RÉSUMÉ
Objective@#To explore the performance of machine learning prediction models in forecasting student absenteeism due to respiratory symptoms caused by air pollution in short term, aiming to provide a methodological reference for early warning systems of school diseases.@*Methods@#Utilizing data from shortterm sequences of student absenteeism due to respiratory symptoms in Jiangsu Province from September 2019 to October 2022, the study integrated average concentrations of atmospheric pollutants. A univariate distributed lag nonlinear model was employed to select optimal lag variables for the pollutants. An extreme gradient boosting(XGBoost) algorithm model was developed to predict the frequency of absenteeism due to respiratory symptoms and compared with the seasonal autoregressive integrated moving average with exogenous factors(SARIMAX) model.@*Results@#Between 2019 and 2022, an average of 9 709 students per day in Jiangsu Province were absent due to respiratory symptoms. The daily average air quality index (AQI) was 76.96,with mass concentrations of PM2.5, PM10, NO2, and O3 averaging at 35.75, 61.13, 28.89, 104.81 μg/m3, respectively. Granger causality tests indicated that AQI, PM2.5, PM10, NO2, and O3 were significant predictors of absenteeism frequency due to respirutory symptoms(F=1.46,1.79,1.67,3.41,2.18,P<0.01). The singleday lag effects of PM2.5, PM10, NO2, and O3 reached their peak relative risk (RR) values at lag4, lag0, lag0, lag4 respectively. When integrating these optimal lag variables for the pollutants, the XGBoost model demonstrated superior predictive performance to the SARIMAX model, reducing the mean absolute error (MAE) from 2.251 to 0.475, mean absolute percentage error (MAPE) from 0.429 to 0.080, and root mean square error (RMSE) from 2.582 to 0.713; at the P75 percentile alert threshold, the sensitivity improved from 0.086 to 0.694 and specificity from 0.979 to 0.988, with the Youden index increasing from 0.065 to 0.682.@*Conclusions@#The XGBoost model exhibits robust predictive performance and effective early warning capabilities for shortterm sequences of student absenteeism due to respiratory symptoms caused by air pollution. Schools could timely adopt this model to preemptively detect and control disease outbreaks, thereby enhancing school health management.
RÉSUMÉ
Objective:To establish an early prediction model with multiple indicators to predict the risk of severe acute pancreatitis (SAP) in hyperlipidemic acute pancreatitis (HLAP).Methods:The clinical data of 92 patients with HLAP admitted to the Emergency Department of our hospital from March 2018 to February 2020 were analyzed retrospectively. Among them, 29 cases deteriorated to SAP and 63 cases did not. Univariate analysis was used to screen predictive indicators related to hyperlipidemic severe acute pancreatitis (HL-SAP), and logistic regression analysis was used to screen independent predictive indicators related to HL-SAP. Then a prediction model was established. The area under (AUC) the receiver operating curve (ROC) was used to evaluate the predictive ability of each predictive indicator and the model for HL-SAP. Bootstrap resampling technology was used to validate the predictive ability of the model.Results:Univariate analysis showed that procalcitonin, D-dimer, C-reactive protein, albumin, cholesterol and CT grade had influence on the progression of HLAP to SAP ( P<0.05). Logistic regression analysis showed that D-dimer ( OR=2.112, 95% CI: 1.022-4.366; P<0.05), CT grade ( OR=5.818, 95% CI: 2.481-13.643; P<0.01) and cholesterol ( OR=1.146, 95% CI: 1.004-1.308; P<0.05) were independent risk factor of HL-SAP. The AUC of D-dimer, CT grade, cholesterol and the model were 0.802, 0.875, 0.665 and 0.927, respectively. Internal validation of the predictive ability of the model showed that the C-index was 0.927. Conclusions:In the early phase, application of the prediction model that composes D-dimer, CT grade and cholesterol has a good predictive effect on HL-SAP.
RÉSUMÉ
Objective To investigate the effects of hydrogen (H2) on myocardium injury post-cardiac arrest (CA) in rabbits.Methods Sixty New Zealand rabbits were randomly divided into H2 treatment group (n = 30) and control group (n = 30) by random number table. The rabbit CA model was established by means of electrical stimulation of external membrane, both groups were mechanically ventilated. Cardiopulmonary resuscitation (CPR) was performed after 6 minutes of nonintervention, and stopped after restoration of spontaneous circulation (ROSC). Inhalation of 2％ H2 gas was conferred to rabbits immediately at the end of CA modeling for 72 hours in H2 treatment group. Air was given to rabbits in control group instead. The survival rate of rabbits was analyzed. Heart rate, ventricular premature beat frequency, and the levels of blood samples cardiac troponin I (cTnI), left ventricular ejection fraction (LVEF), B-type natriuretic peptide (BNP), and blood lactic acid (Lac) were collected before CA and after ROSC in all rabbits. Rabbits were sacrificed and microstructure injury was observed by electric microscope after ROSC 72 hours.Results There were 28 animals ROSC in both groups; the survival number in H2 treatment group was higher than that in control group at 72 hours after ROSC (number: 15 vs. 7,χ2 = 4.791,P = 0.029). In the early stage of ROSC, the heart rate of two groups slowed down, the number of premature ventricular increased, and then gradually recovered; the heart rate in H2 treatment group was returning to normal more quickly than that in control group at 48 hours after ROSC (bpm: 319±63 vs. 362±40,P < 0.05); the ventricular premature beat frequency was lower than that in control group at 72 hours after ROSC (times per minutes: 9.1±4.3 vs. 15.0±8.0,P < 0.05). The animals of two groups had different degrees of myocardial damage and cardiac insufficiency after ROSC, and restored with the extension of time. Compared with control group, the level of BNP in H2 treatment group was significant decreased at 24 hours after ROSC (ng/L: 385±98 vs. 488±174,P < 0.05), the levels of cTnI and Lac were significant decreased at 48 hours after ROSC [cTnI (μg/L:1.83±0.68 vs. 2.83±0.98, Lac (mmol/L): 5.5±1.6 vs. 7.9±2.6, bothP < 0.01], the LVEF was slightly higher than that at 72 hours after ROSC (0.690±0.040 vs. 0.650±0.041,P = 0.051). Compared with control group, less damage to myocardial ultra structure was found in H2 treatment group at 72 hours after ROSC.Conclusion Inhalation of H2 alleviates cardiac dysfunction and myocardial injury after CPR.
RÉSUMÉ
Objective To investigate the effects of hydrogen (H2) on myocardium injury post-cardiac arrest (CA) in rabbits.Methods Sixty New Zealand rabbits were randomly divided into H2 treatment group (n = 30) and control group (n = 30) by random number table. The rabbit CA model was established by means of electrical stimulation of external membrane, both groups were mechanically ventilated. Cardiopulmonary resuscitation (CPR) was performed after 6 minutes of nonintervention, and stopped after restoration of spontaneous circulation (ROSC). Inhalation of 2％ H2 gas was conferred to rabbits immediately at the end of CA modeling for 72 hours in H2 treatment group. Air was given to rabbits in control group instead. The survival rate of rabbits was analyzed. Heart rate, ventricular premature beat frequency, and the levels of blood samples cardiac troponin I (cTnI), left ventricular ejection fraction (LVEF), B-type natriuretic peptide (BNP), and blood lactic acid (Lac) were collected before CA and after ROSC in all rabbits. Rabbits were sacrificed and microstructure injury was observed by electric microscope after ROSC 72 hours.Results There were 28 animals ROSC in both groups; the survival number in H2 treatment group was higher than that in control group at 72 hours after ROSC (number: 15 vs. 7,χ2 = 4.791,P = 0.029). In the early stage of ROSC, the heart rate of two groups slowed down, the number of premature ventricular increased, and then gradually recovered; the heart rate in H2 treatment group was returning to normal more quickly than that in control group at 48 hours after ROSC (bpm: 319±63 vs. 362±40,P < 0.05); the ventricular premature beat frequency was lower than that in control group at 72 hours after ROSC (times per minutes: 9.1±4.3 vs. 15.0±8.0,P < 0.05). The animals of two groups had different degrees of myocardial damage and cardiac insufficiency after ROSC, and restored with the extension of time. Compared with control group, the level of BNP in H2 treatment group was significant decreased at 24 hours after ROSC (ng/L: 385±98 vs. 488±174,P < 0.05), the levels of cTnI and Lac were significant decreased at 48 hours after ROSC [cTnI (μg/L:1.83±0.68 vs. 2.83±0.98, Lac (mmol/L): 5.5±1.6 vs. 7.9±2.6, bothP < 0.01], the LVEF was slightly higher than that at 72 hours after ROSC (0.690±0.040 vs. 0.650±0.041,P = 0.051). Compared with control group, less damage to myocardial ultra structure was found in H2 treatment group at 72 hours after ROSC.Conclusion Inhalation of H2 alleviates cardiac dysfunction and myocardial injury after CPR.
RÉSUMÉ
Objective To compare the screening effects between Wells and revised Geneva scores on suspected acute pulmonary thromboembolism (APTE),and to explore a optimum screening method for APTE in the emergency department of China.Methods The study was carried out by using random,crossed,prospective methods to compare the screening effects between Wells and revised Geneva scores for 167 suspected APTE patients in the emergency department of the First Affiliated Hospital of Xiamen University.Results The areas under the receiver operating characteristic curve of Wells and revised Geneva scores for screening APTE in the emergency department were (0.917 ± 0.022 ) and (0.927 ± 0.020),respectively ( P ＜ 0.05 ).The diagnostic concordance between the two score systems for predicting APTE was poor (Kappa value =0.276 ). In addition, the difference between their hierarchical discrimination for the possibility of APTE was statistically significant ( P ＜ 0.05 ).Compared with revised Geneva score,fewer patients were diagnosed with low clinical probability of APTE and more patients were diagnosed with intermediate or high clinical probability of APTE through Wells score.The patients with low chnical probability of APTE were excluded from pulmonary embolism in Wells or revised Geneva score.At intermediate clinical probability,the accuracy rate of Wells score for predicting APTE (9.64％) was lower than that (32.84％ ) of revised Geneva ( P ＜ 0.05 ).At high clinical probability,there was no significant difference between their accuracy rate [ (67.24％ vs.86.21％),P＞0.05]. Conclusions Revised Geneva score is more suitable than Wells score in screening suspected APTE patients in the emergency department in our country.
RÉSUMÉ
A total of 192 patients with sepsis were tested by Montreal Cognitive Assessment (MoCA) for a preliminary diagnosis of whether or not there was sepsis associated encephalopathy (SAE) according to their test results.SAE was diagnosed or excluded after consultations and comprehensive analysis on the basis of clinical manifestations and auxiliary examination results.The scores of the patients in this group were (25.7 ± 3.3) points.The sensitivity of MoCA for screening SAE was 0.776 and its specificity 0.963.The rate of diagnostic coincidence between MoCA and comprehensive analysis for SAE was 0.880.The diagnostic concordance between two diagnostic methods of SAE was excellent (kappa value =0.753 ± 0.048,P =0.000).The area under the receiver operating characteristic (ROC) curve of MoCA for screening SAE was 0.929 ± 0.019 (P =0.000) ; the optimal cutoff value was 25.5 points; and its sensitivity was 0.779 and specificity 0.962.And negative correlations existed between score of MoCA and age,disease course and co-existing shock or multiple organ dysfunction syndrome (P ＜ 0.05).