To analyze the disease burden of acute lymphoid leukemia(ALL) and its changing trends in China and globally from 1990 to 2021, aiming to provide a theoretical basis for disease prevention, treatment, and policy formulation.

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 feasibility of using whole blood as a source of NK cells for allogeneic CAR NK cell therapy and activated NK cell reinfusion therapy, and initially construct a technical system for the separation and purification of NK cells from whole blood. [Methods] All peripheral blood mononuclear cells (PBMCs) were enriched from 400 mL of whole blood by manual separation and machine separation, respectively. The erythrocyte loss rate, PBMCs number, NK cell purity of the two methods were compared. NK cells were sorted from PBMCs by three separation and enrichment methods as immunomagnetic bead negative selection method, platelet lysate culture expansion and PERCOLL density gradient separation method, and the purity and yield of NK cells, the activity of NK cells and the tumor-killing ability of the three separation and enrichment methods were compared. [Results] The proportion of NK cells in the lymphocyte population was higher in the manual separation method than in the machine separation method[(13.16±5.16)% vs (8.56±3.92)%, P<0.05]; the number PBMCs was lower in the manual separation method than in the machine separation method[(4.09±1.80)×10⁸vs (6.49±2.16)×10⁸, P<0.05], and there was no difference in the red blood cell loss between the two methods (P>0.05). The purity of NK cells isolated and enriched from PBMCs by manual separation method using immunomagnetic was (96.77±2.31)%; the yield was (56.27±10.47)%; the inhibition of tumor proliferation was (38.67±14.05)%; and the tumor killing rate was (19.90±8.05)%. The purity of NK cells isolated and enriched from PBMCs by manual separation method using platelet lysis culture expansion method was the highest at day 7, which was (54.84±15.80)%; the cell expansion multiple could reach 16.92±6.28 at day 7; the in vitro tumor killing rate of NK cells was (15.83±5.5)%; the tumor inhibition rate was (44.33±13.5)%; and there was no difference in the toxicity and activity of NK cells between the two methods (P>0.05). The purity of NK cells isolated and enriched by PERCOLL density gradient separation method was (15.83±5.82)%, and the yield was (14±6.25)%, which was significantly lower than the other two methods. [Conclusion] PBMCs isolated from whole blood by manual separation and NK cells enriched by negative selection with immunomagnetic beads have the potential to provide NK cell materials for CAR-NK cell therapy, and NK cells enriched by platelet lysate-conditioned medium have the potential to provide NK cells for large-scale NK cell activation reinfusion therapy.

BACKGROUND: The contraction behaviors of cardiomyocytes (CMs), especially contraction synchrony, are crucial factors reflecting their maturity and response to drugs. A wider field of view helps to observe more pronounced synchrony differences, but the accompanied greater computational load, requiring more computing power or longer computational time. METHODS: We proposed a method that directly correlates variations in optical field brightness with cardiac tissue contraction status (CVB method), based on principles from physics and photometry, for rapid video analysis in wide field of view to obtain contraction parameters, such as period and contraction propagation direction and speed. RESULTS: Through video analysis of human induced pluripotent stem cell (hiPSC)-derived CMs labeled with green fluorescent protein (GFP) cultured on aligned and random nanofiber scaffolds, the CVB method was demonstrated to obtain contraction parameters and quantify the direction and speed of contraction within regions of interest (ROIs) in wide field of view. The CVB method required less computation time compared to one of the contour tracking methods, the LucasKanade (LK) optical flow method, and provided better stability and accuracy in the results. CONCLUSION This method has a smaller computational load, is less affected by motion blur and out-of-focus conditions, and provides a potential tool for accurate and rapid analysis of cardiac tissue contraction synchrony in wide field of view without the need for more powerful hardware.

ObjectiveTo explore the listed varieties and prescription characteristics of Chinese patent medicines for stroke in China， explore the medication rules of Chinese medicine for stroke， and provide guidance for further clinical research and development of Chinese patent medicines. MethodsExcel 2021 and the Ancient and Modern Medical Record Cloud Platform （V2.3.5） were used to systematically mine and analyze the varieties and prescriptions of Chinese patent medicines for stroke in China. ResultsA total of 244 Chinese patent medicines （two for different dosage forms of the same prescription）， 1 736 approval documents for Chinese patent medicines， 792 manufacturers， and 83 varieties of protected Chinese patent medicines were finally included in the database. The top three dosage forms were capsules （75）， pills （53）， and tablets （42）. There were 28 Chinese patent medicines for stroke in the National Essential Drug Catalogue （2018）， 129 in the National Essential Medical Insurance， Industrial Injury Insurance and Maternity Insurance Drug Catalogue （2023）， and 4 in the National Non-prescription Drug Catalogue. Among the 138 prescriptions screened out， Chinese patent medicines mainly treated stroke patients with the syndrome of Qi deficiency and blood stasis. The top three most frequent medicinal herbs were Chuanxiong Rhizoma （63）， Pheretima （47）， and Salviae Miltiorrhizae Radix et Rhizoma （47）. The medicinal herbs used were mainly warm， pungent， with the meridian tropism to the liver meridian. The correlation analysis showed that the herb pair with the highest support was Astragali Radix-Chuanxiong Rhizoma， and that with the highest confidence was Carthami Flos-Chuanxiong Rhizoma. Five herb combinations were identified based on the cluster analysis. ConclusionThe Chinese patent medicines for stroke mainly treat patients with the syndrome of Qi deficiency and blood stasis. The medicinal herbs used in the prescriptions mainly have the functions of activating blood and resolving stasis， extinguishing wind and stopping convulsions. Drug compatibility usually focuses on activating blood and resolving stasis， as well as expelling phlegm and opening orifices. This review of the varieties and prescription characteristics of Chinese patent medicines for stroke helps optimize clinical decision-making， guide drug research and development， promote medical research and scientific progress， and provide more effective support and guarantee for the treatment of stroke patients.

Objective To summarize the changing prevalence of carbapenem resistance in Enterobacterales based on the data of CHINET Antimicrobial Resistance Surveillance Program from 2015 to 2021 for improving antimicrobial treatment in clinical practice.Methods Antimicrobial susceptibility testing was performed using a commercial automated susceptibility testing system according to the unified CHINET protocol.The results were interpreted according to the breakpoints of the Clinical & Laboratory Standards Institute(CLSI)M100 31st ed in 2021.Results Over the seven-year period(2015-2021),the overall prevalence of carbapenem-resistant Enterobacterales(CRE)was 9.43％(62 342/661 235).The prevalence of CRE strains in Klebsiella pneumoniae,Citrobacter freundii,and Enterobacter cloacae was 22.38％,9.73％,and 8.47％,respectively.The prevalence of CRE strains in Escherichia coli was 1.99％.A few CRE strains were also identified in Salmonella and Shigella.The CRE strains were mainly isolated from respiratory specimens(44.23±2.80)％,followed by blood(20.88±3.40)％and urine(18.40±3.45)％.Intensive care units(ICUs)were the major source of the CRE strains(27.43±5.20)％.CRE strains were resistant to all the β-lactam antibiotics tested and most non-β-lactam antimicrobial agents.The CRE strains were relatively susceptible to tigecycline and polymyxins with low resistance rates.Conclusions The prevalence of CRE strains was increasing from 2015 to 2021.CRE strains were highly resistant to most of the antibacterial drugs used in clinical practice.Clinicians should prescribe antimicrobial agents rationally.Hospitals should strengthen antibiotic stewardship in key clinical settings such as ICUs,and take effective infection control measures to curb CRE outbreak and epidemic in hospitals.

Objective To characterize the changing species distribution and antibiotic resistance profiles of respiratory isolates in hospitals participating in the CHINET Antimicrobial Resistance Surveillance Program from 2015 to 2021.Methods Commercial automated antimicrobial susceptibility testing systems and disk diffusion method were used to test the susceptibility of respiratory bacterial isolates to antimicrobial agents following the standardized technical protocol established by the CHINET program.Results A total of 589 746 respiratory isolates were collected from 2015 to 2021.Overall,82.6％of the isolates were Gram-negative bacteria and 17.4％were Gram-positive bacteria.The bacterial isolates from outpatients and inpatients accounted for(6.0±0.9)％and(94.0±0.1)％,respectively.The top microorganisms were Klebsiella spp.,Acinetobacter spp.,Pseudomonas aeruginosa,Staphylococcus aureus,Haemophilus spp.,Stenotrophomonas maltophilia,Escherichia coli,and Streptococcus pneumoniae.Each microorganism was isolated from significantly more males than from females(P＜0.05).The overall prevalence of methicillin-resistant S.aureus(MRSA)was 39.9％.The prevalence of penicillin-resistant S.pneumoniae was 1.4％.The prevalence of extended-spectrum β-lactamase(ESBL)-producing E.coli and K.pneumoniae was 67.8％and 41.3％,respectively.The overall prevalence of carbapenem-resistant E.coli,K.pneumoniae,Enterobacter cloacae,Pseudomonas aeruginosa,and Acinetobacter baumannii was 3.7％,20.8％,9.4％,29.8％,and 73.3％,respectively.The prevalence of β-lactamase was 96.1％in Moraxella catarrhalis and 60.0％in Haemophilus influenzae.The H.influenzae isolates from children(＜18 years)showed significantly higher resistance rates to β-lactam antibiotics than the isolates from adults(P＜0.05).Conclusions Gram-negative bacteria are still predominant in respiratory isolates associated with serious antibiotic resistance.Antimicrobial resistance surveillance should be strengthened in clinical practice to support accurate etiological diagnosis and appropriate antimicrobial therapy based on antimicrobial susceptibility testing results.

Lung transplantation is an important means to treat end-stage lung disease and improve the survival rate and quality of life of patients. However, many postoperative complications seriously affect the prognosis of recipients. Accurate identification of key prognostic factors and construction of individualized and accurate prediction models are of great significance for postoperative prognosis evaluation, treatment strategy formulation and clinical decision-making. In recent years, the clinical prediction model of lung transplantation has gradually changed from traditional statistical methods to machine learning-driven. Compared with traditional models such as Cox regression and Logistic regression, machine learning models such as random forest, support vector machine and artificial neural network have certain advantages in postoperative survival rate prediction, early warning of complications and pulmonary function evaluation. However, their application is also affected by insufficient sample size and poor interpretability of models. Under the condition of small samples, the traditional model still has important value in prediction accuracy. The appropriate prediction model should be selected according to the clinical status of lung transplantation in China, considering the factors such as sample size, variable complexity and model interpretability. In the future, a multi-center, large-sample lung transplantation database should be constructed to further optimize and tap the potential of machine learning algorithms to improve the robustness and clinical applicability of the model.

Non-human primate animal models are core tools for the study of highly pathogenic microorganisms and are irreplaceable in the fields of pathology and drug discovery.However,anatomical sampling of non-human primate infection models in high-level biosafety laboratories carries potential risk and related risk assessment and control measures require clarification.Based on biosafety regulations and practical experience,we systematically discuss the risk control strategies of anatomical operations with respect to personal protection,instrument selection,anatomical specifications,documentation,and personnel training.Our review will help to improve the management of high-level biosafety laboratories,reduce the risk of pathogen escape and human infection,and provide support for the safe research of highly pathogenic microorganisms.