Objective: To analyze the epidemiological characteristics of category C intestinal infectious diseases among children and adolescents in Shenzhen from 2012 to 2024 and the association with meteorological factors, so as to provide a scientific basis for the targeted prevention and control of infectious diseases for children and adolescents. Methods: Using data from the "Infectious Disease Reporting Information Management System" of the "China Disease Prevention and Control Information System" covering the period from January 1, 2012 to December 31, 2024, the study analyzed clinical and confirmed cases of hand, foot, and mouth disease, other infectious diarrhea, and acute hemorrhagic conjunctivitis among individuals aged 6-19 years old to describe demographic and temporal characteristics. It used Joinpoint regression to calculate the average annual percent change (AAPC) and annual percent change (APC) to analyze incidence trends, and Spearman s correlation was combined to generalize linear models so as to assess the association between category C intestinal infectious diseases and meteorological factors. Results: From 2012 to 2024, a cumulative total of 61 019 cases of hand, foot, and mouth disease among children and adolescents, 58 498 cases of other infectious diarrhea, and 6 377 cases of acute hemorrhagic conjunctivitis were reported. The AAPC in the incidence rates of these three diseases was 19.19%, 31.03% and 31.48 %, respectively(all P <0.05). Notably, the incidence of hand, foot, and mouth disease increased significantly after 2022 (APC= 133.66 %, P <0.01). The temporal distribution showed that hand,foot,and mouth disease was most prevalent in May,June and July (seasonal index of 2.39,3.64,1.97), other infectious diarrhea was most prevalent in February,March and December (seasonal index of 1.22,1.25,1.47), and acute hemorrhagic conjunctivitis peaked in September and October (seasonal index of 4.22,2.16). Monthly average temperature could increase the risk of hand,foot,and mouth disease( β = 0.18 ,95% CI =0.11-0.25); as monthly average wind speed increased, the incidence of other infectious diarrhea ( β =-0.86, 95% CI = -1.50 to -0.22) and acute hemorrhagic conjunctivitis ( β =-1.32, 95% CI =-2.60 to -0.05) both decreased (all P < 0.05 ). Conclusions Among children and adolescents in Shenzhen, category C intestinal infectious diseases remain prevalent throughout the year；the number of reported hand, foot, and mouth disease cases has shown an upward trend in recent years.Temperature and wind speed significantly affect the number of reported cases of three types with category C intestinal infectious diseases.

Accurate prediction of molecular properties is crucial for selecting compounds with ideal properties and reducing the costs and risks of trials.Traditional methods based on manually crafted features and graph-based methods have shown promising results in molecular property prediction.However,traditional methods rely on expert knowledge and often fail to capture the complex structures and interactions within molecules.Similarly,graph-based methods typically overlook the chemical structure and function hidden in molecular motifs and struggle to effectively integrate global and local molecular information.To address these limitations,we propose a novel fingerprint-enhanced hierarchical graph neural network(FH-GNN)for molecular property prediction that simultaneously learns information from hierarchical molecular graphs and fingerprints.The FH-GNN captures diverse hierarchical chemical information by applying directed message-passing neural networks(D-MPNN)on a hierarchical molecular graph that integrates atomic-level,motif-level,and graph-level information along with their relationships.Addi-tionally,we used an adaptive attention mechanism to balance the importance of hierarchical graphs and fingerprint features,creating a comprehensive molecular embedding that integrated hierarchical mo-lecular structures with domain knowledge.Experiments on eight benchmark datasets from MoleculeNet showed that FH-GNN outperformed the baseline models in both classification and regression tasks for molecular property prediction,validating its capability to comprehensively capture molecular informa-tion.By integrating molecular structure and chemical knowledge,FH-GNN provides a powerful tool for the accurate prediction of molecular properties and aids in the discovery of potential drug candidates.

Objective To investigate the predictive value of serum interferon regulatory factor 4 (IRF4) and receptor activator of nuclear factor-κB ligand (RANKL) levels for recurrence and correlation with other immunological indicators in children with recurrent oral ulcers after treatment. Methods Eighty children with recurrent oral ulcers admitted to Beijing Chaoyang Hospital,Capital Medical University from January 2019 to June 2022 were collected as the diseased group. They were separated into a nonrecurrent group (n=61) and a recurrent group (n=19) based on follow-up results. Additionally,49 oral health volunteers who underwent physical examinations were regarded as the control group. ELISA was applied to detect serum levels of IRF4 and RANKL. General clinical data of patients were collected and analyzed. Multivariate logistic regression was applied to analyze the factors affecting the recurrence of children with recurrent oral ulcers. The Pearson method was applied to analyze the correlation between IRF4,RANKL,and immunological indicators. The receiver operating characteristic (ROC) curve was plotted to analyze the predictive value of serum IRF4 and RANKL levels for recurrence in children with recurrent oral ulcers after treatment. Results Compared with the control group,the levels of IRF4(9.67±1.03ng/ml,7.86±0.92ng/ml vs 6.19±0.71ng/ml) and RANK(192.95±19.86pg/ml,152.56±15.98pg/ml vs 83.96±9.85pg/ml) in the non recurrent group and the recurrent group,compared with the non recurrent group,the levels of IRF4 and RANKL in the recurrent group were increased,and the differences were statistically significant (F=121.514,487.250,all P<0.05). There were statistical difference between the two groups in terms of inflammatory factors such as IL-1β,TNF-α,IL-6 and CD4+,CD4+/CD8+(t=6.926～15.648,all P<0.05). Logistic regression analysis,IRF4[OR(95％CI):1.529(1.079～2.167)],RANKL[OR(95％CI):1.421(1.049～1.925)]and IL-1β[OR(95％CI):1.322(1.007～1.736)]were all risk factors for recurrence in children with recurrent oral ulcers ( all P<0.05),while CD4+[OR(95％CI):0.788(0.641～0.968)]was protective factor for recurrence in children with recurrentoral ulcers (P<0.05). Pearson correlation analysis,IRF4,RANKL were positively correlated with inflamnatory factors (r=0.453～0.533,all P<0.05). CD4+(r=-0.407,-0.409) and CD4+/CD8+(r=-0.425,-0.412) were positively correlated with inflammatory factors (P<0.05),ROC curve results revealed that the AUC(95％CI)of serum IRF4,RANK,and their combination in predicting recurrence in children with recurrent oral ulcers was 0.840(0.741～0.913),0.832(0.732～0.906) and 0.928(0.847～0.974) respectively. The combined prediction was better than that of IRF4 and RANKL alone prediction(Z=1.984,2.071,P=0.047,0.038). Conclusion The serum levels of IRF4 and RANKL in children with recurrent oral ulcers obviously increase,both of which are risk factors affecting recurrence in children after treatment,and have certain auxiliary predictive values for recurrence in children after treatment.

Objective:To investigate the knowledge, attitudes, practices (KAP), and demands of medical staff in family physician teams regarding the application of artificial intelligence (AI) in contracted services, and to analyze the influencing factors.Methods:A cross-sectional study was conducted from June to July 2023. A total of 602 medical staff members from family physician teams in Shanghai Minhang District were selected as subjects. Data on demographics (age, gender, institution, position, education, work experience, household registration, professional title, marital status, fertility status) and KAP/demand regarding AI application in contracted services were collected using a self-designed questionnaire. Intergroup differences were analyzed. Multiple stepwise linear regression was employed to identify the main factors influencing AI application demand.Results:Among the 602 participants, 484 (80.4%) were aged 30-49 years, 466 (77.40%) were females, 559 (92.9%) held a bachelor′s degree or higher, and 505 (83.9%) had intermediate or senior professional titles. The awareness rate for knowledge, positive attitude rate, and practice implementation rate regarding AI application were 47.2% (284/602), 73.1% (440/602), and 32.1% (193/602), respectively. The mean scores for knowledge, attitude, and practice were 15.72±3.40, 18.34±3.41, and 14.60±3.89, respectively. Significant differences were found among the items within each KAP dimension (knowledge: F=7.688, P<0.001; attitude: F=5.106, P<0.001; practice: F=6.763, P<0.001). Within knowledge, item K3 (awareness of intelligent elderly monitoring devices) scored lowest (3.00±0.79), differing significantly from K1, K2, K4, and K5 (all P<0.05). Within attitude, item A5 (willingness to fully trust AI′s accuracy and convenience in contracted services) scored lowest (3.57±0.75), differing significantly from A3 and A4 (all P<0.05). Within practice, item P3 (increasing reliance on AI in daily contracted services) scored lowest (2.79±0.93), differing significantly from P1 and P2 (all P<0.05). KAP scores differed significantly across demographic subgroups. Knowledge scores differed significantly by age, gender, and marital status (all P<0.05). Attitude scores differed significantly by gender, household registration, and fertility status (all P<0.05). Practice scores differed significantly by gender, position, and marital status (all P<0.05). Regarding demand, the most frequently selected areas were follow-up services (28.74%, 173/602), data management (26.25%, 158/602), and data collection (25.42%, 153/602). Univariate analysis identified age, gender, education, professional title, fertility status, and KAP scores as significant factors influencing AI application demand (all P<0.05). Multiple stepwise linear regression revealed that older age ( t=3.905, P<0.001), female gender ( t=3.548, P<0.001), and higher practice scores ( t=-3.044, P=0.002) were significant predictors of greater AI application demand. Conclusions:Significant variations exist in the KAP levels regarding AI application among family physician team members. Gender, age, and practice behavior significantly influence demand. Tailored strategies for different subgroups, coupled with timely targeted training and practical exercises, are recommended to enhance the effective and widespread adoption of AI technology in family physician contracted services.

Bone marrow edema (BME) of the temporomandibular joint is an abnormal change within the bone marrow that appears as low signal intensity on T1-weighted images and high signal intensity on T2-weighted images in MRI. It is commonly seen in patients with temporomandibular joint osteoarthritis. Studies have shown that BME is closely related to joint pain, disc displacement and bony degeneration, but its specific mechanism is not yet fully understood, potentially involving abnormal mechanical loading, inflammatory responses and imbalances in bone metabolism. Although BME is considered an early sign of osteoarthritis, BME persists in approximately 71.4% of patients after pain resolution , suggesting the complexity of its association with clinical symptoms. In recent years, advanced MRI techniques (such as T2 mapping and fat suppression sequences) have improved the detection rate of BME and facilitated the assessment of its dynamic changes. However, the exact role of BME in temporomandibular disorders remains controversial. This article reviews the etiology and mechanism, pathology, imaging characteristics, and correlation with clinical symptoms of BME in the temporomandibular joint, and proposing future research directions to to elucidate its significance in disease progression and to provide guidance for clinical diagnosis and treatment.

Bone marrow edema (BME) of the temporomandibular joint is an abnormal change within the bone marrow that appears as low signal intensity on T1-weighted images and high signal intensity on T2-weighted images in MRI. It is commonly seen in patients with temporomandibular joint osteoarthritis. Studies have shown that BME is closely related to joint pain, disc displacement and bony degeneration, but its specific mechanism is not yet fully understood, potentially involving abnormal mechanical loading, inflammatory responses and imbalances in bone metabolism. Although BME is considered an early sign of osteoarthritis, BME persists in approximately 71.4% of patients after pain resolution , suggesting the complexity of its association with clinical symptoms. In recent years, advanced MRI techniques (such as T2 mapping and fat suppression sequences) have improved the detection rate of BME and facilitated the assessment of its dynamic changes. However, the exact role of BME in temporomandibular disorders remains controversial. This article reviews the etiology and mechanism, pathology, imaging characteristics, and correlation with clinical symptoms of BME in the temporomandibular joint, and proposing future research directions to to elucidate its significance in disease progression and to provide guidance for clinical diagnosis and treatment.

Objective:To investigate the clinical effect of free anterolateral thigh myocutaneous flap combined with oral repair membrane in the reconstruction of nasal mucosa defect after maxillary malignant tumor surgery. Methods:A total of 12 patients with maxillary gingival squamous cell carcinoma and maxillary sinus cancer who had been treated in Department of Oral and Maxillofacial Surgery, Beijing Anzhen Nanchong Hospital, Capital Medical University & Nanchong Central Hospital, were selected from November 2020 to November 2023. Free anterolateral thigh musculocutaneous flap transplantation combined with oral repair membrane were used in all patients. Meanwhile, maxillary soft and hard tissue defects and nasal mucosa defects left after tumor operation were repaired and reconstructed. The clinical effect was evaluated after 6-12 months follow-up. Results:Subtotal maxillary resection was performed in 1 case, total maxillary resection in 9 cases and extended maxillary resection in 2 cases. The musculocutaneous flaps of all patients survived, the facial appearance was basically symmetrical, no obvious depression deformity, the swallowing and speech function recovered well, the mouth and nasal cavity were closed completely, the food could be eaten through the mouth, and the lower nasal passage was not blocked. Conclusion:The free anterolateral thigh musculoflap combined with oral repair membrane can be used to repair and reconstruct maxillary malignant tumor complicated with extensive maxillary tissue and nasal mucosa defect after operation, and the appearance and function can be recovered well after operation, which is a choice for maxillary malignant tumor complicated with nasal mucosa defect.
Humans ; Myocutaneous Flap ; Plastic Surgery Procedures/methods* ; Maxillary Neoplasms/surgery* ; Carcinoma, Squamous Cell/surgery* ; Male ; Middle Aged ; Female ; Nasal Mucosa/surgery* ; Maxilla/surgery* ; Thigh/surgery* ; Maxillary Sinus Neoplasms/surgery*

Accurate prediction of molecular properties is crucial for selecting compounds with ideal properties and reducing the costs and risks of trials. Traditional methods based on manually crafted features and graph-based methods have shown promising results in molecular property prediction. However, traditional methods rely on expert knowledge and often fail to capture the complex structures and interactions within molecules. Similarly, graph-based methods typically overlook the chemical structure and function hidden in molecular motifs and struggle to effectively integrate global and local molecular information. To address these limitations, we propose a novel fingerprint-enhanced hierarchical graph neural network (FH-GNN) for molecular property prediction that simultaneously learns information from hierarchical molecular graphs and fingerprints. The FH-GNN captures diverse hierarchical chemical information by applying directed message-passing neural networks (D-MPNN) on a hierarchical molecular graph that integrates atomic-level, motif-level, and graph-level information along with their relationships. Additionally, we used an adaptive attention mechanism to balance the importance of hierarchical graphs and fingerprint features, creating a comprehensive molecular embedding that integrated hierarchical molecular structures with domain knowledge. Experiments on eight benchmark datasets from MoleculeNet showed that FH-GNN outperformed the baseline models in both classification and regression tasks for molecular property prediction, validating its capability to comprehensively capture molecular information. By integrating molecular structure and chemical knowledge, FH-GNN provides a powerful tool for the accurate prediction of molecular properties and aids in the discovery of potential drug candidates.

Objective To construct large medical model named by "Huaxi HongYi"and explore its application effectiveness in assisting medical record generation. Methods By the way of a full-chain medical large model construction paradigm of "data annotation - model training - scenario incubation", through strategies such as multimodal data fusion, domain adaptation training, and localization of hardware adaptation, "Huaxi HongYi" with 72 billion parameters was constructed. Combined with technologies such as speech recognition, knowledge graphs, and reinforcement learning, an application system for assisting in the generation of medical records was developed. Results Taking the assisted generation of discharge records as an example, in the pilot department, after using the application system, the average completion times of writing a medical records shortened (21 min vs. 5 min) with efficiency increased by 3.2 time, the accuracy rate of the model output reached 92.4%. Conclusion It is feasible for medical institutions to build independently controllable medical large models and incubate various applications based on these models, providing a reference pathway for artificial intelligence development in similar institutions.

OBJECTIVE To explore comprehensive management and potential issues associated with long-term prescriptions medications of psychiatry， in order to provide a reference for the comprehensive management of long-term prescriptions of psychiatry in psychiatric hospitals and other medical institutions’ pharmacies. METHODS Starting from the applicable principles for long-term prescriptions of psychiatry， this study introduced the standardized assessment and precautions before issuing long-term prescriptions， the formulation and adjustment of the drug list， as well as the rational management of the long-term prescriptions. It also analyzed potential issues that may arise in the comprehensive management of long-term prescription medications and proposed corresponding countermeasures and suggestions. RESULTS & CONCLUSIONS Prior to initiating long-term prescriptions， a standardized assessment should be conducted on patients from the aspects of their psychiatric condition and long-term potential risk factors， pharmacological treatment plans and other non-pharmacological therapies， physical illnesses. Additionally， healthcare providers should fulfill their obligation to inform patients or their family members. The comprehensive management of long-term prescription medications should be jointly established and improved by multiple departments， and the formulation of drug catalogs should avoid including drugs with potential social harm or medication risks while complying with policy requirements. Furthermore， measures such as adding special identifiers to long-term prescriptions， providing patients with reminders about （No.YGLX202537） prescription expiration， or offering online consultations can also effectively enhance the rationality of medication use under long-term prescriptions. Currently， the implementation of long-term prescriptions in psychiatry remains challenged by inconsistencies in prescription duration， incomplete coverage of diagnostic categories， poor patient adherence， and the risk of deviation in clinical assessments. In this regard， measures such as collaborating with multiple departments to strengthen long-term prescription information management， providing matching pharmaceutical services， ensuring the quality and rationality of long-term prescription implementation， and using modern methods to screen high-risk patients can be taken to improve patient medication compliance and safety.