Deep learning method can be used to automatically analyze electrocardiogram (ECG) data and rapidly implement arrhythmia classification, which provides significant clinical value for the early screening of arrhythmias. How to select arrhythmia features effectively under limited abnormal sample supervision is an urgent issue to address. This paper proposed an arrhythmia classification algorithm based on an adaptive multi-feature fusion network. The algorithm extracted RR interval features from ECG signals, employed one-dimensional convolutional neural network (1D-CNN) to extract time-domain deep features, employed Mel frequency cepstral coefficients (MFCC) and two-dimensional convolutional neural network (2D-CNN) to extract frequency-domain deep features. The features were fused using adaptive weighting strategy for arrhythmia classification. The paper used the arrhythmia database jointly developed by the Massachusetts Institute of Technology and Beth Israel Hospital (MIT-BIH) and evaluated the algorithm under the inter-patient paradigm. Experimental results demonstrated that the proposed algorithm achieved an average precision of 75.2%, an average recall of 70.1% and an average F ₁-score of 71.3%, demonstrating high classification accuracy and being able to provide algorithmic support for arrhythmia classification in wearable devices.
Humans ; Arrhythmias, Cardiac/diagnosis* ; Algorithms ; Electrocardiography/methods* ; Neural Networks, Computer ; Signal Processing, Computer-Assisted ; Deep Learning ; Classification Algorithms

Objective In the context of rapid development and widespread application of new generation information technologies represented by big data,artificial intelligence,the Internet of Things,and 5 G,a hospital has initiated the construc-tion and application of an integrated smart emergency medical information management system project for"Pre-Hospital Emergen-cy+Emergency Department+Specialized Disease Center."Methods T This paper primarily discusses the construction of the"Pre-Hospital Emergency+Emergency Department+Specialized Disease Center"integrated smart emergency medical information management system project,summarizing and exploring the implementation of a new efficient emergency treatment model charac-terized by"hospital admission upon boarding"and"no waiting for emergency care."Results A comprehensive integrated sys-tem covering pre-hospital emergency,triage,in-hospital treatment,and five major centers has been established.This system aims to optimize the treatment process,enhance treatment efficiency,and create a better medical experience for patients.Conclusion The paper discusses the optimization and improvement of the emergency medical information platform under the new circum-stances.

Objective:The objective of this study is to sequence whole genome of the rabies virus in the saliva specimen of a suspected rabies case with onset of rabies from a wildlife (hog badger) bite wound in Sichuan province, where the genetic variation characteristics of the virus was analyzed at the molecular level, so as to understand the prevalence and mutation of wildlife rabies virus in Sichuan province.Methods:Total viral RNA was extracted from the saliva specimen of the suspected rabies case. Then, rabies virus sequences were amplified using PCR with specific primers; the gene fragments obtained were sequenced, and the sequences obtained were spliced using biological software to obtain the whole genome sequence of the rabies virus strain. The genetic variation characteristics of the whole genome were analyzed.Results:The whole genome nucleotide sequence of a strain of rabies virus of hog badger origin (hereinafter referred to as SCR23-052) was obtained by sequencing, and NCBI online BLAST and comparison with several reference sequences showed that the composition and structure of the whole genome sequence of SCR23-052 conformed to the characteristics of the Lyssavirus under the Rhabdoviridae; the highest similarity in nucleotide and amino acid sequences in various gene regions was observed between SCR23-052 and the strains of Ningxia (J) and Chongqing (CQ92, 02050CHI). The sequence variability of SCR23-052 genome was significantly lower at the amino acid level than that at the nucleotide level, which indicated that most of the nucleotide variants in the protein-coding genes belonged to synonymous mutations. Phylogenetic analysis showed that SCR23-052 belonged to genotype V, which did not show any obvious mutation in the major antigenic site of the glycoprotein, underwent amino acid glycosylation at positions 56 and 338 by the online site prediction, and showed the least amino acid difference compared with the signal peptide sequence of the vaccine strain CTN181. The virus in this study has an A→T mutation at position 332 in the nucleoprotein major antigenic site with all reference vaccine strains, and an L→V mutation at position 379 in the B-cell epitope with the CTN181 vaccine strain. SCR23-052 was consistent with both genotype V reference strains at the nucleoprotein study site.Conclusions:The whole genome sequence of a wildlife strain of genotype V rabies virus of hog badger origin was obtained, which was different from that of the genotype I strain of rabies virus of dog origin that previously reported to be prevalent in Sichuan. The genome sequence of SCR23-052 differed from that of the reference vaccine strains to a varying degree, but the main virulence characteristics remained unaltered.

The purpose of this study is to construct a muscle-specific synthetic promoter library, screen out muscle-specific promoters with high activity, analyze the relationship between element composition and activity of highly active promoters, and provide a theoretical basis for artificial synthesis of promoters. In this study, 19 promoter fragments derived from muscle-specific elements, conserved elements, and viral regulatory sequences were selected and randomLy connected to construct a muscle-specific synthetic promoter library. The luciferase plasmids pCMV-Luc and pSPs-Luc were constructed and transfected into the myoblast cell line C2C12. The activities of the synthesized promoters were evaluated by the luciferase activity assay. Two non-muscle-derived cell lines HeLa and 3T3 were used to verify the muscle specificity of the highly active promoters. The sequences of promoters with high activity, good muscle specificity, and correct sequences were analyzed to explore the relationship between the element composition and activity of promoters. We successfully constructed a muscle-specific promoter library and screened out 321 effective synthetic promoter plasmids. Among them, the activity of SP-301 promoter was 5.63 times that of CMV. The 15 promoters with high activity were muscle-specific. In the promoters with high activity and correct sequences, there was a relationship between their element composition and activity. Muscle-specific elements accounted for a high proportion in the promoters, while they had weak correlations with the promoter activity, being tissue-specific determinants. Viral elements accounted for no less than 20% in highly active promoters, which may be the key elements for the promoter activity. The content of conserved elements was proportional to the promoter activity. This study lays a theoretical foundation for the synthesis of tissue-specific efficient promoters and provides a new idea for the construction and application of in-situ gene delivery systems.
Promoter Regions, Genetic ; Humans ; Animals ; Mice ; Gene Library ; Cell Line ; Transfection ; HeLa Cells ; Luciferases/metabolism* ; Muscle, Skeletal/metabolism* ; Plasmids/genetics* ; Myoblasts/cytology*

Background Steel workers are exposed to occupational hazardous factors such as dust, noise, and heat, and often work in shifts, making them prone to sleep disorders. Objective To explore potential influencing factors of sleep disorders among workers in a steel enterprise in Gansu Province, and provide a basis for reducing the risk of sleep disorders among them. Methods From January to March 2022, a self-made questionnaire combined with the Pittsburgh Sleep Quality Index (PSQI) were used to investigate the employees of a steel enterprise in Gansu Province. According to their PSQI scores, they were divided into a normal sleep group and a sleep disorder group. The general demographic variables of the two groups were balanced by 1∶1 propensity score matching (PSM). Multiple logistic regression was used to analyze the contributing factors of sleep disorders. Restricted cubic spline (RCS) model was used to analyze potential dose-response relationship between weekly working hours and sleep disorders. Results The prevalence of sleep disorders in the steel workers was 48.06% (6029/12544). After PSM, 5847 pairs were successfully matched, and the distributions of matched variables were well balanced between the two groups. The results of multiple logistic regression showed that hypertension (OR=1.39, 95%CI: 1.24, 1.56), diabetes mellitus (OR=1.34, 95%CI: 1.07, 1.66), three-shift system (OR=1.26, 95%CI: 1.12, 1.41), dust exposure (OR=1.14, 95%CI: 1.01, 1.29), noise exposure (OR=1.23, 95%CI: 1.09, 1.39), heat exposure (OR=1.16, 95%CI: 1.04, 1.29), and work injury (OR=1.22, 95%CI: 1.02, 1.46) increased the risk of sleep disorders. Compared with workers with < 10 years of service, those with 10-20 years (OR=1.31, 95%CI: 1.19, 1.44), 20-30 years (OR=1.34, 95%CI: 1.19, 1.52), and ≥30 years of service (OR=1.35, 95%CI: 1.19, 1.53) had a higher risk of sleep disorders. Compared with non-exercise workers, the risk of developing sleep disorders was lower in workers with occasional exercise (OR=0.61, 95%CI: 0.56, 0.66) and regular exercise (OR=0.55, 95%CI: 0.49, 0.62). The RCS model showed that the weekly working hours and sleep disorders in the steel workers showed a nonlinear dose-response relationship (P<0.05 for overall trend, P<0.05 for nonlinear test). The relationship between weekly working hours and sleep disorders showed a "U" shaped distribution, with a significant increase in the risk of sleep disorders when the weekly working hours exceeded 49 h. Conclusion The non-occupational influencing factors of sleep disorders of employees in the steel enterprise include hypertension, diabetes, physical exercise, and occupational influencing factors include length of service, weekly working hours, shifts, dust exposure, noise exposure, heat exposure, and work injuries. It is recommended to consider both occupational and non-occupational factors to formulate appropriate sleep disorder prevention and control measures for steel employees to reduce the risk of sleep disorders.

{L-End}Objective To analyze the correlation between occupational burnout and sleep quality among steelworkers. {L-End}Methods A total of 11 491 steelworkers from a large steel enterprise in Gansu Province were selected as the research subjects using convenient sampling method. The Burnout Questionnaire and the Pittsburgh Sleep Quality Index Scale were used to investigate their occupational burnout and sleep quality. Hierarchical regression analysis was used to analyze the effects of occupational burnout on the sleep quality. {L-End}Results The detection rate of occupational burnout and sleep disorder were 50.4% and 39.0%, respectively. There was a positive correlation between the level of occupational burnout and the total score of sleep quality (Spearman correlation coefficient=0.454, P<0.05). The results of hierarchical regression analysis, adjusting for confounding factors such as gender, age, marital status, education level, alcohol consumption, exercise, weekly working hours, seniority, work shift, noise exposure, dust exposure, and high-temperature work, showed that the score of occupational burnout was positively related to the score of sleep quality(P<0.01), explaining 16.0% of the variance in the score of sleep quality among these steelworkers. {L-End}Conclusion The detection rate of occupational burnout and sleep disorders are relatively high among the steelworkers in this enterprise. Higher levels of occupational burnout are associated with poorer sleep quality. Alleviating occupational burnout among steelworkers may contribute to improving their sleep quality.

Purpose: Oligometastatic non–small cell lung cancer (NSCLC) patients have been increasingly regarded as a distinct group that could benefit from local treatment to achieve a better clinical outcome. However, current definitions of oligometastasis are solely numerical, which are imprecise because of ignoring the biological heterogeneity caused by genomic characteristics. Our study aimed to profile the molecular alterations of oligometastatic NSCLC and elucidate its potential difference from polymetastasis. Materials and Methods: We performed next-generation sequencing to analyze tumors and paired peripheral blood from 77 oligometastatic and 21 polymetastatic NSCLC patients to reveal their genomic characteristics and assess the genetic heterogeneity. Results: We found ERBB2, ALK, MLL4, PIK3CB, and TOP2A were mutated at a significantly lower frequency in oligometastasis compared with polymetastasis. EGFR and KEAP1 alterations were mutually exclusive in oligometastatic group. More importantly, oligometastasis has a unique significant enrichment of apoptosis signaling pathway. In contrast to polymetastasis, a highly enriched COSMIC signature 4 and a special mutational process, COSMIC signature 14, were observed in the oligometastatic cohort. According to OncoKB database, 74.03% of oligometastatic NSCLC patients harbored at least one actionable alteration. The median tumor mutation burden of oligometastasis was 5.00 mutations/Mb, which was significantly associated with smoking, DNA damage repair genes, TP53 mutation, SMARCA4 mutation, LRP1B mutation, ABL1 mutation. Conclusion Our results shall help redefine oligometastasis beyond simple lesion enumeration that will ultimately improve the selection of patients with real oligometastatic state and optimize personalized cancer therapy for oligometastatic NSCLC.