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.

Methods: Fifty patients who underwent UBE-TLIF and 50 patients who underwent W-TLIF, each with >2 years of follow-up, were retrospectively analyzed. Outcomes included operative parameters, time to postoperative mobilization, paravertebral muscle atrophy and fat infiltration rates, clinical scores (Visual Analog Scale [VAS], Oswestry Disability Index [ODI], Japanese Orthopaedic Association [JOA]), modified Macnab criteria, fusion rates, and complications. Results: Compared with W-TLIF, the UBE-TLIF group had significantly less intraoperative blood loss, shorter operative times, and lower postoperative drainage volumes (p <0.05). The UBE-TLIF group showed faster postoperative recovery and shorter hospital stays. At 6 months, 1 year, and 2 years, W-TLIF patients had higher multifidus and erector spinae atrophy, and greater paravertebral muscle fat infiltration (p <0.05). The UBE-TLIF group also had lower VAS and ODI scores at 1 year and 2 years (p <0.05) and fewer surgical complications (6% vs. 10%). Fusion rates (94% vs. 92%) and modified Macnab outcomes (88% vs. 86%) were comparable (p >0.05). Conclusions UBE-TLIF is associated with reduced intraoperative trauma, quicker recovery, and fewer complications. In the long-term, it better preserves paravertebral muscle integrity and provides superior pain and functional outcomes.

Pancreatic cancer （PC）， a highly malignant tumor of the digestive system， is characterized by extremely high incidence and mortality rates， imposing a severe burden on the healthcare system. Despite significant advancements in comprehensive therapeutic strategies for PC， clinical benefits for patients remain limited due to the high heterogeneity and unique microenvironment of the tumor. Therefore， further elucidating the molecular mechanisms underlying PC progression and developing novel， effective therapeutic strategies have become an urgent necessity in the current research. Programmed cell death （PCD） is a genetically orchestrated process of autonomous cell demise that is crucial for maintaining cellular homeostasis. Recent studies have demonstrated that tumor cells can remodel the tumor microenvironment by modulating various PCD pathways—including apoptosis， autophagy， pyroptosis， and ferroptosis—thereby driving tumor proliferation， invasion， and drug resistance. Traditional Chinese medicine （TCM） possesses unique advantages in anti-tumor therapy， characterized by multi-component， multi-target， and multi-pathway regulation. Accumulating evidence suggests that TCM formulas and their natural active compounds can effectively delay the malignant progression of PC by modulating PCD in tumor cells. This article systematically summarizes the molecular mechanisms of PCD in PC progression and reviews the mechanisms by which TCM formulas and natural active compounds intervene in PC through PCD modulation. This review aims to provide theoretical support and future research directions for the prevention and treatment of PC with TCM.

Objective:To investigate the relationship between lipid levels and cognitive impairment in the elderly Chinese population using prospective cohort data.Methods:Based on the China-PAR (Prediction for Atherosclerotic Cardiovascular Disease Risk in China) cohort, this study included 24 380 individuals aged ≥60 years who participated in the cognitive function follow-up survey from 2018 to 2019. Cognitive function was assessed using the Mini-Mental State Examination (MMSE), with cognitive impairment defined according to different educational levels: MMSE ≤17 for illiterate individuals, MMSE ≤20 for those with primary education and MMSE ≤24 for those with secondary education or above. Multivariable linear regression and logistic regression models were employed to examine the associations between six baseline lipid indicators and cognitive scores, as well as cognitive impairment. Additionally, restricted cubic splines were used to explore the exposure-dose relationship between lipid levels and cognitive function.Results:The study population had a median follow-up time of 11.6 years, with a baseline age of (59.7±6.8) years. Among the participants, 9 510 (39.0%) were males, and the mean MMSE score was 24.7±6.8. A total of 3 887 individuals (15.9%) were identified as cognitively impaired. The results of multivariable linear regression and logistic regression indicated that total cholesterol (TC), low-density lipoprotein cholesterol (LDL-C) and non-high-density lipoprotein cholesterol (non-HDL-C) levels were not only significantly positively associated with cognitive scores but also significantly associated with a lower risk of cognitive impairment. Each 1 mmol/L increase in these lipid levels corresponded to β values (95% CI) of 0.267 (0.173-0.361), 0.385(0.271-0.499) and 0.331(0.231-0.431), respectively. Each 1 mmol/L increase in these lipid levels corresponded to odds ratio ( OR) (95% CI) values of 0.915 (0.876-0.956), 0.875 (0.830-0.923) and 0.886 (0.848-0.927), respectively. The dose-response curve demonstrated that the negative association was primarily observed within the guideline-recommended optimal lipid level range. Specifically, when LDL-C was less than 3.4 mmol/L and non-HDL-C was less than 4.1 mmol/L, the corresponding OR (95% CI) values were 0.859 (0.796-0.926) and 0.876 (0.818-0.939). Conclusion:Lipid levels exhibit a certain linear negative association with cognitive impairment in elderly Chinese adults, with LDL-C and non-HDL-C demonstrating a stronger effect, particularly within the guideline-recommended optimal range.

Objective:To investigate clinical efficacy and safety of dupilumab in the treatment of moderate to severe atopic dermatitis (AD) complicated with asthma.Methods:A self-controlled study before and after treatment was conducted to retrospectively analyze 45 patients with moderate to severe atopic dermatitis combined with asthma who received dupilumab in the respiratory allergy clinic of North Theater Command General Hospital from January 2021 to May 2024, which age ≥12 years, including 27 males, 18 females. The treatment period was 4 to 12 months. All patients were treated with dupilumab combined with inhaled glucocorticoids and long-acting beta2-receptor agonists, as well as symptomatic drugs for atopic dermatitis. Analyze the clinical data of the patients before and after treatment, including lung function, asthma and AD-related assessment scales. Generalized estimation equation was used to analyze the simple effect of time on the repeated measurement data following non-normal distribution, and Wilcoxon signed rank test was used to compare the differences of each observation index before and after treatment.Results:Among 45 patients with moderate to severe atopic dermatitis complicated with asthma, after treatment with dupilumab, the FEV 1 increased from 2.39 (1.87, 2.83) L at baseline to 2.50 (1.84, 2.97) L 3 months after treatment ( Z=2.417, P=0.016), 2.60 (1.95, 3.14) L 6 months after treatment ( Z=2.896, P=0.004); the FEV 1pred% increased from 74.10% (67.70%, 78.75%) at baseline to 77.09% (68.40%, 80.24%) at 3 months after treatment ( Z=2.574, P=0.010), and 77.20% (71.10%, 80.72%) at 6 months after treatment ( Z=2.861, P=0.004). Meanwhile, there were statistically significant differences in the ACT and Mini-AQLQ scales at 3, 6, and 12 months after treatment compared with those before treatment (ACT score Z=3.170, 4.216, 5.723; Mini-AQLQ score Z=3.231, 4.133, 5.826; all P<0.05). The EASI scale decreased from baseline 25.90 (18.95, 33.45) to 6.20 (1.15, 8.35) at 4 months after treatment ( Z=5.842, P<0.05) and 4.90 (2.75, 8.35) at 6 months after treatment ( Z=5.841, P<0.05), 4.00 (3.15, 5.05) at 12 months after treatment ( Z=5.841, P<0.05); The scores of each scale of IGA, NRS and DLQI decreased significantly compared with the baseline after 4 months, 6 months and 12 months of treatment, and this trend became more obvious with the extension of treatment time. The differences were statistically significant (IGA score Z=6.247, 6.070, 5.946; NRS score Z=5.960, 5.893, 5.879; DLQI score Z=5.880, 5.850, 5.848; all P<0.05). During treatment, 1 patient had local adverse reactions at the injection site and 1 patient had conjunctivitis. Conclusion:Dupilumab may have a positive effect on improving the clinical efficacy of patients with moderate to severe atopic dermatitis complicated with asthma. During the 12-month observation period, this biological agent generally demonstrated good safety characteristics.

Objective:To analyze the epidemiological characteristics of cases involving monkey injuries at medical institutions surrounding Qianlingshan Park in Guiyang City, and to provide a reference basis for preventive measures to reduce monkey injuries and standardized post-exposure treatment.Methods:A retrospective cross-sectional study was conducted, collecting 1 900 cases of monkey-induced injuries in Qianlingshan Park treated at the outpatient clinic of Guizhou Provincial Center for Disease Control and Prevention and the Department of Surgery at Qianling Hospital, Guiyang City, from 2021 to 2024. Statistical analysis was performed using Pearson′s chi-square test.Results:Total of 1 900 cases of monkey-related injuries in Qianlingshan Park were collected from 2021 to 2024. The exposure time distribution exhibited significant seasonality, with 48.58% of cases occurring during July and August, totaling 923 cases, indicating a peak in the summer. There were 774 male patients and 1 126 female patients, with a ratio of 1∶1.45.and significant differences were observed between different age groups and genders (χ2=195.00, P<0.001), with the highest number of cases occurring in the 0-9 and 20-29 age groups, accounting for 22.05%(419 cases) and 21.79%(414 cases), respectively. The upper limbs were the most common injury site, accounting for 50.84% of the total cases(966 cases in total), with significant differences between gender and injury location (χ2=22.00, P<0.001), Among females, the proportion of injuries to the upper and lower limbs (30.11% and 16.47%, respectively) was higher than that among males (20.74% and 8.63%, respectively). The majority of injuries were classified as Grade Ⅲ, making up 57.38% of cases(1 069 cases in total). Self-treatment after exposure was the most common approach(60.44%), with significant differences observed between wound severity and treatment method (χ2=6.90, P=0.032), Patients with Grade Ⅱ and Grade Ⅲ wounds were more likely to choose self-management (26.84% and 33.23%, respectively) than outpatient management (15.14% and 24.15%). Approximately 98.05% (1 863 cases) of monkey-injured patients had received rabies vaccinations. Conclusions:This study analyzes monkey-related injuries in Qianlingshan Park from 2021 to 2024, clarifying the temporal distribution of injuries, demographic characteristics, injury sites, and treatment methods. The findings provide references for optimizing human-monkey conflict management and the prevention and control of zoonotic diseases in urban ecological parks.

Objective To investigate the relationship between serum microRNA-132(miRNA-132)and mi-croRNA-345(miRNA-345)expression,tumor markers and prognosis in gastric cancer patients.Methods A total of 103 gastric cancer patients from the hospital from January 2020 to January 2022 were selected as study group.82 patients with benign gastric diseases in the hospital during the same period were selected as the benign group.Another 61 healthy individuals who underwent physical examinations in the hospital during the same period were selected as control group.Real-time polymerase chain reaction(qPCR)was used to measure the expression of miRNA-132 and miRNA-345.The levels of carcinoembryonic antigen(CEA),carbohydrate anti-gen 199(CA199)and glycoprotein antigen 724(CA724)were measured using electrochemiluminescence.All patients were followed up until June 2024,and the overall survival(OS)of gastric cancer patients was calcu-lated through outpatient or telephone follow-up.Results The relative expression level of miRNA-132 in the serum of the study group was higher than that of the benign group and the control group(P＜0.05),while the relative expression level of miRNA-345 was lower than that of the benign group and the control group(P＜0.05).The serum levels of CEA,CA199,and CA724 in the study group were higher than those in the be-nign group and the control group(P＜0.05).Pearson analysis showed that miRNA-132 was positively correla-ted with CEA,CA199,and CA724(P＜0.05),while miRNA-345 was negatively correlated with CEA,CA199,and CA724(P＜0.05).Patients with a relative expression level of miRNA-132≤2.76 had better OS than those with a relative expression level of miRNA-132＞2.76(P＜0.05).Patients with a relative expres-sion level of miRNA-132 greater than 0.37 have better OS than those with a relative expression level of miR-NA-132≤0.37(P＜0.05).High expression of miRNA-132 and low expression of miRNA-345 were risk fac-tors affecting the prognosis of gastric cancer patients(P＜0.05).Conclusion Patients with gastric cancer have high expression of miRNA-132 and low expression of miRNA-345,and the expression of miRNA-132 and miRNA-345 is closely related to tumor markers and prognosis,which is worthy of clinical reference.

Objective:To investigate the self-perceived cognitive decline status in the community population,and to explore the association between different dietary habits and self-perceived cognitive decline.Methods:A cross-sectional study was carried out in 11 879 community residents in the three regions of Weifang,Jining,and Zoucheng in Shandong Province.The Ascertain Dementia-8 and dietary habits information questionnaire were used to assess self-perceived cognitive decline and dietary habits,and their association were analyzed using single factor and multivariate logistic regression.Results:The detection rate of self-perceived cognitive decline was 21.4％.Lo-gistic regression showed that smoking in the past was positively associated with self-perceived cognitive decline(OR=1.40,95％CI:1.14-1.73).However,intake of fruits(often,OR=0.70,95％CI:0.52-0.94;everyday,OR=0.60,95％CI:0.44-0.81),nuts(daily,OR=0.62,95％CI:0.44-0.88),mushrooms(often,OR=0.74,95％CI:0.57-0.92)and high tryptophan foods(sometimes,OR=0.79,95％CI:0.68-0.91;everyday,OR=0.54,95％CI:0.34-0.87)were negatively associated with self-perceived cognitive decline.Conclusion:Smoking history might be a risk factor for self-perceived cognitive decline,and high frequency intake of fruits,nuts,mush-rooms,and high tryptophan foods might protective factors for it.

Objective:To investigate the effects of 10° head up tilt bed rest (HUT) on human cardiac function via 3D speckle tracking echocardiography (3D-STE), and to study the difference in cardiac function under the submaximal exercise load between the horizontal position and 10° HUBR.Methods:Thirty young healthy volunteers were recruited as the subjects, who were randomly divided into an 10° HUT exercise group and horizontal exercise group with 15 subjects in each. Subjects in both groups were asked to ride the bicycle ergometer in the 10° HUBR position and supine position respectively. The load started with 50 W and was increased by 25 W every 3 min until it reached the maximum of 125 W. Before the exercise (resting state), 1 min after the load was increased each time, and 3 min after exercise (recovery period), the following indices were collected: ①basic cardiac function indices: heart rate (HR), systolic blood pressure (SBP) and diastolic blood pressure (DBP), ②conventional cardiac ultrasound indices: left ventricular ejection fraction (LVEF), stroke volume (SV) and cardiac output (CO), ③left ventricular strain indices: global longitudinal strain (GLS), global circumferential strain (GCS), and global area strain (GAS) measured by 3D-SET. The changes of these indices in the 2 groups of subjects under different exercise loads were observed.Results:The differences in the major effect of the basic heart indices (HR, SBP and DBP), conventional cardiac ultrasound indices (LVEF, SV and CO) and left ventricular strain indices (GLS, GCS and GAS) in response to the exercise load were statistically significant ( F=194.90, 113.66, 17.19, P=0.017, 0.018, 0.001). With the increase of the exercise load, the basic heart indices and conventional cardiac ultrasound indices kept rising, the left ventricular strain indices reached the minimum under a moderate exercise load (75 W), HR, SBP and CO were higher than those of the resting state ( P<0.05 or 0.01). Both LVEF under exercise loads of 75, 100, 125 W and during recovery, and SV under exercise loads of 100, 125 W and during recovery were significantly higher than those of the resting state ( P<0.05 or 0.01), while GLS and GCS under exercise loads of 50, 75, 125 W ( P<0.05 or 0.01), and GAS under exercise loads of 50, 75 W ( P<0.01) were significantly lower. There were statistically significant differences not only in GCS across the groups ( F=4.60, P=0.026) but also in DBP due to the interactions between the grouping and exercise loads ( F=3.13, P=0.031). DBP was higher than that of the resting state when the exercise load was 125 W in both groups. Conclusions:During submaximal exercise, myocardial contractility shows sustained enhancement with the increase of the exercise load. The results of GLS, GCS and GAS indicate that myocardial strain reaches its lowest value under a moderate exercise load, suggesting that moderate exercise can be used to evaluate cardiac function via 3D-SET. Under a simulated lunar gravity of 10° HUT, there is less deformation in the short axis direction of the myocardium, indicating that GCS can be used as a sensitive indicator to detect changes in cardiac function under different gravities.