OBJECTIVE:Nowadays,machine learning algorithms are gradually being applied to predict stroke and cardiovascular disease.Compared with traditional regression models,machine learning can learn from data to achieve high prediction accuracy by exploring the flexible relationship between a large number of predictive features and outcome variables,providing a new method for the formulation of individualized treatment and rehabilitation programs.This study aims to systematically evaluate stroke functional recovery and prognosis prediction models based on machine learning,comprehensively assessing their predictive performance and clinical application potential to provide references for the development,application,and promotion of related predictive models.METHODS:This review was conducted following the PRISMA(Preferred Reporting Items for Systematic Reviews and Meta-Analyses)guidelines.Relevant literature on stroke prognosis prediction using machine learning methods was selected by searching PubMed,EMbase,Web of Science Core Collection,CNKI,WanFang,and the China Biomedical Literature Database,with the search period from January 1,2014,to July 1,2024.Two researchers independently screened the literature and extracted data based on inclusion and exclusion criteria,using the Prediction model Risk Of Bias ASsessment Tool(PROBAST)to assess model quality.RESULTS:(1)A total of 3 126 articles were obtained in the preliminary search.After screening and exclusion,18 articles were finally included.150 prediction models were constructed using 13 machine learning methods.The three most frequently used methods are Logistic Regression,Random Forest,and Extreme Gradient Boosting(XGBoost).Only one study was externally validated.Eight studies reported how the missing data were handled.(2)In terms of outcome indicators,8 studies used the combination of clinical data and imaging data to build models,9 studies only used clinical data to build models,and 1 study only used imaging data to build models.(3)Each of the 18 studies gave the most important characteristics of the study,with the most mentioned being the National Institute of Health Stroke Scale and age.All studies reported area under curve values ranging from 0.74 to 0.96,with the highest area under curve being 0.96.The overall risk of bias in all models was high.The high risk of bias in the field of model analysis was the main reason for the high risk of overall bias in all models.(4)The results of meta-analysis showed that age and National Institute of Health Stroke Scale score had significant influence on stroke prognosis,with age[MD=8.49,95％CI(6.24,10.75),P＜0.01]and National Institute of Health Stroke Scale score[MD=4.78,95％CI(2.56,7.00),P＜0.01].CONCLUSION:This study systematically evaluated the predictive model of functional recovery and prognosis of stroke based on machine learning,and all the models have good predictive potential.However,future studies should increase the sample size of the included model,adopt prospective studies,and add external validation of the model to improve the stability and prediction accuracy of the model,control the risk of bias,and contribute to the validation and promotion of the model in practical clinical applications.At the same time,the interpolation of missing values is more transparent and accurate.Although existing machine learning models show good predictive performance,it is also important to focus on the functionality and usability of the model,and the inclusion of features will reduce ease of use.We should develop easy to use model interfaces and user-friendly clinical tools to enable medical staff to better apply the model for clinical decision.

Objective:To analyze the effects of shift work on the renal function of oil workers and investigate whether there is a measured response relationship between shift work and renal dysfunction.Methods:In this study, oil workers who participated in physical examinations at the North China Oilfield Downhole Hospital were selected as the study subjects, and the physical examinations as well as questionnaires of the study subjects in 2017 and 2018 were collected as the baseline data, which included blood biochemical indexes, socio-demographic characteristics, history of life behaviors, occupational exposures, and occupational histories. Three follow-up surveys were subsequently conducted in April 2019, April 2020, and January 2021. The presence of renal dysfunction in the study population was determined based on the glomerular filtration rate tested at the medical examination hospital. The exposure of the study subjects to shift work was assessed using the weighted shift index (WSI), the relationship between different levels of shift work and renal dysfunction was analyzed using Cox regression, and the measure of WSI and renal dysfunction was explored by restricted cubic spline (RCS). response relationship.Results:A total of 2292 study participants were included in this study, and the prevalence density of renal dysfunction was 87.44 k/year, of which the prevalence of renal dysfunction in females (30.31%), those with per capita monthly income <2, 000 yuan (27.00%), those who consume alcohol (27.10%), those who are hypertensive (23.05%), those who are exposed to high temperatures (27.37%), those who are exposed to organic solvents (30.42%), and those who are engaged in shift work (25.87%) were to be found had a higher prevalence of renal dysfunction ( P<0.05). After correcting for age, sex, and other risk factors, there was a nonlinear association between intensity of shift work and renal dysfunction, with a hazard ratio ( HR) of 1.29 (95% CI: 0.98-1.59) for the development of renal dysfunction in petroleum workers for shift work performed at higher intensities, and moderate intensity of exposure to shift work reduced the risk of renal dysfunction in petroleum workers ( HR=0.54 with a 95% CI: 0.39-0.75, P<0.001) . Conclusion:Prolonged shift work increases the risk of renal dysfunction in oil workers, and the occurrence of renal dysfunction in oil workers is influenced by multiple factors.

To understand the pathogenicity and drug resistance of swine-derived E.coli and its bio-logical characteristics in some areas in Hebei,E.coli was isolated and identified from diarrheal fe-ces of piglets collected from swine farms,and the isolated strains were subjected to drug sensitivity test,detection of the ability to form biofilm,pathogenicity test,virulence gene test,drug resistance gene test,and identification of phylogenetic subgroups.The results showed that a total of 35 patho-genic E.coli strains were isolated from the feces of diarrheic piglets,and most of the isolates were multidrug-resistant,and were resistant to at least three antibiotics,including amoxicillin(88.57％),ampicillin(88.57％),doxycycline(88.75％),sulfisoxazole(77.17％),lincomycin(100％),and chloramphenicol(100％);the isolates were severely resistant.The isolates all carried virulence genes,with a total of five virulence genes detected,namely,EAST1(77.14％),eaeA(17.14％),stx2e(5.71％),LT(2.86％),and STb(2.86％),and the isolates also carried multi-re-sistance genes,with a total of five virulence genes detected,namely,bla TEM-1(65.71％),bla CTX-M(20.00％),tetA(82.86％),tetB(14.29％),aadA2(17.14％),aac(6')-Ib(14.29％),qnrS(17.14％),sul 1(40.00％),sul2(34.29％),and floR(60.00％);the phylogenetic grouping showed that the isolates had a high proportion of group B1 and group A;and all 35 isolates showed differ-ent pathogenicity after infection of mice.This study provides a reference for the selection of effec-tive therapeutic drugs and the development of prevention and control programs for swine-origin pathogenic E.coli in Hebei Province.

Objective:To establish a quantitative analysis of multi-components by single marker(QAMS)for the determination of 6 alkaloid components,which is benzoylmesaconine,benzoyl-hypaconine,benzoylaconine,mesaconitine,hypaconitine,and aconitine in Zhachong Shisanwei Pills,and prove the scientificity and feasibility of the method in the quality analysis.Methods:The chromatographic separation was performed on an Agilent Eclipse Plus C18(250 mm×4.6 mm,5 μm)with gradient elution using 0.1 mol·L-1 ammonium acetate(0.5 mL of gla-cial acetic acid per 1 000 mL)(A)-acetonitrile:tetrahydrofuran(25∶15)(B),as the mobile phase(0-50 min,18%B-28%B),the detection wavelength was switched from 235 nm,the column temperature was kept at 40℃and the flow rate was 1.0 mL·min-1.The relative correction factors(fs/i)were established with the other 5 compo-nents to be measured using benzoylaconine as the internal reference,which were used to calculate the mass fraction of each component.At the same time,the mass fractions of the 6 effective constituents in Zhachong Shisanwei Pills were calculated by the external standard method(ESM).By comparing the content results of ESMand QAMS,the accura-cy of QAMS method were evaluated.Results:The relative correction factors(fs/i)of benzoylmesaconine,benzoylhyp-aconine,mesaconitine,hypaconitine,and aconitine in Mongolian medicine Zhachong Shisanwei Pills were reproduci-ble with good reproducibility,which were 0.680 4,0.450 6,0.850 8,0.676 1 and 0.757 0,the result obtained by QAMS approximated those obtained by external standard method(ESM).Conclusion:The method is simple,stable and reproducible,and can be used for the quality control of 6 alkaloid components in Zhachong Shisanwei Pills.

OBJECTIVE:Nowadays,machine learning algorithms are gradually being applied to predict stroke and cardiovascular disease.Compared with traditional regression models,machine learning can learn from data to achieve high prediction accuracy by exploring the flexible relationship between a large number of predictive features and outcome variables,providing a new method for the formulation of individualized treatment and rehabilitation programs.This study aims to systematically evaluate stroke functional recovery and prognosis prediction models based on machine learning,comprehensively assessing their predictive performance and clinical application potential to provide references for the development,application,and promotion of related predictive models.METHODS:This review was conducted following the PRISMA(Preferred Reporting Items for Systematic Reviews and Meta-Analyses)guidelines.Relevant literature on stroke prognosis prediction using machine learning methods was selected by searching PubMed,EMbase,Web of Science Core Collection,CNKI,WanFang,and the China Biomedical Literature Database,with the search period from January 1,2014,to July 1,2024.Two researchers independently screened the literature and extracted data based on inclusion and exclusion criteria,using the Prediction model Risk Of Bias ASsessment Tool(PROBAST)to assess model quality.RESULTS:(1)A total of 3 126 articles were obtained in the preliminary search.After screening and exclusion,18 articles were finally included.150 prediction models were constructed using 13 machine learning methods.The three most frequently used methods are Logistic Regression,Random Forest,and Extreme Gradient Boosting(XGBoost).Only one study was externally validated.Eight studies reported how the missing data were handled.(2)In terms of outcome indicators,8 studies used the combination of clinical data and imaging data to build models,9 studies only used clinical data to build models,and 1 study only used imaging data to build models.(3)Each of the 18 studies gave the most important characteristics of the study,with the most mentioned being the National Institute of Health Stroke Scale and age.All studies reported area under curve values ranging from 0.74 to 0.96,with the highest area under curve being 0.96.The overall risk of bias in all models was high.The high risk of bias in the field of model analysis was the main reason for the high risk of overall bias in all models.(4)The results of meta-analysis showed that age and National Institute of Health Stroke Scale score had significant influence on stroke prognosis,with age[MD=8.49,95％CI(6.24,10.75),P＜0.01]and National Institute of Health Stroke Scale score[MD=4.78,95％CI(2.56,7.00),P＜0.01].CONCLUSION:This study systematically evaluated the predictive model of functional recovery and prognosis of stroke based on machine learning,and all the models have good predictive potential.However,future studies should increase the sample size of the included model,adopt prospective studies,and add external validation of the model to improve the stability and prediction accuracy of the model,control the risk of bias,and contribute to the validation and promotion of the model in practical clinical applications.At the same time,the interpolation of missing values is more transparent and accurate.Although existing machine learning models show good predictive performance,it is also important to focus on the functionality and usability of the model,and the inclusion of features will reduce ease of use.We should develop easy to use model interfaces and user-friendly clinical tools to enable medical staff to better apply the model for clinical decision.

Objective:To analyze the effects of shift work on the renal function of oil workers and investigate whether there is a measured response relationship between shift work and renal dysfunction.Methods:In this study, oil workers who participated in physical examinations at the North China Oilfield Downhole Hospital were selected as the study subjects, and the physical examinations as well as questionnaires of the study subjects in 2017 and 2018 were collected as the baseline data, which included blood biochemical indexes, socio-demographic characteristics, history of life behaviors, occupational exposures, and occupational histories. Three follow-up surveys were subsequently conducted in April 2019, April 2020, and January 2021. The presence of renal dysfunction in the study population was determined based on the glomerular filtration rate tested at the medical examination hospital. The exposure of the study subjects to shift work was assessed using the weighted shift index (WSI), the relationship between different levels of shift work and renal dysfunction was analyzed using Cox regression, and the measure of WSI and renal dysfunction was explored by restricted cubic spline (RCS). response relationship.Results:A total of 2292 study participants were included in this study, and the prevalence density of renal dysfunction was 87.44 k/year, of which the prevalence of renal dysfunction in females (30.31%), those with per capita monthly income <2, 000 yuan (27.00%), those who consume alcohol (27.10%), those who are hypertensive (23.05%), those who are exposed to high temperatures (27.37%), those who are exposed to organic solvents (30.42%), and those who are engaged in shift work (25.87%) were to be found had a higher prevalence of renal dysfunction ( P<0.05). After correcting for age, sex, and other risk factors, there was a nonlinear association between intensity of shift work and renal dysfunction, with a hazard ratio ( HR) of 1.29 (95% CI: 0.98-1.59) for the development of renal dysfunction in petroleum workers for shift work performed at higher intensities, and moderate intensity of exposure to shift work reduced the risk of renal dysfunction in petroleum workers ( HR=0.54 with a 95% CI: 0.39-0.75, P<0.001) . Conclusion:Prolonged shift work increases the risk of renal dysfunction in oil workers, and the occurrence of renal dysfunction in oil workers is influenced by multiple factors.

Objective:To establish a quantitative analysis of multi-components by single marker(QAMS)for the determination of 6 alkaloid components,which is benzoylmesaconine,benzoyl-hypaconine,benzoylaconine,mesaconitine,hypaconitine,and aconitine in Zhachong Shisanwei Pills,and prove the scientificity and feasibility of the method in the quality analysis.Methods:The chromatographic separation was performed on an Agilent Eclipse Plus C18(250 mm×4.6 mm,5 μm)with gradient elution using 0.1 mol·L-1 ammonium acetate(0.5 mL of gla-cial acetic acid per 1 000 mL)(A)-acetonitrile:tetrahydrofuran(25∶15)(B),as the mobile phase(0-50 min,18%B-28%B),the detection wavelength was switched from 235 nm,the column temperature was kept at 40℃and the flow rate was 1.0 mL·min-1.The relative correction factors(fs/i)were established with the other 5 compo-nents to be measured using benzoylaconine as the internal reference,which were used to calculate the mass fraction of each component.At the same time,the mass fractions of the 6 effective constituents in Zhachong Shisanwei Pills were calculated by the external standard method(ESM).By comparing the content results of ESMand QAMS,the accura-cy of QAMS method were evaluated.Results:The relative correction factors(fs/i)of benzoylmesaconine,benzoylhyp-aconine,mesaconitine,hypaconitine,and aconitine in Mongolian medicine Zhachong Shisanwei Pills were reproduci-ble with good reproducibility,which were 0.680 4,0.450 6,0.850 8,0.676 1 and 0.757 0,the result obtained by QAMS approximated those obtained by external standard method(ESM).Conclusion:The method is simple,stable and reproducible,and can be used for the quality control of 6 alkaloid components in Zhachong Shisanwei Pills.

To understand the pathogenicity and drug resistance of swine-derived E.coli and its bio-logical characteristics in some areas in Hebei,E.coli was isolated and identified from diarrheal fe-ces of piglets collected from swine farms,and the isolated strains were subjected to drug sensitivity test,detection of the ability to form biofilm,pathogenicity test,virulence gene test,drug resistance gene test,and identification of phylogenetic subgroups.The results showed that a total of 35 patho-genic E.coli strains were isolated from the feces of diarrheic piglets,and most of the isolates were multidrug-resistant,and were resistant to at least three antibiotics,including amoxicillin(88.57％),ampicillin(88.57％),doxycycline(88.75％),sulfisoxazole(77.17％),lincomycin(100％),and chloramphenicol(100％);the isolates were severely resistant.The isolates all carried virulence genes,with a total of five virulence genes detected,namely,EAST1(77.14％),eaeA(17.14％),stx2e(5.71％),LT(2.86％),and STb(2.86％),and the isolates also carried multi-re-sistance genes,with a total of five virulence genes detected,namely,bla TEM-1(65.71％),bla CTX-M(20.00％),tetA(82.86％),tetB(14.29％),aadA2(17.14％),aac(6')-Ib(14.29％),qnrS(17.14％),sul 1(40.00％),sul2(34.29％),and floR(60.00％);the phylogenetic grouping showed that the isolates had a high proportion of group B1 and group A;and all 35 isolates showed differ-ent pathogenicity after infection of mice.This study provides a reference for the selection of effec-tive therapeutic drugs and the development of prevention and control programs for swine-origin pathogenic E.coli in Hebei Province.

Objective Based on ultra-high-performance liquid chromatography-quadrupole time-of-flight tandem mass spectrometry(UPLC-QTOF-MS/MS),a nontarget screening strategy was adopted in analyzing and identifying emerging pollutants(EPs)in tap water across the country,and studying its spatial and temporal distribution characteristics.Methods After extracting tap water samples by solid phase extraction,mobile phases(0.1%aqueous formic acid and methanol solutions)was used to elute the sample on a C18 chromatographic column.The nontarget screening strategy was used to acquire the MS information in full scan mode.We extracted and analyzed the chromatographic and mass spectral peaks,then searched the spectral library to compare the exact mass numbers,and the secondary MS/MS spectra fragment ion information was compared one by one.Finally,the retention time and the mass spectrum data of candidate EPs were identified and analyzed with the data of standard samples.then quantified by the internal standard method.Results A total of 135 EPs were initially screened from tap water across the country and 24 with high chromatographic peak response were finally selected and verified by standard products,including 6 pharmaceuticals,13 pesticides,3 industrial compounds and 2 food additives.Nine of them showed detection rates of more than 60%,such as canrenone,medroxyprogesterone,hydrocortisone acetate,etc.The concentrations of detected pollutants range from ND 422.63 ng/L.And the four contaminants with higher average concentrations were canrenone,medroxyprogesterone,hydrocortisone acetate and tris(2-butoxyethyl)phosphate.Conclusion In this study,a nontarget strategy based on UPLC-QTOF-MS/MS was adopted to screen potential unknown pollutants with no-standards,and explored the overall contamination status of tap water samples.This contributed to more comprehensive understanding of the EPs distribution,and provided technical support for monitoring the EPs in tap water.

Objective:To systematically summarize and comparatively analyze the development, establishment and usage of oncology drugs speedy review approaches in China and in the United States between 2012 and 2021.Methods:Based on National Medical Products Administration (NMPA) and Food and Drug Administration (FDA) websites, the development and current status of the speedy review approaches were consulted and summarized. Approved oncology drugs in China and in the United States (87 in China, 118 in the United States) over the past decade were analyzed using chi-square test for group comparison.Results:Five speedy approaches have been established in China and in the United States, three of which are the same, priority review, conditional approval or accelerated approval and breakthrough therapy. The rest two are special review and approval, special examination and approval in China, and fast track and real-time oncology review in the United States. Compared to the United States, speedy review approaches in China set up late (1992 vs. 2005). The overall utilization rates of the oncology drugs speedy review approaches were similar between the China and United States (90.8% vs. 92.4%, P=0.800) in the previous 10 years, and priority review have highest utilization rates in both China and the United States without significant group difference (77.0% vs. 82.2%, P=0.381); relatively low utilization rates of conditional approval (31.0% vs. 44.9%, P=0.041) and breakthrough therapy (2.3% vs. 50.0%, P＜0.001) were seen in China. 52.9% of new drugs applied for special examination and approval in China and 40.7% of new drugs applied for fast track in the United States. Overall, the priority review both in China and the United States are stable, with a similar average annual utilization rate (84.8% vs. 83.7%); accelerated approval and breakthrough therapies in the United States fluctuate wildly, but the situation is tending towards stability in the last 3 years. Conclusions:Both China and the United States have established a relatively complete accelerated review system, with an overall utilization rate over 90%; China's accelerated review started late, although the overall utilization rate is close to that of the United States. The utilization rates of conditional approval and breakthrough therapy are still relatively low. Flexible usage of speedy review approaches, gaining regulatory recognition to use alternative endpoints, achieving real-time review and guidance are keys to accelerate new drug development in China.