as to provide the reference for optimizing community management strategies and improving the target achievement rate for blood lipids among this population. Methods: From May to June 2023, a multi-stage stratified random sampling method was employed to select patients with dyslipidemia from primary healthcare institutions in Jiaxing City, Zhejiang Province. Eligible participants were randomly assigned to either a control group or an intervention group. The control group received routine management, while the intervention group was subjected to a community comprehensive management model in addition to the routine care. Both groups were followed up for 24 months. Data on demographic characteristics, lifestyle behaviors, physical examination indices, and blood biochemical indicators were collected at baseline and after the intervention through questionnaires, physical examinations, and laboratory tests. Changes in obesity rate, central obesity rate, target achievement rates for blood lipids, blood pressure, and blood glucose, as well as lifestyle modifications, were analyzed. Differences between the two groups before and after the intervention were assessed using generalized estimating equations (GEE). Results: The control group consisted of 560 patients, including 303 females (54.11%) and 430 individuals aged ≥65 years (76.79%). The intervention group also included 560 patients, with 300 females (53.57%) and 431 individuals aged ≥65 years (76.96%). Before the intervention, no statistically significant differences were observed between the two groups in terms of gender, age, educational level, history of chronic diseases, and atherosclerotic cardiovascular disease risk stratification (all P>0.05). After 24 months of intervention, interaction effects between group and time were observed for obesity rate, central obesity rate, target achievement rate for blood lipids, target achievement rate for blood glucose, composite target achievement rate, physical activity rate, and medication adherence (all P<0.05). Specifically, the intervention group demonstrated lower rates of obesity and central obesity, and higher target achievement rate of blood lipids, target achievement rate of blood glucose, composite target achievement rate, physical activity rate, and medication adherence compared to the control group. Conclusion The community comprehensive management model contributed to improvements in multiple metabolic parameters (including body weight, waist circumference, blood lipids, and blood glucose) among patients with dyslipidemia, and was associated with increased physical activity rate and medication adherence.

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 evaluate the predictive ability and clinical application value of artificial intelligence (AI) systems in the benign and malignant differentiation and pathological type of pulmonary nodules, and to summarize clinical application experience. Methods A retrospective analysis was conducted on the clinical data of patients with pulmonary nodules admitted to the Department of Thoracic Surgery, Second Hospital of Lanzhou University, from February 2016 to February 2025. Firstly, pulmonary nodules were divided into benign and non-benign groups, and the discriminative abilities of AI systems and clinicians were compared. Subsequently, lung nodules reported as precursor glandular lesions (PGL), microinvasive adenocarcinoma (MIA), and invasive adenocarcinoma (IAC) in postoperative pathological results were analyzed, comparing the efficacy of AI systems and clinicians in predicting the pathological type of pulmonary nodules. Results In the analysis of benign/non-benign pulmonary nodules, clinical data from a total of 638 patients with pulmonary nodules were included, of which there were 257 males (10 patients and 1 patient of double and triple primary lesions, respectively) and 381 females (18 patients and 1 patient of double and triple primary lesions, respectively), with a median age of 55.0 (47.0, 61.0) years. Different lesions in the same patient were analyzed as independent samples. Univariate analysis of the two groups of variables showed that, except for nodule location, the differences in the remaining variables were statistically significant (P<0.05). Multivariate logistic regression analysis showed that age, nodule type (subsolid pulmonary nodule), average density, spicule sign, and vascular convergence sign were independent influencing factors for non-benign pulmonary nodules, among which age, nodule type (subsolid pulmonary nodule), spicule sign, and vascular convergence sign were positively correlated with non-benign pulmonary nodules, while average density was negatively correlated with the occurrence of non-benign pulmonary nodules. The area under the receiver operating characteristic curve (AUC) of the malignancy risk value given by the AI system in predicting non-benign pulmonary nodules was 0.811, slightly lower than the 0.898 predicted by clinicians. In the PGL/MIA/IAC analysis, clinical data from a total of 411 patients with pulmonary nodules were included, of which there were 149 males (8 patients of double primary lesions) and 262 females (17 patients of double primary lesions), with a median age of 56.0 (50.0, 61.0) years. Different lesions in the same patient were analyzed as independent samples. Univariate analysis results showed that, except for gender, nodule location, and vascular convergence sign, the differences in the remaining variables among the three groups of PGL, MIA, and IAC patients were statistically significant (P<0.05). Multinomial multivariate logistic regression analysis showed that the differences between the parameters in the PGL group and the MIA group were not statistically significant (P>0.05), and the maximum diameter and average density of the nodules were statistically different between the PGL and IAC groups (P<0.05), and were positively correlated with the occurrence of IAC as independent risk factors. The average AUC value, accuracy, recall rate, and F1 score of the AI system in predicting lung nodule pathological type were 0.807, 74.3%, 73.2%, and 68.5%, respectively, all better than the clinical physicians’ prediction of lung nodule pathological type indicators (0.782, 70.9%, 66.2%, and 63.7% respectively). The AUC value of the AI system in predicting IAC was 0.853, and the sensitivity, specificity, and optimal cutoff value were 0.643, 0.943, and 50.0%, respectively. Conclusion This AI system has demonstrated high clinical value in predicting the benign and malignant nature and pathological type of lung nodules, especially in predicting lung nodule pathological type, its ability has surpassed that of clinical physicians. With the optimization of algorithms and the adequate integration of multimodal data, it can better assist clinical physicians in formulating individualized diagnostic and treatment plans for patients with lung nodules.

Objective: To investigate the detection of depressive symptoms and its influencing factors among middle school students in Huzhou City, so as to provide insights for improving the mental health levels among middle school students. Methods: From September to November 2024, a total of 4 729 middle school students from five counties (districts) in Huzhou City were selected through the stratified cluster random sampling method. Demographic information, lifestyle, and school bullying were collected through questionnaire surveys. Depressive symptoms were assessed using the Center for Epidemiological Studies Depression Scale (CES-D). Factors affecting depressive symptoms among middle school students were analyzed using a multivariable logistic regression model. Results: A total of 4 729 middle school students were surveyed, including 2 200 boys (46.52%) and 2 529 girls (53.48%). Depressive symptoms were detected in 1 026 students, with a detection rate of 21.70%. Multivariable logistic regression analysis showed that girl (OR=1.960, 95%CI: 1.659-2.317), high school (ordinary high school, OR=1.789, 95%CI: 1.465-2.186; vocational high school, OR=1.581, 95%CI: 1.105-2.263), consumption of sugar-sweetened beverages >0 time/day (<1 time/day, OR=1.363, 95%CI: 1.009-1.841; ≥1 time/day, OR=1.568, 95%CI: 1.098-2.239), fried food intake ≥1 time/day (OR=1.890, 95%CI: 1.291-2.769), skipping breakfast daily (OR=2.178, 95%CI: 1.825-2.599), TV viewing time ≥2 hours/day (OR=1.457, 95%CI: 1.154-1.838), insufficient sleep duration (OR=1.761, 95%CI: 1.422-2.181), smoking (OR=2.798, 95%CI: 1.834-4.269), alcohol consumption (OR=2.282, 95%CI: 1.861-2.798), experiencing school bullying (OR=5.440, 95%CI: 3.148-9.402) and parental physical/verbal abuse (OR=3.954, 95%CI: 3.189-4.902) were associated with a higher risk of depressive symptoms among middle school students. Conversely, the middle school students who engaged in moderate-to-vigorous physical activity ≥3 times/week (OR=0.784, 95%CI: 0.668-0.921) and attended physical education classes ≥3 sessions/week (OR=0.736, 95%CI: 0.613-0.884) were associated with a lower risk of depressive symptoms. Conclusion The prevalence of depressive symptoms among middle school students in Huzhou City was lower than national average, and was influenced by dietary habits, physical exercise, sleep duration, smoking, alcohol consumption, and experiencing school bullying.

Objective: To explore the association between mental health status and adverse childhood experiences (ACEs) among sexual minority college students, so as to provide a scientific basis for mental health education and health promotion in universities. Methods: From January to February 2024, convenience and cluster sampling methods were used to select 1 792 college students from 11 colleges in Guangxi. A self reporting method was applied to identify 476 sexual minority individuals. The Symptom Check-List 90 (SCL-90) and the Simplified Chinese Adverse Childhood Experiences International Questionnaire (SC-ACE-IQ) were employed to assess mental health and ACEs. Multivariate Logistic regression analysis was conducted to examine the associations. Results: The detection rates of all psychological issues among sexual minority college students in Guangxi were significantly higher than those of non sexual minority college students ( χ 2=56.01-91.39, all P <0.01). Except for physical neglect, bullying, and community violence, sexual minority students exhibited higher reporting rates of other ACEs types compared to nonsexual minority students ( χ 2= 4.52-13.34, all P <0.05). The total ACEs score for college students was 1.00 (1.00, 2.00), while the SCL-90 total score was 96.00 (113.00, 160.00). Spearman correlation analysis revealed a positive correlation between ACEs total scores and SCL-90 total scores ( r=0.29, P <0.05). Additionally, all ACEs subscales, including emotional neglect, physical neglect, emotional abuse, sexual abuse, parental loss, domestic violence, and community violence were positively correlated with corresponding SCL-90 subscale scores ( r =0.05-0.22, all P <0.05). Multivariate Logistic regression analysis showed that family violence increased the risk of mental health issues for sexual minority students ( OR=1.61, 95%CI =1.26-2.09); emotional neglect ( OR= 1.05 , 95%CI =1.00-1.10), physical neglect ( OR=1.20, 95%CI =1.06-1.35), sexual abuse ( OR=1.49, 95%CI =1.15-1.93) increased mental health risks for non sexual minority students (all P <0.05). The cumulative effects of ACEs were all statistically significant in the total sample and both subgroups (all P <0.05). Conclusion Mental health status among sexual minority college students in Guangxi is associated with ACEs, and their well being requires active attention

Objective　To analyse the crude detection rate and trends of newly detected HIV/AIDS cases in medical institutions in China from 2017 to 2023, and to provide a reference for optimizing HIV testing strategies in medical institutions. Methods　Data on HIV testing and newly reported HIV/AIDS cases were analysed using data from the Comprehensive AIDS Prevention and Control Information System of the China Information System for Disease Control and Prevention for the period from 2017 to 2023. HIV testing in medical institutions includes patients tested preoperatively, those tested before transfusion, those tested in sexually transmitted disease (STD) clinics, prenatal care clinics, and other types of patients. Descriptive statistical analysis and χ2 test were performed using SAS 9.4 software. Joinpoint regression was performed using Joinpoint 4.9.0 software to analyse trends of the crude detection rates over time. Results　From 2017 to 2023, the person-times of HIV tests in medical institutions increased from 143 million to 255 million, with an increase of 78.07%. The number of newly detected HIV/AIDS cases increased from 74 000 to 88 000 and then declined to 69 000. The crude detection rate of new HIV/AIDS cases declined from 5.18/10 000 to 2.71/10 000, showed a declining trend, the mean annual percentage change was -9.99%(P<0.001). The crude detection rate of new HIV/AIDS cases in STD clinics was the highest among all types of clinic visits (12.79/10 000-24.47/10 000), and the crude detection rate of new cases among all types of clinic visits showed a decreasing trend(P<0.05). Among different medical institutions, general hospitals were the most important source of the number of tests and the number of newly detected HIV/AIDS cases, accounting for more than 62.93% and 62.68%, respectively. Specialised medical institutions had the highest crude detection rate of new cases, which was maintained at more than 5.13/10 000. The crude detection rate of new cases for all four types of medical institutions, except for primary medical institutions, showed a decreasing trend (P<0.05). Conclusions　The detection rate of new cases in medical institutions showed a decreasing trend in 2017-2023, and the efficiency of STD clinics testing and detection was higher among all types of attendees. General hospitals are the main source of new cases detection, and testing in specialised medical institutions is more efficient. Testing should be strengthened in key groups of patients and in key medical institutions.

Objective: To investigate the current status and influencing factors of overweight and obesity among adults in Jiaxing City, Zhejiang Province, so as to provide a basis for developing targeted weight management measures. Methods: In 2024, a multistage stratified random cluster sampling method was employed to recruit permanent residents aged ≥18 years from Jiaxing City for questionnaire surveys. Data on basic information, lifestyle behaviors, and history of chronic diseases were collected. Height and body weight were measured, and overweight and obesity were determined based on body mass index (BMI). The influencing factors of overweight and obesity among adults were analyzed by a multivariable logistic regression model. Results: Totally 10 509 questionnaires were allocated, and 9 802 valid questionnaires were recovered, with an effective recovery rate of 93.27%. Among the respondents, 4 808 (49.05%) were males and 4 994 (50.95%) were females, with a mean age of (51.27±17.26) years. A total of 4 884 overweight and obesity individuals were identified, with a detection rate of 49.83%. Multivariable logistic regression analysis showed that gender (male, OR=1.719, 95%CI: 1.578-1.873), age (≥60 years, OR=0.802, 95%CI: 0.652-0.986), educational level (bachelor and above, OR=0.640, 95%CI: 0.518-0.791), marital status (being married/cohabiting, OR=1.224, 95%CI: 1.009-1.486), adequate nut intake (OR=0.910, 95%CI: 0.832-0.995), hypertension (OR=2.462, 95%CI: 2.219-2.732), and dyslipidemia (OR=1.629, 95%CI: 1.444-1.837) were statistically associated with overweight and obesity among adults. Conclusion The detected rate of overweight and obesity among adults in Jiaxing City was relatively high, and was mainly associated with gender, age, education level, marital status, nut intake, hypertension, and dyslipidemia.