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.

Objective:To explore the incidence of Crohn's disease-like pouch (CDP) after ileal pouch-anal anastomosis (IPAA) and analyze the clinical characteristics and risk factors in ulcerative colitis (UC) patients.Methods:A retrospective cohort study was conducted. One hundred and eighty-two UC patients undergoing IPAA at Jinling Hospital affiliated to Nanjing University from November 2003 to November 2024 were enrolled. Patients were categorized into CDP and non-CDP groups. Clinical features and prognosis were compared, and multivariate Cox regression was performed to identify risk factors for CDP.Results:A total of 182 UC patients were included, with a median follow-up time of 45.00 (30.00, 75.25) months. The patients were divided into two groups based on the diagnosis of CDP, with 23 patients (12.64%) in the CDP group and 159 patients (87.30%) in the non-CDP group. Compared to the non-CDP group, patients in the CDP group had a lower body mass index (BMI) ( Z=-2.87, P=0.004), and were more likely to develop early postoperative pouchitis (χ 2=4.50, P=0.034). The median time from ileostomy closure to the development of CDP was 12 .00 (6.00, 28.00) months. Cox regression analysis showed that a preoperative BMI<18.5 kg/m 2 ( HR=2.84, 95% CI: 1.24~6.49, P=0.013) and early postoperative pouchitis ( HR=3.11, 95% CI: 1.22~7.93, P=0.018) were associated with an increased risk of CDP. Conclusions:Preoperative low BMI and pouchitis occurring within 3 months postoperatively are significant risk factors for CDP. Close monitoring and early intervention are recommended for high-risk patients.

AIM To investigate the effects of rice wine type and wine processing method on chemical constituents and anti-coagulation effect of Angelicae sinensis Radix.METHODS Wine-washed products and wine-stir-fried products were prepared by different types and ages of rice wine,respectively,after which HPLC was adopted in the content determination of tryptophan,chlorogenic acid,vanillic acid,phthalic acid,ferulic acid,senkyunolide I,senkyunolide H,coniferyl ferulate and ligustilide,and PT,APTT,TT were detected in rabbit plasma.RESULTS Phenolic acids and volatile constituents demonstrated lower contents in the wine-stir-fried products than those in the raw product(P＜0.05),while those in the wine-washed products displayed no obvious changes(except for senkyunolide I)(P＞0.05).The contents of volatile constituents in the wine-washed products were higher than those in the wine-stir-fried products(P＜0.05).After being processed with dry rice wine,various constituents exhibited increased contents as compared with those after being processed with sweet rice wine(P＜0.05).Compared with the raw product,prolonged PT,APTT and TT were observable in the processed products prepared by 3-year semi-dry rice wine(P＜0.05).CONCLUSION The optimal rice wine type is determined to be 3-year semi-dry.Wine-washed Angelicae sinensis Radix shows high contents of ferulic acid and volatile constituents,whose activating blood and resolving stasis effect may be stronger.

AIM To investigate the clinical effects of Huoxue Qingre Formula combined with radiofrequency ablation on patients with thyroid nodules due to Phlegm Stagnation and Blood Stasis.METHODS Ninety patients were randomly assigned into control group(45 cases)for 3-month intervention of both Huoxue Qingre Formula placebo and radiofrequency ablation,and observation group(45 cases)for 3-month intervention of both Huoxue Qingre Formula and radiofrequency ablation.The changes in clinical effects,VRR,TCM syndrome score,cytokines(TNF-α,IFN-γ,IL-23,IL-1β),biochemical indices[Ca2+,25(OH)D,CRP],thyroid function indices(TSH,FT3,FT4)and safety indices were detected.RESULTS At postoperative 12 months,the observation group demonstrated higher total effective rate than the control group(P＜0.05).At postoperative 3-12 months,the two groups displayed increased VRR(P＜0.05),especially for the observation group(P＜0.05,P＜0.01).At the 7th,30th postoperative days,the two groups displayed decreased TCM syndrome score and cytokines(P＜0.05,P＜0.01),especially for the observation group(P＜0.05,P＜0.01).At the 30th postoperative day,the two groups displayed decreased Ca2+(P＜0.01)and increased 25(OH)D(P＜0.01),especially for the observation group(P＜0.01).At the 7th postoperative day,the two groups displayed increased CRP(P＜0.01),especially for the control group(P＜0.05),which were decreased at the 30th postoperative day(P＜0.05,P＜0.01),especially for the observation group(P＜0.01).No significant differences in thyroid function indices and safety indices were found between the two groups(P＞0.05).CONCLUSION For the patients with thyroid nodules due to Phlegm Stagnation and Blood Stasis,Huoxue Qingre Formula combined with radiofrequency ablation can safely and effectively enhance nodule VRR,alleviate discomfort symptoms,reduce TCM syndrome score,inflammatory factor levels,and accelerate recovery.

Objective To explore a deep learning-based automatic bone age estimation model for elbow joint X-ray images of Chinese Han adolescents and children and evaluate its performance.Methods A total of 943(517 males and 426 females)elbow joint frontal view X-ray images of Chinese Han ado-lescents and children aged 6.00 to<16.00 years were collected from East,South,Central and North-west China.Three experimental schemes were adopted for bone age estimation.Scheme 1:Directly in-put preprocessed images into the regression model;Scheme 2:Train a segmentation network using"key elbow joint bone annotations"as labels,then input segmented images into the regression model;Scheme 3:Train a segmentation network using"full elbow joint bone annotations"as labels,then in-put segmented images into the regression model.For segmentation,the optimal model was selected from U-Net,UNet++and TransUNet.For regression,VGG16,VGG19,InceptionV2,InceptionV3,ResNet34,ResNet50,ResNet101 and DenseNet121 models were selected for bone age estimation.The dataset was randomly split into 80%(754 samples)for training and validation for model fitting and hyperparameter tuning,and 20%(189 samples)as an internal test set to test the performance of the trained model.An additional 104 elbow joint X-ray images from the same demographic and age group were col-lected and used as an external test set.Model performance was evaluated by comparing the mean ab-solute error(MAE),root mean square error(RMSE),accuracies within±0.7 years(P±0.7 years)and±1.0 years(P±1.0 years)between the estimated age and the actual age,and by drawing radar charts,scat-ter plots,and heatmaps.Results When segmented with Scheme 3,the UNet++model achieved good segmentation performance with a segmentation loss of 0.000 4 and an accuracy of 93.8%at a learning rate of 0.000 1.In the internal test set,the DenseNet121 model with Scheme 3 yielded the best results with MAE,P±0.7 years and P±1.0 years being 0.83 years,70.03%,and 84.30%,respectively.In the external test set,the DenseNet121 model with Scheme 3 also performed best,with an average MAE of 0.89 years and an average RMSE of 1.00 years.Conclusion When performing automatic bone age estima-tion using elbow joint X-ray images in Chinese Han adolescents and children,it is recommended to use the UNet++model for segmentation.The DenseNet121 model with Scheme 3 achieves optimal per-formance.Using segmentation networks,especially that trained with annotation areas encompassing the full elbow joint including the distal humerus,proximal radius,and proximal ulna,can improve the ac-curacy of bone age estimation based on elbow joint X-ray images.

Objective To explore a deep learning network model suitable for bone age estimation using shoulder joint X-ray images in Chinese Han adolescents.Methods A retrospective collection of 1 286 shoulder joint X-ray images of Chinese Han adolescents aged 12.0 to<18.0 years(708 males and 578 females)was conducted.Using random sampling,approximately 80%of the samples(1 032 cases)were selected as the training and validation sets for model learning,selection and optimization,and the other 20%samples(254 cases)were used as the test set to evaluate the model's generalization ability.The original single-channel shoulder joint X-ray images and dual-channel inputs combining original images with segmentation labels(manually annotated shoulder joint regions multiplied pixel-by-pixel with original images,followed by segmentation via the U-Net++network to retain only key shoulder joint region information)were respectively input into four network models,namely VGG16,ResNet18,ResNet50 and DenseNet121 for bone age estimation.Additionally,manual bone age estimation was con-ducted on the test set data,and the results were compared with the four network models.The mean absolute error(MAE),root mean square error(RMSE),coefficient of determination(R2),and Pear-son correlation coefficient(PCC)were used as main evaluation indicators.Results In the test set,the bone age estimation results of the four models with dual-channel input of shoulder joint X-ray images outperformed those with single-channel input in all four evaluation indicators.Among them,DenseNet121 with dual-channel input achieved best results with MAE of 0.54 years,RMSE of 0.82 years,R2 of 0.76,and PCC(r)of 0.88.Manual estimation yielded an MAE of 0.82 years,ranking second only to dual-channel DenseNet121.Conclusion The DenseNet121 model with dual-channel input combined with original images and segmentation labels is superior to manual evaluation results,and can effectively estimate the bone age of Chinese Han adolescents.

Purpose: The aim of this study was to investigate the feasibility of an intelligent handheld ultrasound (US) device for assisting non-expert general practitioners (GPs) in detecting carotid plaques (CPs) in community populations. Methods: This prospective parallel controlled trial recruited 111 consecutive community residents. All of them underwent examinations by non-expert GPs and specialist doctors using handheld US devices (setting A, setting B, and setting C). The results of setting C with specialist doctors were considered the gold standard. Carotid intima-media thickness (CIMT) and the features of CPs were measured and recorded. The diagnostic performance of GPs in distinguishing CPs was evaluated using a receiver operating characteristic curve. Inter-observer agreement was compared using the intragroup correlation coefficient (ICC). Questionnaires were completed to evaluate clinical benefits. Results: Among the 111 community residents, 80, 96, and 112 CPs were detected in settings A, B, and C, respectively. Setting B exhibited better diagnostic performance than setting A for detecting CPs (area under the curve, 0.856 vs. 0.749; P<0.01). Setting B had better consistency with setting C than setting A in CIMT measurement and the assessment of CPs (ICC, 0.731 to 0.923). Moreover, measurements in setting B required less time than the other two settings (44.59 seconds vs. 108.87 seconds vs. 126.13 seconds, both P<0.01). Conclusion Using an intelligent handheld US device, GPs can perform CP screening and achieve a diagnostic capability comparable to that of specialist doctors.

Purpose: The aim of this study was to investigate the feasibility of an intelligent handheld ultrasound (US) device for assisting non-expert general practitioners (GPs) in detecting carotid plaques (CPs) in community populations. Methods: This prospective parallel controlled trial recruited 111 consecutive community residents. All of them underwent examinations by non-expert GPs and specialist doctors using handheld US devices (setting A, setting B, and setting C). The results of setting C with specialist doctors were considered the gold standard. Carotid intima-media thickness (CIMT) and the features of CPs were measured and recorded. The diagnostic performance of GPs in distinguishing CPs was evaluated using a receiver operating characteristic curve. Inter-observer agreement was compared using the intragroup correlation coefficient (ICC). Questionnaires were completed to evaluate clinical benefits. Results: Among the 111 community residents, 80, 96, and 112 CPs were detected in settings A, B, and C, respectively. Setting B exhibited better diagnostic performance than setting A for detecting CPs (area under the curve, 0.856 vs. 0.749; P<0.01). Setting B had better consistency with setting C than setting A in CIMT measurement and the assessment of CPs (ICC, 0.731 to 0.923). Moreover, measurements in setting B required less time than the other two settings (44.59 seconds vs. 108.87 seconds vs. 126.13 seconds, both P<0.01). Conclusion Using an intelligent handheld US device, GPs can perform CP screening and achieve a diagnostic capability comparable to that of specialist doctors.

PURPOSE: To investigate the protective effect of sub-hypothermic mechanical perfusion combined with membrane lung oxygenation on ischemic hypoxic injury of yorkshire brain tissue caused by traumatic blood loss. METHODS: This article performed a random controlled trial. Brain tissue of 7 yorkshire was selected and divided into the sub-low temperature anterograde machine perfusion group (n = 4) and the blank control group (n = 3) using the random number table method. A yorkshire model of brain tissue injury induced by traumatic blood loss was established. Firstly, the perfusion temperature and blood oxygen saturation were monitored in real-time during the perfusion process. The number of red blood cells, hemoglobin content, NA⁺, K⁺, and Ca²⁺ ions concentrations and pH of the perfusate were detected. Following perfusion, we specifically examined the parietal lobe to assess its water content. The prefrontal cortex and hippocampus were then dissected for histological evaluation, allowing us to investigate potential regional differences in tissue injury. The blank control group was sampled directly before perfusion. All statistical analyses and graphs were performed using GraphPad Prism 8.0 Student t-test. All tests were two-sided, and p value of less than 0.05 was considered to indicate statistical significance. RESULTS: The contents of red blood cells and hemoglobin during perfusion were maintained at normal levels but more red blood cells were destroyed 3 h after the perfusion. The blood oxygen saturation of the perfusion group was maintained at 95% - 98%. NA⁺ and K⁺ concentrations were normal most of the time during perfusion but increased significantly at about 4 h. The Ca²⁺ concentration remained within the normal range at each period. Glucose levels were slightly higher than the baseline level. The pH of the perfusion solution was slightly lower at the beginning of perfusion, and then gradually increased to the normal level. The water content of brain tissue in the sub-low and docile perfusion group was 78.95% ± 0.39%, which was significantly higher than that in the control group (75.27% ± 0.55%, t = 10.49, p < 0.001), and the difference was statistically significant. Compared with the blank control group, the structure and morphology of pyramidal neurons in the prefrontal cortex and CA1 region of the hippocampal gyrus were similar, and their integrity was better. The structural integrity of granulosa neurons was destroyed and cell edema increased in the perfusion group compared with the blank control group. Immunofluorescence staining for glail fibrillary acidic protein and Iba1, markers of glial cells, revealed well-preserved cell structures in the perfusion group. While there were indications of abnormal cellular activity, the analysis showed no significant difference in axon thickness or integrity compared to the 1-h blank control group. CONCLUSIONS Mild hypothermic machine perfusion can improve ischemia and hypoxia injury of yorkshire brain tissue caused by traumatic blood loss and delay the necrosis and apoptosis of yorkshire brain tissue by continuous oxygen supply, maintaining ion homeostasis and reducing tissue metabolism level.
Animals ; Perfusion/methods* ; Disease Models, Animal ; Brain Injuries/etiology* ; Swine ; Male ; Hypothermia, Induced/methods*