Abstract: Objective To explore the trend of maternal mortality in Hainan Province during the period of 2003-2022, both in the province and in urban and rural areas, and to forecast the maternal mortality rate for the period 2023-2025. Methods The 2003-2022 data collected from Hainan Province's three-tier network for maternal mortality surveillance was examined using the Chi-square test for trend (CMH) to analyze the province-wide and urban-rural maternal mortality trends. The time series model forecasting using exponential smoothing was used to predict the maternal mortality rate in Hainan Province from 2023 to 2025. Results　From 2003 to 2022, a total of 201 011 616 live births and 463 maternal deaths were reported in Hainan Province, with a maternal mortality rate of 23.03 per 100 000. Over 20 years, the maternal mortality rate in Hainan Province showed a downward trend, with an annual rate of decline of -4.13%. The rate decreased significantly during this period.. From 2003 to 2022, the maternal mortality rate in rural areas of Hainan Province was 25.74/100 000 (373/1 448 943), and it was significantly higher than that in urban areas, 16.04/100 000 (90/561 173). In the first 10 years, the gap between urban and rural areas progressively reduced, but it widened significantly in the last decade, especially after 2017. The maternal mortality rate was significantly lower in urban than rural areas, and the differences had statistical significance. The annual rates of decline in maternal mortality in Hainan Province and in urban and rural areas from 2003 to 2022 were -5.0% and -3.71%, respectively, showing a negative growth with the decrease rate in rural areas lower than urban areas. The maternal mortality rate in Hainan Province showed a fluctuating downward trend, different from the stable trend of national decrease. The mortality rates for direct obstetric causes of death (233 cases) and indirect obstetric causes of death (230 cases) were 11.59/100 000 and 11.44/100 000, respectively. The results of the maternal mortality review showed a predominance of avoidable deaths (315 cases, 68.03%). Brown's cubic exponential smoothing predicted the maternal mortality rate in Hainan Province for 2023-2025 as 9.45/100 000, 8.17/100 000, and 6.89/100 000. Conclusions　The maternal mortality rate in Hainan Province is largely influenced by maternal deaths in rural areas, and maternal health care in rural areas should be emphasized. Measures such as intervening to address the main factors influencing avoidable maternal deaths, strengthening high-risk maternal management, improving the level of critical maternal care, and providing subsidies for critical maternal care can sustainably reduce the maternal mortality rate in Hainan.

{L-End}Objective To analyze the characteristics of hearing loss and the influencing factors of high-frequency hearing loss (HFHL) among noise-exposed workers in an urban rail transit enterprise over five consecutive years. {L-End}Methods A total of 1 268 noise-exposed workers, who exposed to the average noise intensity of <85.0 dB(A), in an urban rail transit enterprise was selected as the research subjects using a judgment sampling method. The pure-tone audiometry results from 2019 to 2023 were collected to analyze the result of hearing loss. The influencing factors of HFHL (average hearing threshold ≥40.0 dB at high frequencies in both ears) were analyzed using the generalized estimating equations (GEE). {L-End}Results The detection rates of threshold elevations at frequencies of 0.5-6.0 kHz increased with increasing frequency from 2019 to 2023 (all P<0.01), with the highest detection rate at 6.0 kHz. The detection rate of speech frequency hearing loss (hearing threshold weighted value≥26.0 dB in the better ear) was 0.1%, 0.0%, 0.4%, 0.2%, and 0.2%, respectively. The detection rate of HFHL from 2019 to 2023 was 2.4%, 2.8%, 2.8%, 2.1%, and 2.8%, respectively. The GEE analysis results showed that the risk of HFHL of the workers in 2022 and 2023 was lower than that in 2019 (all P<0.01), with the odds ratios and 95% confidence intervals ［OR (95%CI)］ of 0.57 (0.41-0.81) and 0.65 (0.48-0.87), respectively. The risk of HFHL was higher among vehicle maintenance worker than train drivers (P<0.05), with OR (95%CI) of 2.37 (1.18-4.77). The risk of HFHL increased with age and length of service among the workers (all P<0.05), with the OR (95%CI) of 2.05 (1.22-3.46) and 1.69 (1.12-2.54), respectively. No interaction was found between type of job and age, type of job and length of service, or age and length of service in the risk of HFHL among the research subjects(all P<0.05). {L-End}Conclusion Noise exposure below the national occupational exposure limits can lead to hearing loss in noise-exposed workers of urban rail transit enterprises, possibly affecting the hearing threshold at 6.0 kHz first. The influencing factors for HFHL in workers of rail transit are age, length of service, and type of job. There is a dose-effect relationship with age and length of service.

Objective To investigate the learning curve of contrast-enhanced ultrasonography ( CEUS) in sentinel lymph node( SLN ) of breast cancer and provide a theoretical basis for leaners to learn SLN CEUS . Methods The multi-center study of SLN CEUS in breast cancer" was planned by Sichuan Cancer Hospital . According to the uniform inclusion and exclusion criteria , 511 patients with complete clinical data and follow-up results from 9 hospitals in Multi-center were included in this study . According to the inspection time ,the patients were divided into 3 groups named as group A ( 170 patients) ,group B ( 170 patients) and group C ( 171 patients ) ,respectively . The basic clinical data ,ultrasound imaging data , intraoperative and postoperative pathological findings of all patients were recorded . With the accumulation of cases examined ,analysis was performed to find the learning curve of the SLN CEUS examination time , SLN CEUS detection rate ,SLN CEUS surface marking accuracy rate and SLN CEUS diagnosis rate ,the learning curve was analyzed . Results ① There was no statistical significant difference in patients ages , tumors sizes ,tumors locations ,SLNs numbers and LCs numbers among the three groups( all P > 0 .05) . ②As the number of cases examined increases ,the examination time was reduced gradually ,but SLN detection rate ,surface marking accuracy and SLN diagnostic coincidence rate were increased gradually( F = 151 .75 , 1 .96 ,7 .49 ,5 .50 ; P = 0 .000 ,0 .143 ,0 .001 ,0 .005 ) . Conclusions The skill of the doctor is improved gradually when learning SLN CEUS . With the number of the cases increase ,the operating time of SLN CEUS is shorted ,and the SLN detection rate ,surface marking accuracy and SLN diagnostic coincidence rate of SLN-CEUS are gradually increased . It has an important clinical significance for beginners to learn the SLN CEUS technology .

Objective To explore the therapeutic mechanism of Modified Lugen Formula(Phragmitis Rhizoma,Cicadae Periostracum,Batryticatus Bombyx,Lonicerae Japonicae Flos,Glycyrrhiza,Menthae Haplocalycis Herba,Notopterygii Rhizoma et Radix,Puerariae Lobatae Radix,Bupleuri Radix)in treating influenza from the virus-host interaction interface.Methods The phytocompounds were first collected from the HERB database,and then potential active compounds were screened out by Lipinski's rules of five.The targets of active compounds were further predicted through the SwissTargetPrediction platform.Differentially expressed genes(DEGs)were determined from the human H1N1 influenza dataset GSE90732 available in the Gene Expression Omnibus database(GEO).H1N1-Homo sapiens-related protein-protein interactions(PPIs)were gathered from the Pathogen-Host Interaction Search Tool(PHISTO).The above mentioned bioinformatic datasets were integrated.Then a PPI network and a Formula-virus-host interaction network were constructed using Cytoscape.Functional enrichment analyses were performed by using R software.Finally,molecular docking was carried out to evaluate the binding activities between the key compounds and targets.Results A total of 1 252 active compounds,1 415 targets,951 influenza-related DEGs,and 10 142 H1N1-Homo sapiens-related PPIs were obtained.There were 72 intersection targets between the Modified Lugen Formula and influenza.Functional enrichment analyses showed that these targets are closely related to host defense and programmed cell death.The network topological analysis showed that active compounds in the Modified Lugen Formula,such as oleanolic acid,γ-undecalactone,and longispinogenin,regulate viral proteins M2,NA,NS1,and HA and/or the host factors HSP90AA1,NRAS,and ITGB1,thus exert therapeutic effect.Molecular docking results confirmed that these compounds had a good binding ability with the targets.Conclusion Multiple active ingredients in Modified Lugen Formula directly target influenza virus proteins and/or host factors,thereby play an anti-influenza role in multiple dimensions,including inhibiting virus replication,regulating host defense and cell death.This study provides a theoretical basis for further experimental analysis of the action mechanism of the Modified Lugen Formula in treating influenza.