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.

Background: Visceral white adipose tissue (vWAT) inflammation is a critical pathology of obesity-caused heart damage and is closely associated with adipocyte endoplasmic reticulum (ER) dysfunction. Serine (or cysteine) peptidase inhibitor, clade A, member 3C (Serpina3c) has been identified as an adipokine with anti-vWAT inflammatory effects. However, it remains unclear whether Serpina3c deficiency promotion of vWAT inflammation involves adipocyte ER dysfunction and whether it further contributes to heart damage in obesity. Methods: Wild type and Serpina3c knockout (Serpina3c–/–) mice were fed a high-fat diet (HFD) for 12 weeks. An adeno-associated virus (AAV) was injected locally into epididymal white adipose tissue (eWAT) of Serpina3c–/– mice to induce eWAT-adipocyte- specific overexpression of Serpina3c (AAV-Serpina3c) or knockdown of hypoxia-inducible factor 1α (AAV-shHIF1α). In vitro experiments were performed in 3T3-L1 adipocytes. Results: Serpina3c–/– mice exhibited more severe eWAT, serum and heart inflammation after HFD feeding. Consistently, these adverse phenotypes were mitigated in AAV-Serpina3c and AAV-shHIF1α mice. Mechanistically, ER oxidoreductase 1α (Ero1α) and protein disulfide isomerase (PDI) family members PDIA3 and PDIA4 were found to be target genes of HIF1α. In the obese mice, Serpina3c deficiency caused adipocyte more hypertrophy, and activated HIF1α-Ero1α/PDI mediated ER overoxidation and ER stress in eWAT. Subsequently, this led to increased adipocyte apoptosis and chemokine production and decreased adiponectin expression, which promoted macrophage infiltration and M1 polarization in eWAT, thus exacerbating eWAT inflammation and ultimately facilitating serum and distal heart inflammation. Conclusion These findings indicate that Serpina3c is a significant regulator of adipocyte ER redox homeostasis, thus highlighting Serpina3c as a potential therapeutic target for obesity-related eWAT inflammation and heart damage.

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: Humans are exposed to complex mixtures of environmental chemicals and other factors that can affect their health. Analysis of these mixture exposures presents several key challenges for environmental epidemiology and risk assessment, including high dimensionality, correlated exposure, and subtle individual effects. METHODS: We proposed a novel statistical approach, the generalized functional linear model (GFLM), to analyze the health effects of exposure mixtures. GFLM treats the effect of mixture exposures as a smooth function by reordering exposures based on specific mechanisms and capturing internal correlations to provide a meaningful estimation and interpretation. The robustness and efficiency was evaluated under various scenarios through extensive simulation studies. RESULTS: We applied the GFLM to two datasets from the National Health and Nutrition Examination Survey (NHANES). In the first application, we examined the effects of 37 nutrients on BMI (2011-2016 cycles). The GFLM identified a significant mixture effect, with fiber and fat emerging as the nutrients with the greatest negative and positive effects on BMI, respectively. For the second application, we investigated the association between four pre- and perfluoroalkyl substances (PFAS) and gout risk (2007-2018 cycles). Unlike traditional methods, the GFLM indicated no significant association, demonstrating its robustness to multicollinearity. CONCLUSION GFLM framework is a powerful tool for mixture exposure analysis, offering improved handling of correlated exposures and interpretable results. It demonstrates robust performance across various scenarios and real-world applications, advancing our understanding of complex environmental exposures and their health impacts on environmental epidemiology and toxicology.
Humans ; Environmental Exposure/analysis* ; Linear Models ; Nutrition Surveys ; Environmental Pollutants ; Body Mass Index

Objective To explore the correlation of bilateral knee joint strength asymmetry with balance,walking ability,and motor function in hemiplegic stroke patients,providing a reference for clinical assessment of stroke patients.Methods A total of 46 hemiplegic stroke patients admitted to the Rehabilitation Medicine Department of People's Hospital of Shijiazhuang from February to December 2023 were selected.According to the Berg Balance Scale(BBS)scores,patients were divided into Group A(BBS score≤20,n=23)and Group B(BBS score>20,n=23).The peak torque and differences of bilateral knee flexors and extensors were compared between two groups.Isokinetic technology was used to assess the differences in peak torque of bilateral knee joints at 60°/s and 120°/s.BBS,Functional Ambulation Classification(FAC),and Fugl-Meyer Assessment of Lower Extremity(FMA-LE)were used to evaluate patients'balance,walking ability,and lower limb motor function.The correlation between bilateral knee joint peak torque and its difference with the scores of three functional scales was analyzed.Results The peak torque of knee flexors and extensors at 60°/s in group A was significantly lower than that in group B(P<0.05).At both 60°/s and 120°/s the differences in peak torque between the healthy and affected sides of knee flexors and extensors were greater than those in group B(P<0.05).At 60°/s,the difference in peak torque of bilateral knee extensors in hemiplegic stroke patients was negatively correlated with the scores of BBS,FAC,and FMA-LE(r=-0.569,-0.582,-0.606,P<0.01),as did the knee flexors(r=-0.534,-0.386,-0.458,P<0.05).At 120°/s,similar negative correlations were observed for both knee extensors(r=-0.304,-0.304,-0.443,P<0.05)and flexors(r=-0.337,-0.349,-0.370,P<0.05).Conclusions Bilateral knee joint strength asymmetry in hemiplegic stroke patients is negatively correlated with balance and walking ability.The difference in strength between the two sides of knee joint may be one of the clinical indicators for evaluating the motor function of stroke patients.

Objective:To investigate the changes in N100, P300, and N400 of event-related potential(ERP) in children with obstructive sleep apnea (OSA), and provide the basis for evaluating cognitive and neurological impairment in pediatric OSA.Methods:Totally 108 children aged 5-10 years who visited the Sleep Center of Beijing Children's Hospital due to snoring or mouth breathing were recruited from June to September, 2023, and ultimately 90 children were included in the study.According to the obstructive sleep apnea hypopnea index (OAHI) in their polysomnography (PSG) results, children with OAHI>1 time/h were classified as OSA group ( n=74), and children with OAHI ≤ 1 time/h were classified as non-OSA group ( n=16).All participants completed the auditory oddball and Peabody image vocabulary test tasks, and the EEG data collected through ERP technology were compared between the two groups.SPSS 26.0 software was used for statistical analysis, and independent samples t-test or non parametric test was used for comparison between the two groups. Results:The P300 latency of OSA children in lead Fz was significantly longer than that of non OSA children (330.00(308.00, 396.00) ms, 309.00(294.50, 337.50)ms), and the difference was statistically significant ( Z=-2.143, P=0.032). The latency of P300 was positively correlated with apnea hypopnea index(AHI)(Fz lead: r=0.332, Cz lead: r=0.239, Pz lead: r=0.213, all P<0.05). There was no statistically significant difference in P300 latency between Cz and Pz leads ( Z=-1.615, P=0.106; Z=-1.055, P=0.291). There was no statistically significant difference in the amplitude of P300 among the leads (all P>0.05). There was no statistically significant difference in the amplitude and latency of N100 and N400 (both P>0.05). Conclusion:The latency of P300 in OSA children is significantly longer than that in non-OSA children, indicating impaired cognitive function. The latency of auditory P300 might serve as an early neuroelectrophysiological biomarker for identifying cognitive impairment in OSA children.

Aim To study the mechanism of salidro-side combined with rosavin in rats with ischemic stroke.Methods The MCAO rats was established by using thread-embolic method.The rats were divided into the sham group,MCAO group,salidroside com-bined with rosavin group,and positive control group;the drug was given continuously for seven days.Western blot was used to detect apoptosis indicators.Proteomics was used to analyse differential proteins(DEPs).STEP receptor inhibitor was injected into the lateral ventricles,the rats were administered for seven days,then the apoptosis indicators were detected.Re-sults Salidroside combined with rosavin could reduce neurological function scores in MCAO rats and inhibit cell apoptosis.Quantitative proteomics identified 496 DEPs in brain tissue and discovered core proteins STEP,p38,and CRTC1.Salidroside combined with rosavin could promote the STEP and CRTC1 while in-hibiting p38 protein.After treatment with STEP inhibi-tor,those effects were reversed.Conclusion Salidro-side combined with rosavin can inhibit cell apoptosis in MCAO rats,which is closely related to the regulation of the STEP/p38/CRTC1 signaling pathway.

Aim To study the mechanism of salidro-side combined with rosavin in rats with ischemic stroke.Methods The MCAO rats was established by using thread-embolic method.The rats were divided into the sham group,MCAO group,salidroside com-bined with rosavin group,and positive control group;the drug was given continuously for seven days.Western blot was used to detect apoptosis indicators.Proteomics was used to analyse differential proteins(DEPs).STEP receptor inhibitor was injected into the lateral ventricles,the rats were administered for seven days,then the apoptosis indicators were detected.Re-sults Salidroside combined with rosavin could reduce neurological function scores in MCAO rats and inhibit cell apoptosis.Quantitative proteomics identified 496 DEPs in brain tissue and discovered core proteins STEP,p38,and CRTC1.Salidroside combined with rosavin could promote the STEP and CRTC1 while in-hibiting p38 protein.After treatment with STEP inhibi-tor,those effects were reversed.Conclusion Salidro-side combined with rosavin can inhibit cell apoptosis in MCAO rats,which is closely related to the regulation of the STEP/p38/CRTC1 signaling pathway.

Objective:To investigate the changes in N100, P300, and N400 of event-related potential(ERP) in children with obstructive sleep apnea (OSA), and provide the basis for evaluating cognitive and neurological impairment in pediatric OSA.Methods:Totally 108 children aged 5-10 years who visited the Sleep Center of Beijing Children's Hospital due to snoring or mouth breathing were recruited from June to September, 2023, and ultimately 90 children were included in the study.According to the obstructive sleep apnea hypopnea index (OAHI) in their polysomnography (PSG) results, children with OAHI>1 time/h were classified as OSA group ( n=74), and children with OAHI ≤ 1 time/h were classified as non-OSA group ( n=16).All participants completed the auditory oddball and Peabody image vocabulary test tasks, and the EEG data collected through ERP technology were compared between the two groups.SPSS 26.0 software was used for statistical analysis, and independent samples t-test or non parametric test was used for comparison between the two groups. Results:The P300 latency of OSA children in lead Fz was significantly longer than that of non OSA children (330.00(308.00, 396.00) ms, 309.00(294.50, 337.50)ms), and the difference was statistically significant ( Z=-2.143, P=0.032). The latency of P300 was positively correlated with apnea hypopnea index(AHI)(Fz lead: r=0.332, Cz lead: r=0.239, Pz lead: r=0.213, all P<0.05). There was no statistically significant difference in P300 latency between Cz and Pz leads ( Z=-1.615, P=0.106; Z=-1.055, P=0.291). There was no statistically significant difference in the amplitude of P300 among the leads (all P>0.05). There was no statistically significant difference in the amplitude and latency of N100 and N400 (both P>0.05). Conclusion:The latency of P300 in OSA children is significantly longer than that in non-OSA children, indicating impaired cognitive function. The latency of auditory P300 might serve as an early neuroelectrophysiological biomarker for identifying cognitive impairment in OSA children.