Identifying drug-drug interactions (DDIs) is essential to prevent adverse effects from polypharmacy. Although deep learning has advanced DDI identification, the gap between powerful models and their lack of clinical application and evaluation has hindered clinical benefits. Here, we developed a Multi-Dimensional Feature Fusion model named MDFF, which integrates one-dimensional simplified molecular input line entry system sequence features, two-dimensional molecular graph features, and three-dimensional geometric features to enhance drug representations for predicting DDIs. MDFF was trained and validated on two DDI datasets, evaluated across three distinct scenarios, and compared with advanced DDI prediction models using accuracy, precision, recall, area under the curve, and F1 score metrics. MDFF achieved state-of-the-art performance across all metrics. Ablation experiments showed that integrating multi-dimensional drug features yielded the best results. More importantly, we obtained adverse drug reaction reports uploaded by Xiangya Hospital of Central South University from 2021 to 2023 and used MDFF to identify potential adverse DDIs. Among 12 real-world adverse drug reaction reports, the predictions of 9 reports were supported by relevant evidence. Additionally, MDFF demonstrated the ability to explain adverse DDI mechanisms, providing insights into the mechanisms behind one specific report and highlighting its potential to assist practitioners in improving medical practice.

Objective To investigate the factors influencing mood and sleep among special operation personnel,so as to provide evidence for the prevention and intervention of related issues.Methods A total of 600 special operation personnel during their special duty were enrolled by random sampling method.Basic information was collected by a general information questionnaire.Mood and sleep status were evaluated by self-rating anxiety scale(SAS),self-rating depression scale(SDS),and Pittsburgh sleep quality index scale(PSQI).Results Spearman correlation analysis showed that factors associated with anxiety were years of work experience(r=0.099,P=0.016),family residence(r=-0.153,P＜0.001),and sleep status(r=0.199,P＜0.001);the factors associated with depression were relationship with parents(r=-0.134,P=0.001),family residence(r=-0.144,P＜0.001),and sleep status(r=0.122,P=0.003);and the factors associated with sleep were age(r=0.088,P=0.031),years of work experience(r=0.079,P=0.049),and number of children(r=0.083,P=0.043).Binary logistic regression analysis showed that depression was not an influencing factor of sleep(odds ratio[OR]=1.302,95%confidence interval[95%CI]0.759-1.758,P=0.086),while anxiety was an influencing factor of sleep(OR=2.187,95%CI 1.608-4.695,P=0.001).Conclusion The sleep status of special operation personnel investigated in this survey is related to age and years of work experience,and is influenced by anxiety.Therefore,greater attention should be given to sleep issues in the occupational health support of special operations.Additionally,years of work experience,family residence,relationship with parents,and sleep status are related to anxiety and depression among special operation personnel.It is necessary to focus on the mental health of older workers and those with longer tenure.

Identifying drug-drug interactions(DDIs)is essential to prevent adverse effects from polypharmacy.Although deep learning has advanced DDI identification,the gap between powerful models and their lack of clinical application and evaluation has hindered clinical benefits.Here,we developed a Multi-Dimensional Feature Fusion model named MDFF,which integrates one-dimensional simplified molec-ular input line entry system sequence features,two-dimensional molecular graph features,and three-dimensional geometric features to enhance drug representations for predicting DDIs.MDFF was trained and validated on two DDI datasets,evaluated across three distinct scenarios,and compared with advanced DDI prediction models using accuracy,precision,recall,area under the curve,and F1 score metrics.MDFF achieved state-of-the-art performance across all metrics.Ablation experiments showed that integrating multi-dimensional drug features yielded the best results.More importantly,we obtained adverse drug reaction reports uploaded by Xiangya Hospital of Central South University from 2021 to 2023 and used MDFF to identify potential adverse DDIs.Among 12 real-world adverse drug reaction reports,the predictions of 9 reports were supported by relevant evidence.Additionally,MDFF demon-strated the ability to explain adverse DDI mechanisms,providing insights into the mechanisms behind one specific report and highlighting its potential to assist practitioners in improving medical practice.

【Objective】 To evaluate the clinical implications of ARL5B in esophageal cancer and its underlying mechanisms by using bioinformatics methods. 【Methods】 ARL5B transcriptomic expression data were obtained from The Cancer Genome Atlas (TCGA), R software was employed to detect the differential expression mRNAs, and related clinical information was collected for survival analysis. To validate the bioinformatics results, Real-time quantitative PCR (qRT-PCR) and Western blotting were carried out for clinical specimens of esophageal cancer tumor tissues and adjacent tissues. Immunohistochemistry was used to evaluate the expression of ARL5B and its associated clinicopathologic features. The underlying mechanisms of ARL5B in esophageal cancer were preliminarily explored by bioinformatics and qRT-PCR. 【Results】 Bioinformatics method showed that the expression of ARL5B in human esophageal cancer tissues was significantly higher than in adjacent tissues and correlated with poor prognosis. Clinical specimens were detected, the expressions of ARL5B mRNA and protein were the highest in metastases lymph node, followed by esophageal cancer tissues and adjacent tissues, which corresponded with bioinformatics results. The expression of ARL5B was strongly correlated with lymph node metastases and advanced clinical stage. Kaplan-Meier analysis results denoted high ARL5B level, indicating poor prognosis. Enrichment analysis showed that ARL5B was associated the biological processes such as vacuolar transport, late endosome to lysosome transport, and organelle localization. Protein-protein interaction analysis (PPI) suggested that ARL5B might interact with VPS16, KIF1A and TOM1, whose expressions were verified by qRT-PCR and positively correlated with ARL5B expression. 【Conclusion】 ARL5B was highly expressed in esophageal cancer and associated with lymph node metastases, advanced clinical stage, and poor prognosis. ARL5B may be involved in the progression of esophageal cancer with several molecular mechanisms.

Surgery-based multidisciplinary comprehensive treatment is the preferred treatment strategy for local advanced esophageal cancer. Neoadjuvant chemotherapy and neoadjuvant chemoradiotherapy have been recommended by the Chinese Society of Clinical Oncology (CSCO) guideline. With the advent of immunotherapy, neoadjuvant immunotherapy combined with chemotherapy has received much attention, and the first phase Ⅲ study has also confirmed that neoadjuvant immunotherapy combined chemotherapy is a promising treatment option. This article will review the recent advances and hot spots of neoadjuvant immunotherapy combined with chemotherapy.

ObjectiveTo investigate the effects of acupuncture combined with Du-Moxibustion (ADM) on peripheral blood cell count and levels of immune factors in patients with occupational chronic benzene poisoning. Methods A total of 70 patients with occupational chronic benzene poisoning (leukopenia and neutropenia) were selected as the research subjects by judgement sampling method. They were randomly divided into a control group and an ADM group using a random number table method, with 35 cases in each group. Patients in the control group were treated with conventional Western medicine such as leukocyte boosting and symptomatic treatment. While patients in the ADM group were treated with ADM treatment in addition to treatments of the control group, once per week for five consecutive weeks. Peripheral blood samples of patients were collected before and after treatment from both groups, to detect cell counts and serum levels of immune factors. Results The white blood cell count, red blood cell count, absolute lymphocyte count, absolute neutrophil count, platelet count, and levels of hemoglobin, immunoglobulins (Ig) A, IgM, IgG, complement C3 and complement C4 of patients in both groups improved after treatment compared with those before treatment (all P<0.05). The white blood cell count, levels of IgA, IgM, IgG, complement C3 and complement C4 of patients in the ADM group were higher than those in the control group after treatment (all P<0.05). Conclusion ADM treatment can increase peripheral blood white blood cells and serum levels of immune factor in patients with occupational chronic benzene poisoning (leukopenia, neutropenia), which helps improve patient recovery and can be promoted clinically.

This study aimed to explore the relationship between comorbidity factors and in-hospital mortality related to factors in patients with carbapenem-resistant Klebsiella pneumoniae (CRKP) pneumonia. This study collected clinical data from 218 patients with CRKP pneumonia in Beijing hospital from November 2011 to December 2023, analyzed the number of comorbidities carried by CRKP pneumonia patients, comorbidity patterns, Charlson Comorbidity Index (CCI) scores, and comorbidity of underlying diseases, and explored the relationship between various indicators and comorbidity factors and in-hospital mortality in CRKP pneumonia patients. The Ward.D cluster analysis was performed on the comorbidities of patients and used to draw heatmaps. Using a multiple logistic regression model, a nomogram model was constructed to predict in-hospital mortality in patients with CRKP pneumonia. This study included 218 patients with CRKP pneumonia. The results showed that there were significant differences in the age ( P=0.003), comorbidities such as heart failure ( P<0.001), arrhythmia ( P=0.002), chronic liver disease ( P=0.003), chronic kidney disease ( P=0.002), CCI score ( P=0.007), total number of comorbidities ( P<0.001), and comorbidity patterns (respiratory/immune/psychiatric disease patterns and cardiovascular/tumor/metabolic disease patterns, P=0.003) between the survival and death groups of CRKP pneumonia patients. The multiple logistic regression showed that cardiovascular/tumor/metabolic disease patterns ( P=0.030), CCI score ( P=0.040), concomitant heart failure ( P=0.011), and concomitant arrhythmia ( P=0.025) were independent risk factors for in-hospital mortality in patients with CRKP pneumonia. The nomogram model for predicting the risk of in-hospital mortality in patients with CRKP pneumonia, constructed based on the identified risk factors, had an area under the ROC curve of 0.758. Both the ROC curve and validation curve indicated that the nomogram model had stable performance in predicting in-hospital mortality in patients with CRKP pneumonia. In summary, comorbidity factors are risk factors for predicting in-hospital mortality in patients with CRKP pneumonia, and the role of comorbidity factors in in-hospital mortality in patients with CRKP pneumonia should be taken seriously.

This study aimed to explore the relationship between comorbidity factors and in-hospital mortality related to factors in patients with carbapenem-resistant Klebsiella pneumoniae (CRKP) pneumonia. This study collected clinical data from 218 patients with CRKP pneumonia in Beijing hospital from November 2011 to December 2023, analyzed the number of comorbidities carried by CRKP pneumonia patients, comorbidity patterns, Charlson Comorbidity Index (CCI) scores, and comorbidity of underlying diseases, and explored the relationship between various indicators and comorbidity factors and in-hospital mortality in CRKP pneumonia patients. The Ward.D cluster analysis was performed on the comorbidities of patients and used to draw heatmaps. Using a multiple logistic regression model, a nomogram model was constructed to predict in-hospital mortality in patients with CRKP pneumonia. This study included 218 patients with CRKP pneumonia. The results showed that there were significant differences in the age ( P=0.003), comorbidities such as heart failure ( P<0.001), arrhythmia ( P=0.002), chronic liver disease ( P=0.003), chronic kidney disease ( P=0.002), CCI score ( P=0.007), total number of comorbidities ( P<0.001), and comorbidity patterns (respiratory/immune/psychiatric disease patterns and cardiovascular/tumor/metabolic disease patterns, P=0.003) between the survival and death groups of CRKP pneumonia patients. The multiple logistic regression showed that cardiovascular/tumor/metabolic disease patterns ( P=0.030), CCI score ( P=0.040), concomitant heart failure ( P=0.011), and concomitant arrhythmia ( P=0.025) were independent risk factors for in-hospital mortality in patients with CRKP pneumonia. The nomogram model for predicting the risk of in-hospital mortality in patients with CRKP pneumonia, constructed based on the identified risk factors, had an area under the ROC curve of 0.758. Both the ROC curve and validation curve indicated that the nomogram model had stable performance in predicting in-hospital mortality in patients with CRKP pneumonia. In summary, comorbidity factors are risk factors for predicting in-hospital mortality in patients with CRKP pneumonia, and the role of comorbidity factors in in-hospital mortality in patients with CRKP pneumonia should be taken seriously.

Objective     To explore the causes of conversion to thoracotomy in patients with minimally invasive esophagectomy (MIE) in a surgical team, and to obtain a deeper understanding of the timing of conversion in MIE. Methods     The clinical data of patients who underwent MIE between September 9, 2011 and February 12, 2022 by a single surgical team in the Department of Thoracic Surgery of the Fourth Hospital of Hebei Medical University were retrospectively analyzed. The main influencing factors and perioperative mortality of patients who converted to thoracotomy in this group were analyzed. Results     In the cohort of 791 consecutive patients with MIE, there were 520 males and 271 females, including 29 patients of multiple esophageal cancer, 156 patients of upper thoracic cancer, 524 patients of middle thoracic cancer, and 82 patients of lower thoracic cancer. And 46 patients were converted to thoracotomy for different causes. The main causes for thoracotomy were advanced stage tumor (26 patients), anesthesia-related factors (5 patients), extensive thoracic adhesions (6 patients), and accidental injury of important structures (8 patients). There was a statistical difference in the distribution of tumor locations between patients who converted to thoracotomy and the MIE patients (P<0.05). The proportion of multiple and upper thoracic cancer in patients who converted to thoracotomy was higher than that in the MIE patients, while the proportion of lower thoracic cancer was lower than that in the MIE patients. The perioperative mortality of the thoracotomy patients was not significantly different from that of the MIE patients (P=1.000). Conclusion     In MIE, advanced-stage tumor, anesthesia-related factors,extensive thoracic adhesions, and accidental injury of important structures are the main causes of conversion to thoracotomy. The rate varies at different tumor locations. Intraoperative conversion to thoracotomy does not affect the perioperative mortality of MIE.

Objective To explore the application effect of quality control circle in the quality management improvement of the occupational disease diagnostic and code.Methods To established across departmental team of medical record QCC,we Select 132 cases of the occupational disease departmentin specialized Hospital from May 1,2022 to June31,2022,Analyze the reasons of inaccurate that main diagnosisselection and code mapping.After formulating a series of improvement measures,we Se-lect 71 cases of occupational disease department from November 1,2022 to December 31,2022.To compared effect that before and after the implementation of QCC.Results After6 months of improvement,The utilization rate of main diagnosis mapping Z-code was decreased from 36.3%to 12.7%,goal achievement rate of 106.3%,improvement rate of 65.0%.Enrollment rate of major diagnostic was increased from 53.0%to 86.3%,goal achievement rate of 116.4%,Improvement rate of 62.8%.Conclu-sion We have changed the quality control management mode of diagnosis and coding by the QCC,implement targeted feedback,supervision,and training,establish a reward and punishment mechanism that matches indicators to improved accuracy of enroll-ment.Upgrading the quality management level of medical records and medical safety at the same time.Should continue to pro-mote theimplementation.