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.

Hepatolenticular degeneration is a rare disease,and the number of cases of primary liver cancer occurring on the basis of liver cirrhosis caused by hepatolenticular degeneration is very small.This paper reports a case of hepatolenticular degeneration with primary liver cancer,and then reviews and summarizes current cases of this disease both domestically and internationally.
Humans ; Hepatolenticular Degeneration/complications* ; Liver Neoplasms/complications*

Objective To propose a method for mass spectrometry imaging(MSI)super-resolution reconstruction based on a spatially multi-level and self-supervised deep learning network(SMSDL-Net),aiming to improve the resolution of mass spectrometry images.Methods SMSDL-Net firstly registered histological and mass spectrometry images using a nonlinear transformation.Then a multi-branch Vision Transformer(ViT)was utilized to extract hierarchical features of high-resolution histological images in a self-supervised manner.These features were subsequently combined with the paired low-resolution mass spectrometry data to construct a regression network,which could realize the prediction of high-resolution mass spectrometry information.To validate the performance of the proposed method,the results by the method were compared with those of the traditional bicubic interpolation(BI)methods based on interpolation processing and the deepFERE method based on convolution neural network(CNN)for super-resolution reconstruction of magnesium elemental image of metal mass spectrometry of human liver cancer samples,and the method was also applied to a mouse renal adenocarcinoma metabolite mass spectrometry imaging dataset.Results Compared with the traditional BI methods and the deepFERE multimodal method,the method proposed demonstrated the lowest root mean square error(RMSE=0.015),the highest structural similarity index measure(SSIM=0.84)and the highest R-squared value(R2=0.853)in reconstructing mass spectrometry images.The effectiveness of the method and its potential for precise tissue-specific distinction were validated using the mouse renal adenocarcinoma metabolite MSI dataset.Conclusion Compared with traditional single-modal and pixel-wise regression deep learning methods,the method proposed enhances the quality of high-resolution mass spectrometry image reconstruction and can serve as a novel method for super-resolution reconstruction in the field of mass spectrometry imaging.[Chinese Medical Equipment Journal,2025,46(4):1-8]

Objective To investigate the mechanism of Ginkgo biloba extracts(GB)in the treatment of cerebral ischemia-reperfusion(CIR)injury based on network pharmacology and animal experiments.Methods The intersection targets of CIR and GB were obtained from TCMSP,GeneCards and other databases.Cytoscape software and Metascape database were used to analyze and map the related targets.The model of transient middle cerebral arterial occlusion(t-MCAO)was constructed in mice by suture method,and the effects of GB on the neurological function of mice after t-MCAO were observed by the neurological deficit score and Morris water maze test.HE staining was used to observe the pathological structural changes of neuron in the hippocampus of mice,and Western blot was used to verify the signal pathways screened by network pharmacology.Results Network pharmacology predicted that GB contained 33 active ingredients,and 116 potential targets of GB in treatment of CIR included Caspase3,Bax,etc.In addition,GB may play a protective role through signaling pathways such as PI3K-Akt and AMPK.Animal experiments showed that GB treatment could significantly improve the neural function and learning spatial memory ability of mice,alleviate the brain histopathological injury,and activate p-Akt/Akt signaling pathway.Conclusion GB has the characteristics of multi-target and multi-pathways therapy for CIR,which may reduce neuronal apoptosis by activating the p-Akt/Akt signaling pathway.

BACKGROUND: Prior studies have shown that drivers with type 2 diabetes are more likely to be involved in motor vehicle collisions (MVCs) compared to the general population. Certain meteorological factors have been increasingly recognized as contributors to MVC risk. This study aims to examine the association of MVCs with temperature, rainfall, wind speed, and sunshine duration among drivers with type 2 diabetes. METHODS: Using Taiwan's National Health Insurance data (2019-2021), we identified individuals diagnosed with type 2 diabetes and linked their records to the Police-Reported Traffic Accident Registry to obtain daily MVC counts. Meteorological data were sourced from the Central Weather Administration. Associations between daily weather conditions and MVCs were assessed using a Distributed Lag Non-Linear Model. RESULTS: Over the 1,096-day study period, 170,468 MVC events involving drivers with type 2 diabetes were recorded. A U-shaped association was observed between same-day temperature and MVC rates. Compared with the reference temperature of 17.5 °C, both lower temperatures (≤15 °C; rate ratio [RR] = 1.014-1.053) and higher temperatures (≥30 °C; RR = 1.062) were associated with increased MVC risk. Rainfall showed an inverse relationship with MVCs. Compared with 70 mm of rainfall, the lowest MVC rate occurred at 129 mm (RR = 0.873), while the highest was on rain-free days (0 mm; RR = 1.068). Stronger effects were observed when lag periods up to 14 days were considered. Wind speed and sunshine duration were not significantly associated with MVC risk. CONCLUSIONS These findings suggest that drivers with type 2 diabetes should exercise greater caution on days with extreme temperatures or in days with lesser rainfall, as these conditions may elevate MVC risk.
Humans ; Diabetes Mellitus, Type 2/epidemiology* ; Taiwan/epidemiology* ; Accidents, Traffic/statistics & numerical data* ; Male ; Middle Aged ; Female ; Weather ; Aged ; Adult ; Temperature ; Risk Factors

OBJECTIVE: To explore the mechanism of colon dialysis with Yishen Decoction (YS) in improving the autophagy disorder of intestinal epithelial cells in chronic renal failure (CRF) in vivo and in vitro. METHODS: Thirty male SD rats were randomly divided into normal, CRF, and colonic dialysis with YS groups by a random number table method (n=10). The CRF model was established by orally gavage of adenine 200 mg/(kg•d) for 4 weeks. CRF rats in the YS group were treated with colonic dialysis using YS 20 g/(kg•d) for 14 consecutive days. The serum creatinine (SCr) and urea nitrogen (BUN) levels were detected by enzyme-linked immunosorbent assay. Pathological changes of kidney and colon tissues were observed by hematoxylin and eosin staining. Autophagosome changes in colonic epithelial cells was observed with electron microscopy. In vitro experiments, human colon cancer epithelial cells (T84) were cultured and divided into normal, urea model (74U), YS colon dialysis, autophagy activator rapamycin (Ra), autophagy inhibitor 3-methyladenine (3-MA), and SIRT1 activator resveratrol (Re) groups. RT-PCR and Western blot were used to detect the mRNA and protein expressions of zonula occludens-1 (ZO-1), Claudin-1, silent information regulator sirtuin 1 (SIRT1), LC3, and Beclin-1 both in vitro and in vivo. RESULTS: Colonic dialysis with YS decreased SCr and BUN levels in CRF rats (P<0.05), and alleviated the pathological changes of renal and colon tissues. Expressions of SIRT1, ZO-1, Claudin-1, Beclin-1, and LC3II/I were increased in the YS group compared with the CRF group in vivo (P<0.05). In in vitro study, compared with normal group, the expressions of SIRT1, ZO-1, and Claudin-1 were decreased, and expressions of Beclin-1, and LC3II/I were increased in the 74U group (P<0.05). Compared with the 74U group, expressions of SIRT1, ZO-1, and Claudin-1 were increased, whereas Beclin-1, and LC3II/I were decreased in the YS group (P<0.05). The treatment of 3-MA and rapamycin regulated autophagy and the expression of SIRT1. SIRT1 activator intervention up-regulated autophagy as well as the expressions of ZO-1 and Claudin-1 compared with the 74U group (P<0.05). CONCLUSION Colonic dialysis with YS could improve autophagy disorder and repair CRF intestinal mucosal barrier injury by regulating SIRT1 expression in intestinal epithelial cells.
Animals ; Sirtuin 1/metabolism* ; Drugs, Chinese Herbal/therapeutic use* ; Autophagy/drug effects* ; Male ; Intestinal Mucosa/drug effects* ; Rats, Sprague-Dawley ; Epithelial Cells/metabolism* ; Colon/drug effects* ; Humans ; Kidney Failure, Chronic/drug therapy* ; Signal Transduction/drug effects* ; Renal Dialysis ; Rats ; Kidney/drug effects*

OBJECTIVE: To investigate chronic hepatitis C virus (HCV) infection's effect on gestational liver function, pregnancy and delivery complications, and neonatal development. METHODS: A total of 157 HCV antibody-positive (anti-HCV[+]) and HCV RNA(+) patients (Group C) and 121 anti-HCV(+) and HCV RNA(-) patients (Group B) were included as study participants, while 142 anti-HCV(-) and HCV RNA(-) patients (Group A) were the control group. Data on biochemical indices during pregnancy, pregnancy complications, delivery-related information, and neonatal complications were also collected. RESULTS: Elevated alanine aminotransferase (ALT) rates in Group C during early, middle, and late pregnancy were 59.87%, 43.95%, and 42.04%, respectively-significantly higher than Groups B (26.45%, 15.70%, 10.74%) and A (23.94%, 19.01%, 6.34%) ( P < 0.05). Median ALT levels in Group C were significantly higher than in Groups A and B at all pregnancy stages ( P < 0.05). No significant differences were found in neonatal malformation rates across groups ( P > 0.05). However, neonatal jaundice incidence was significantly greater in Group C (75.16%) compared to Groups A (42.25%) and B (57.02%) ( χ ² = 33.552, P < 0.001). HCV RNA positivity during pregnancy was an independent risk factor for neonatal jaundice ( OR = 2.111, 95% CI 1.242-3.588, P = 0.006). CONCLUSIONS Chronic HCV infection can affect the liver function of pregnant women, but does not increase the pregnancy or delivery complication risks. HCV RNA(+) is an independent risk factor for neonatal jaundice.
Humans ; Female ; Pregnancy ; Adult ; Pregnancy Complications, Infectious/epidemiology* ; Retrospective Studies ; Pregnancy Outcome ; Infant, Newborn ; Viremia/virology* ; Hepatitis C ; Hepacivirus/physiology* ; Hepatitis C, Chronic/virology* ; Young Adult ; Alanine Transaminase/blood*

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: Coronavirus disease 2019 (COVID-19) can result in fatigue and post-exertional malaise; however, whether severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection exacerbates lower urinary tract symptoms (LUTS) is unclear. This study investigated the association between prenatal SARS-CoV-2 infection and postpartum LUTS. METHODS: A multicenter, retrospective cohort study was conducted at two tertiary hospitals in China from November 1, 2022, to November 1, 2023. Participants were classified into infected and uninfected groups based on SARS-CoV-2 antigen results. LUTS prevalence and severity were assessed using self-reported symptoms and the Incontinence Impact Questionnaire-Short Form (IIQ-7). Pelvic floor muscle activity was measured using electromyography following the Glazer protocol. Group comparisons were performed to evaluate the association of SARS-CoV-2 infection with LUTS and electromyography parameters, with stratified analyses conducted using SPSS version 26.0. RESULTS: Among 3,652 participants (681 infected, 2,971 uninfected), no significant differences in LUTS prevalence or IIQ-7 scores were observed. However, SARS-CoV-2 infection was an independent factor influencing the electromyographic activity of the pelvic floor muscles (mean tonic contraction amplitudes), regardless of delivery mode ( P = 0.001). CONCLUSION Prenatal SARS-CoV-2 infection was not significantly associated with an increased risk of postpartum LUTS but independently altered pelvic floor muscle electromyographic activity, suggesting potential neuromuscular effects.
Humans ; Female ; COVID-19/epidemiology* ; Retrospective Studies ; Adult ; Pregnancy ; Lower Urinary Tract Symptoms/virology* ; Postpartum Period ; Pregnancy Complications, Infectious/virology* ; China/epidemiology* ; Electromyography ; SARS-CoV-2/physiology* ; Pelvic Floor/physiopathology* ; Prevalence