Objective: To identify early clinical predictors associated with dysphagia and delayed swallowing recovery in patients with traumatic brain injury (TBI). Methods: In this retrospective study, we enrolled adult TBI patients admitted to the rehabilitation unit of a tertiary medical center between June 2019 and June 2023. Data on baseline characteristics, neurological status, imaging findings, and rehabilitation-related variables were collected. Swallowing function was assessed using two indicators: (1) nasogastric (NG) tube retention and (2) the Functional Oral Intake Scale (FOIS) scores at 1, 4, and 12 weeks post-injury. Regression analyses were conducted to identify predictors associated with dysphagia and swallowing recovery. Results: A total of 160 patients were included. At 1 week post-injury, longer intensive care unit (ICU) stay, poor initial sitting balance and use of sedative medication in ICU were associated with NG tube retention. At 4 weeks, lower initial Rancho Los Amigos Scale (RLAS) scores, immobility-related complications, longer hospitalization, and temporal lobe hematomas were associated with persistent NG tube dependence. By 12 weeks, older age, delayed ability to follow commands, and poor initial sitting balance remained associated with NG tube retention. FOIS outcomes were also associated with older age, delayed time to follow commands, impaired initial sitting balance, prolonged ICU stay, temporal lobe hematomas, lower initial RLAS scores, immobility-related complications, prolonged endotracheal tube placement and extended hospital stays. Conclusion Impaired cognitive status, poor physical function, immobility-related complications, and temporal lobe hematomas were key factors associated with dysphagia and delayed oral intake in individuals with TBI.

Schizophrenia (SZ) stands as a severe psychiatric disorder. This study applied diffusion tensor imaging (DTI) data in conjunction with graph neural networks to distinguish SZ patients from normal controls (NCs) and showcases the superior performance of a graph neural network integrating combined fractional anisotropy and fiber number brain network features, achieving an accuracy of 73.79% in distinguishing SZ patients from NCs. Beyond mere discrimination, our study delved deeper into the advantages of utilizing white matter brain network features for identifying SZ patients through interpretable model analysis and gene expression analysis. These analyses uncovered intricate interrelationships between brain imaging markers and genetic biomarkers, providing novel insights into the neuropathological basis of SZ. In summary, our findings underscore the potential of graph neural networks applied to multimodal DTI data for enhancing SZ detection through an integrated analysis of neuroimaging and genetic features.
Humans ; Schizophrenia/pathology* ; Diffusion Tensor Imaging/methods* ; Male ; Female ; Adult ; Brain/metabolism* ; Young Adult ; Middle Aged ; White Matter/pathology* ; Gene Expression ; Nerve Net/diagnostic imaging* ; Graph Neural Networks

Traditional Chinese medicine(TCM) decoction pieces are a core carrier for the inheritance and innovation of TCM, and their quality and safety are critical to public health and the sustainable development of the industry. Conventional quality control models, while having established a well-developed system through long-term practice, still face challenges such as relatively long inspection cycles, insufficient objectivity in characterizing complex traits, and urgent needs for improving the efficiency of integrating multidimensional quality information when confronted with the dual demands of large-scale production and precision quality control. With the rapid development of artificial intelligence, machine learning can deeply analyze multidimensional data of the morphology, spectroscopy, and chemical fingerprints of decoction pieces by constructing high-dimensional feature space analysis models, significantly improving the standardization level and decision-making efficiency of quality evaluation. This article reviews the research progress in the application of machine learning in the processing, production, and rapid quality evaluation of TCM decoction pieces. It further analyzes current challenges in technological implementation and proposes potential solutions, offering theoretical and technical references to advance the digital and intelligent transformation of the industry.
Machine Learning ; Drugs, Chinese Herbal/standards* ; Quality Control ; Medicine, Chinese Traditional/standards* ; Humans

Testicular histology based on testicular biopsy is an important factor for determining appropriate testicular sperm extraction surgery and predicting sperm retrieval outcomes in patients with azoospermia. Therefore, we developed a deep learning (DL) model to establish the associations between testicular grayscale ultrasound images and testicular histology. We retrospectively included two-dimensional testicular grayscale ultrasound from patients with azoospermia (353 men with 4357 images between July 2017 and December 2021 in The First Affiliated Hospital of Sun Yat-sen University, Guangzhou, China) to develop a DL model. We obtained testicular histology during conventional testicular sperm extraction. Our DL model was trained based on ultrasound images or fusion data (ultrasound images fused with the corresponding testicular volume) to distinguish spermatozoa presence in pathology (SPP) and spermatozoa absence in pathology (SAP) and to classify maturation arrest (MA) and Sertoli cell-only syndrome (SCOS) in patients with SAP. Areas under the receiver operating characteristic curve (AUCs), accuracy, sensitivity, and specificity were used to analyze model performance. DL based on images achieved an AUC of 0.922 (95% confidence interval [CI]: 0.908-0.935), a sensitivity of 80.9%, a specificity of 84.6%, and an accuracy of 83.5% in predicting SPP (including normal spermatogenesis and hypospermatogenesis) and SAP (including MA and SCOS). In the identification of SCOS and MA, DL on fusion data yielded better diagnostic performance with an AUC of 0.979 (95% CI: 0.969-0.989), a sensitivity of 89.7%, a specificity of 97.1%, and an accuracy of 92.1%. Our study provides a noninvasive method to predict testicular histology for patients with azoospermia, which would avoid unnecessary testicular biopsy.
Humans ; Male ; Azoospermia/diagnostic imaging* ; Deep Learning ; Testis/pathology* ; Retrospective Studies ; Adult ; Ultrasonography/methods* ; Sperm Retrieval ; Sertoli Cell-Only Syndrome/diagnostic imaging*

BACKGROUND: Blood glucose and serum albumin have been associated with cardiovascular disease prognosis, but the impact of admission-blood-glucose-to-albumin ratio (AAR) on adverse outcomes in critical ill coronary artery disease (CAD) patients was not investigated. METHODS: Patients diagnosed with CAD were non-consecutively selected from the MIMIC-IV database and categorized into quartiles based on their AAR. The primary outcome was 1-year mortality, and secondary endpoints were in-hospital mortality, acute kidney injury (AKI), and renal replacement therapy (RRT). A restricted cubic splines model and Cox proportional hazard models assessed the association between AAR and adverse outcomes in CAD patients. Kaplan-Meier survival analysis determined differences in endpoints across subgroups. RESULTS: A total of 8360 patients were included. There were 726 patients (8.7%) died in the hospital and 1944 patients (23%) died at 1 year. The incidence of AKI and RRT was 63% and 4.3%, respectively. High AAR was markedly associated with in-hospital mortality (HR = 1.587, P = 0.003), 1-year mortality (HR = 1.502, P < 0.001), AKI incidence (HR = 1.579, P < 0.001), and RRT (HR = 1.640, P < 0.016) in CAD patients in the completely adjusted Cox proportional hazard model. Kaplan-Meier survival analysis noted substantial differences in all endpoints based on AAR quartiles. Stratified analysis and interaction test demonstrated stable correlations between AAR and outcomes. CONCLUSIONS The results highlight that AAR may be a potential indicator for assessing in-hospital mortality, 1-year mortality, and adverse renal prognosis in critical CAD patients.

OBJECTIVE: To investigate the effect of zedoarondiol on neovascularization of atherosclerotic (AS) plaque by exosomes experiment. METHODS: ApoE^-/- mice were fed with high-fat diet to establish AS model and treated with high- and low-dose (10, 5 mg/kg daily) of zedoarondiol, respectively. After 14 weeks, the expressions of anti-angiogenic protein thrombospondin 1 (THBS-1) and its receptor CD36 in plaques, as well as platelet activation rate and exosome-derived miR-let-7a were detected. Then, zedoarondiol was used to intervene in platelets in vitro, and miR-let-7a was detected in platelet-derived exosomes (Pexo). Finally, human umbilical vein endothelial cells (HUVECs) were transfected with miR-let-7a mimics and treated with Pexo to observe the effect of miR-let-7a in Pexo on tube formation. RESULTS: Animal experiments showed that after treating with zedoarondiol, the neovascularization density in plaques of AS mice was significantly reduced, THBS-1 and CD36 increased, the platelet activation rate was markedly reduced, and the miR-let-7a level in Pexo was reduced (P<0.01). In vitro experiments, the platelet activation rate and miR-let-7a levels in Pexo were significantly reduced after zedoarondiol's intervention. Cell experiments showed that after Pexo's intervention, the tube length increased, and the transfection of miR-let-7a minics further increased the tube length of cells, while reducing the expressions of THBS-1 and CD36. CONCLUSION Zedoarondiol has the effect of inhibiting neovascularization within plaque in AS mice, and its mechanism may be potentially related to inhibiting platelet activation and reducing the Pexo-derived miRNA-let-7a level.
Animals ; MicroRNAs/genetics* ; Exosomes/drug effects* ; Plaque, Atherosclerotic/genetics* ; Neovascularization, Pathologic/genetics* ; Human Umbilical Vein Endothelial Cells/metabolism* ; Humans ; Blood Platelets/drug effects* ; Apolipoproteins E/deficiency* ; Thrombospondin 1/metabolism* ; CD36 Antigens/metabolism* ; Platelet Activation/drug effects* ; Male ; Mice ; Mice, Inbred C57BL

Peptide-based radiopharmaceuticals targeting integrin α5β1 show promise for precise tumor diagnosis and treatment. However, current peptide-based radioligands that target α5β1 demonstrate inadequate in vivo performance owing to limited tumor retention. The use of PEGylation to enhance the tumor retention of radiopharmaceuticals by prolonging blood circulation time poses a risk of increased blood toxicity. Therefore, a PEGylation strategy that boosts tumor retention while minimizing blood circulation time is urgently needed. Here, we developed a PEGylation-enabled peptide multidisplay platform (PEGibody) for PR_b, an α5β1 targeting peptide. PEGibody generation involved PEGylation and self-assembly. [⁶⁴Cu]QM-2303 PEGibodies displayed spherical nanoparticles ranging from 100 to 200 nm in diameter. Compared with non-PEGylated radioligands, [⁶⁴Cu]QM-2303 demonstrated enhanced tumor retention time due to increased binding affinity and stability. Importantly, the biodistribution analysis confirmed rapid clearance of [⁶⁴Cu]QM-2303 from the bloodstream. Administration of a single dose of [¹⁷⁷Lu]QM-2303 led to robust antitumor efficacy. Furthermore, [⁶⁴Cu]/[¹⁷⁷Lu]QM-2303 exhibited low hematological and organ toxicity in both healthy and tumor-bearing mice. Therefore, this study presents a PEGibody-based radiotheranostic approach that enhances tumor retention time and provides long-lasting antitumor effects without prolonging blood circulation lifetime. The PEGibody-based radiopharmaceutical [⁶⁴Cu]/[¹⁷⁷Lu]QM-2303 shows great potential for positron emission tomography imaging-guided targeted radionuclide therapy for α5β1-overexpressing tumors.

Objective:To construct an intelligent evaluation model based on the grey whitening weight cluster algorithm for risk in supply chain of medical consumables,and explore its application value in the management for medical consumables.Methods:The risk source of supply chain of medical consumables under the supply-processing-distribution(SPD)mode was analyzed,and the problems in the management for supply chain were evaluated by using grey whitening weight cluster,and an intelligent evaluation model for risk in supply chain of medical consumables was constructed to conduct control and management for risk in supply chain of medical consumables.A total of 10.61 million pieces of 400 types of medical consumables that were purchased and used by the National Center for Children's Health,China,Beijing Children's Hospital,Capital Medical University from 2022 to 2023 were selected.In them,the 5.15 million pieces of 200 types of medical consumables that were purchased and used during January and December 2022 were controlled and managed for risk through the management mode of assessment and prediction with experts.The 5.46 million pieces of 200 types of medical consumables that were purchased and used during January and December 2023 were controlled and managed for risk through used the intelligent evaluation model based on the gray whitening weight cluster algorithm under the SPD mode in supply chain of medical consumables for risk(prediction management mode with evaluation model).The incidences of risk,and the accuracy of data in supply chain between the two management modes were compared.A self-made satisfaction questionnaire was used to investigate the satisfaction rates of medical staffs,medical technicians,managers of department,and keepers of warehouse regarding to the supply of medical consumables.Results:The incidence rates of risks in suppliers,inventory,finance,technical support and information security of supply chain were respectively 1.8%,2.2%,0.4%,1.5%and 3.1%by adopting prediction management mode with evaluation model in 100,000 randomly inspected cases of medical consumables,all of which were lower than those of the management mode of assessment and prediction with experts,and the differences were statistically significant(x2=9.239,23.013,11.706,21.141,42.331,P<0.05).The accuracy rates of supply data of spot check for the consumables of auxiliary examination,the nursing consumables in ward,the consumables of surgical treatment,the consumables of oral treatment and other disposable consumables of prediction management mode with evaluation model were all higher than those of the management mode of assessment and prediction with experts,and the differences were statistically significant(x2=11.628,15.842,7.790,7.289,7.448,P<0.05).The satisfaction rates of medical staffs,medical technicians,managers of department and keepers of warehouse for clinical supply of medical consumables in prediction management mode with evaluation model were all higher than those in the management mode of assessment and prediction with experts,and the differences were statistically significant(x2=4.824,5.703,5.529,5.143,P<0.05).Conclusion:The intelligent evaluation model based on the grey whitening weight cluster algorithm under the SPD mode for risk in supply chain of medical consumables can reduce the incidence rate of risk in the supply chain of medical consumables,and improve the accuracy of supply data of medical consumables,and enhance the satisfaction of staffs in hospital.

In recent years, the deep integration of artificial intelligence (AI) into medical education has created new opportunities for teaching Biochemistry and Molecular Biology, while also offering innovative solutions to the pedagogical challenges associated with protein structure and function. Focusing on the case of anaplastic lymphoma kinase (ALK) gene mutations in non-small-cell lung cancer (NSCLC), this study integrates AI into case-based learning (CBL) to develop an AI-CBL hybrid teaching model. This model features an intelligent case-generation system that dynamically constructs ALK mutation scenarios using real-world clinical data, closely linking molecular biology concepts with clinical applications. It incorporates AI-powered protein structure prediction tools to accurately visualize the three-dimensional structures of both wild-type and mutant ALK proteins, dynamically simulating functional abnormalities resulting from conformational changes. Additionally, a virtual simulation platform replicates the ALK gene detection workflow, bridging theoretical knowledge with practical skills. As a result, a multidimensional teaching system is established—driven by clinical cases and integrating molecular structural analysis with experimental validation. Teaching outcomes indicate that the three-dimensional visualization, dynamic interactivity, and intelligent analytical capabilities provided by AI significantly enhance students’ understanding of molecular mechanisms, classroom engagement, and capacity for innovative research. This model establishes a coherent training pathway linking “fundamental theory-scientific research thinking-clinical practice”, offering an effective approach to addressing teaching challenges and advancing the intelligent transformation of medical education.

OBJECTIVE: This study aimed to identify high-risk areas for type 2 diabetes mellitus (T2DM) mortality to provide relevant evidence for interventions in emerging economies. METHODS: Empirical Bayesian Kriging and a discrete Poisson space-time scan statistic were applied to identify the spatiotemporal clusters of T2DM mortality. The relationships between economic factors, air pollutants, and the mortality risk of T2DM were assessed using regression analysis and the Poisson Log-linear Model. RESULTS: A coastal district in East Guangdong, China, had the highest risk (Relative Risk [RR] = 4.58, P < 0.01), followed by the 10 coastal districts/counties in West Guangdong, China (RR = 2.88, P < 0.01). The coastal county in the Pearl River Delta, China (RR = 2.24, P < 0.01), had the third-highest risk. The remaining risk areas were two coastal counties in East Guangdong, 16 districts/counties in the Pearl River Delta, and two counties in North Guangdong, China. Mortality due to T2DM was associated with gross domestic product per capita (GDP per capita). In pilot assessments, T2DM mortality was significantly associated with carbon monoxide. CONCLUSION High mortality from T2DM occurred in the coastal areas of East and West Guangdong, especially where the economy was progressing towards the upper middle-income level.
Diabetes Mellitus, Type 2/epidemiology* ; China/epidemiology* ; Humans ; Risk Factors ; Spatio-Temporal Analysis ; Air Pollutants/analysis* ; Socioeconomic Factors ; Bayes Theorem ; Female ; Male ; Middle Aged