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.

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.

Objective To understand the current situation of healthcare-associated infection(HAI)in China,pro-vide data support and decision-making basis for formulating scientific and effective strategies for HAI prevention and control.Methods A nationwide cross-sectional survey on HAI was conducted among various types and levels of medical institutions in China according to a unified protocol of bedside surveys and case investigations.Results In 2024,a total of 5 736 medical institutions and 2 751 765 patients were surveyed.Among them,34 889 HAI cases were identified,with a prevalence rate of 1.27％.The number of HAI episodes was 38 032,and case prevalence rate was 1.38％.The prevalence rate of HAI in medical institutions in different regions of China ranged from 0.66％to 2.35％.Among medical institutions of different scales,those with a bed capacity of ≥900 had the high-est incidence of HAI,reaching 1.65％.The most common infection site was the lower respiratory tract(44.66％),followed by the urinary tract(12.94％),surgical site(9.32％),upper respiratory tract(7.02％),and bloodstream infection(5.78％).The top 3 departments with the highest HAI rates were the general intensive care unit(10.02％),department of neurosurgery(5.51％),and department(group)of hematology(5.34％).A total of 23 238 strains of HAI pathogens were detected,with 10 714 strains(46.10％)from lower respiratory tract speci-mens.The top 5 detected strains were Klebsiella pneumoniae(14.76％),Pseudomonas aeruginosa(13.33％),Escherichia coli(12.79％),Acinetobacter baumannii(9.23％),and Staphylococcus aureus(7.88％).231 944 pa-tients underwent class Ⅰ incision surgery were monitored,with 1 647 cases experienced surgical site infection,and the prevalence rate of surgical site infection was 0.71％.The number of patients who should undergo pathogen de-tection(patients receiving therapeutic and therapeutic combined prophylactic antimicrobial agents)was 715 179,while the actual number was 480 492,with a pathogen detection rate of 67.18％.425 225 patients received patho-genic detection before treatment,with a detection rate of 59.46％.Conclusion The overall HAI prevalence in Chi-na is lower,showing disparities among medical institutions of different regions and scales.Therefore,precise imple-mentation of measures is necessary for HAI prevention and control,with a focus on high-risk institutions and high-risk departments,key areas,and critical procedures.All levels of medical institutions should continuously reduce the incidence of HAI by strengthening monitoring,standardizing the use of antimicrobial agents,and reinforcing basic HAI prevention and control measures.

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.

Objective:To investigate the structural changes in the spinal cord in multiple sclerosis (MS) and neuromyelitis optica spectrum disorders (NMOSD) and their relationship with clinical disability.Methods:This study was cross-sectional. A retrospective analysis of clinical and imaging data from 124 patients with MS (MS group), 101 patients with aquaporin-4 antibody-positive NMOSD (NMOSD group), and 110 healthy controls (HC group) from seven medical centers were conducted from January 2018 to October 2021. All subjects underwent 3D T 1WI, and the upper cervical spinal cord cross-sectional area (MUCCA) was segmented and measured. All patients completed the expanded disability status scale (EDSS) assessments at baseline and during follow-up, as well as the baseline 25-foot walk test (T25FW) and the nine-hole peg test (NHPT). Patients were classified into EDSS progression and non-progression groups based on follow-up EDSS scores. Comparisons of MUCCA among the three groups were conducted using analysis of covariance, controlling for age and sex as covariates. Pairwise comparisons between groups were performed using the HSD test. Univariate linear regression and logistic models were employed to identify candidate predictors of baseline clinical disability status or EDSS progression in the MS and NMOSD groups. L1 regularized multivariable linear regression analysis was used to determine independent predictors of baseline clinical disability status or EDSS progression. Independent predictors were then combined to establish a logistic regression model, and the model′s performance in predicting EDSS progression was evaluated using receiver operating characteristic analysis and the area under the curve (AUC). Results:A total of 144 patients completed follow-up EDSS assessments, with a follow-up duration of 3.30 (1.10, 6.42) years, including 82 patients in the MS group and 62 patients in the NMOSD group. Controlling for sex and age as covariates, the overall difference in MUCCA among the MS, NMOSD, and HC groups was statistically significant ( P=0.001). The MUCCA in the MS group was lower than that in the HC group, with a significant difference ( t=-2.54, P=0.007); the MUCCA in the NMOSD group was also lower than that in the HC group, with a significant difference ( t=-2.80, P=0.002). However, the difference in MUCCA between the MS and NMOSD groups was not statistically significant ( t=-0.40, P=0.882). In the MS group, MUCCA was an independent predictor of baseline EDSS score (β=-0.03), baseline T25FW score (β=-0.09), and baseline NHPT score (β=-0.30). In the NMOSD group, MUCCA (β=-0.08), age (β=0.06), and baseline EDSS score (β=-0.43) were independent predictors of EDSS progression, and the logistic regression model incorporating these three factors predicted EDSS progression with an AUC of 0.82. Conclusions:Significant spinal cord atrophy occurs in patients with both MS and NMOSD. Atrophy of the upper cervical spinal cord can predict the degree of disability in MS patients and the progression of clinical disability in NMOSD patients.

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.

Ischemic stroke (IS) is a globally life-threatening disease. Presently, few therapeutic medicines are available for treating IS, and rt-PA is the only drug approved by the US Food and Drug Administration (FDA) in the US. In fact, many agents showing excellent neuroprotection but no blood flow-improving activity in animals have not achieved ideal clinical efficacy, while thrombolytic drugs only improving blood flow without neuroprotection have limited their wider application. To address these challenges and meet the huge unmet clinical need, we have designed and identified a novel compound AAPB with dual effects of neuroprotection and cerebral blood flow improvement. AAPB significantly reduced cerebral infarction and neural function deficit in tMCAO rats, pMCAO rats, and IS rhesus monkeys, as well as displayed exceptional safety profiles and excellent pharmacokinetic properties in rats and dogs. AAPB has now entered phase I of clinical trials fighting IS in China.

Objective:To develop an artificial intelligence (AI)-based platelet review strategy to identify abnormal platelet histograms with no significant difference between initial impedance platelet count (PLT-I) and PLT-F results.Methods:This study included 5 119 routine blood analysis in Nanfang Hospital of Southern Medical University and its Ganzhou branch from July 2023 and March 2024. Specimens exhibiting abnormal platelet histograms and an initial platelet count >40×10?/L underwent review using the fluorescent platelet count (PLT-F) channel. Consistency of the results was defined as a difference between impedance platelet count (PLT-I) and PLT-F less than ±20% of the PLT-F results. A deep learning model was developed using platelet and red blood cell histogram data from a training set of 3 807 specimens. The model′s diagnostic performance was evaluated on an independent external validation set ( n=805) using receiver operating characteristic (ROC) curve analysis. Changes in the number of reviewed samples and sample turnaround time were analyzed to assess its clinical utility. Results:The deep learning model based on platelet and red blood cell histograms achieved an area under the ROC curve (AUC) of 0.854 in the training set. At a cutoff value of 0.1, the sensitivity was 0.954 and specificity was 0.358. The model could reduce review by 16.80% (190/1 131). In the validation set, the AUC was 0.805, with a sensitivity of 0.955 and specificity of 0.307, corresponding to a reduction of 17.41% (47/270) in reviewed specimens.Conclusion:The platelet review prediction model developed based on deep learning technology can efficiently identify samples with consistent results before and after review, reducing unnecessary reviews and shortening specimen testing time, thereby improving the efficiency of platelet test.