Objective: Machine learning (ML) has been reported to have better predictive capability than traditional statistical techniques. The aim of this study was to assess the efficacy of ML algorithms and logistic regression (LR) for predicting depressive symptoms during the COVID-19 pandemic. Methods: Analyses were carried out in a national cross-sectional study involving 21,916 participants. The ML algorithms in this study included random forest (RF), support vector machine (SVM), neural network (NN), and gradient boosting machine (GBM) methods. The performance indices were sensitivity, specificity, accuracy, precision, F1-score, and area under the receiver operating characteristic curve (AUC). Results: LR and NN had the best performance in terms of AUCs. The risk of overfitting was found to be negligible for most ML models except for RF, and GBM obtained the highest sensitivity, specificity, accuracy, precision, and F1-score. Therefore, LR, NN, and GBM models ranked among the best models. Conclusion Compared with ML models, LR model performed comparably to ML models in predicting depressive symptoms and identifying potential risk factors while also exhibiting a lower risk of overfitting.

Objective: Machine learning (ML) has been reported to have better predictive capability than traditional statistical techniques. The aim of this study was to assess the efficacy of ML algorithms and logistic regression (LR) for predicting depressive symptoms during the COVID-19 pandemic. Methods: Analyses were carried out in a national cross-sectional study involving 21,916 participants. The ML algorithms in this study included random forest (RF), support vector machine (SVM), neural network (NN), and gradient boosting machine (GBM) methods. The performance indices were sensitivity, specificity, accuracy, precision, F1-score, and area under the receiver operating characteristic curve (AUC). Results: LR and NN had the best performance in terms of AUCs. The risk of overfitting was found to be negligible for most ML models except for RF, and GBM obtained the highest sensitivity, specificity, accuracy, precision, and F1-score. Therefore, LR, NN, and GBM models ranked among the best models. Conclusion Compared with ML models, LR model performed comparably to ML models in predicting depressive symptoms and identifying potential risk factors while also exhibiting a lower risk of overfitting.

Objective: Machine learning (ML) has been reported to have better predictive capability than traditional statistical techniques. The aim of this study was to assess the efficacy of ML algorithms and logistic regression (LR) for predicting depressive symptoms during the COVID-19 pandemic. Methods: Analyses were carried out in a national cross-sectional study involving 21,916 participants. The ML algorithms in this study included random forest (RF), support vector machine (SVM), neural network (NN), and gradient boosting machine (GBM) methods. The performance indices were sensitivity, specificity, accuracy, precision, F1-score, and area under the receiver operating characteristic curve (AUC). Results: LR and NN had the best performance in terms of AUCs. The risk of overfitting was found to be negligible for most ML models except for RF, and GBM obtained the highest sensitivity, specificity, accuracy, precision, and F1-score. Therefore, LR, NN, and GBM models ranked among the best models. Conclusion Compared with ML models, LR model performed comparably to ML models in predicting depressive symptoms and identifying potential risk factors while also exhibiting a lower risk of overfitting.

Objective: Machine learning (ML) has been reported to have better predictive capability than traditional statistical techniques. The aim of this study was to assess the efficacy of ML algorithms and logistic regression (LR) for predicting depressive symptoms during the COVID-19 pandemic. Methods: Analyses were carried out in a national cross-sectional study involving 21,916 participants. The ML algorithms in this study included random forest (RF), support vector machine (SVM), neural network (NN), and gradient boosting machine (GBM) methods. The performance indices were sensitivity, specificity, accuracy, precision, F1-score, and area under the receiver operating characteristic curve (AUC). Results: LR and NN had the best performance in terms of AUCs. The risk of overfitting was found to be negligible for most ML models except for RF, and GBM obtained the highest sensitivity, specificity, accuracy, precision, and F1-score. Therefore, LR, NN, and GBM models ranked among the best models. Conclusion Compared with ML models, LR model performed comparably to ML models in predicting depressive symptoms and identifying potential risk factors while also exhibiting a lower risk of overfitting.

Objective: Machine learning (ML) has been reported to have better predictive capability than traditional statistical techniques. The aim of this study was to assess the efficacy of ML algorithms and logistic regression (LR) for predicting depressive symptoms during the COVID-19 pandemic. Methods: Analyses were carried out in a national cross-sectional study involving 21,916 participants. The ML algorithms in this study included random forest (RF), support vector machine (SVM), neural network (NN), and gradient boosting machine (GBM) methods. The performance indices were sensitivity, specificity, accuracy, precision, F1-score, and area under the receiver operating characteristic curve (AUC). Results: LR and NN had the best performance in terms of AUCs. The risk of overfitting was found to be negligible for most ML models except for RF, and GBM obtained the highest sensitivity, specificity, accuracy, precision, and F1-score. Therefore, LR, NN, and GBM models ranked among the best models. Conclusion Compared with ML models, LR model performed comparably to ML models in predicting depressive symptoms and identifying potential risk factors while also exhibiting a lower risk of overfitting.

Objective To investigate the effects of epithelial transformation sequence 2(ECT2)and p33ING1 on the metastatic activity of esophageal squamous cell carcinoma(ESCC)cells.Methods The expressions of ECT2 and p33ING1 in esophageal squamous cell carcinoma tissues and adjacent tissues were detected by immunohistochemistry and Western blotting.Human esophageal squamous carcinoma cell line KYSE140 cells were divided into 4 groups:blank group,negative control(pcDNA 3.1 NC)group,overexpression group(pcDNA 3.1 ECT2)and inhibited expression group(si ECT2).MTT assay and cell colony formation assay were used to study the proliferation and growth ability of cells,Transwell assay and scratch assay used to study the invasion and migration ability of cells,and flow cytometry used to detect apoptosis and cell cycle,Western blotting used to detect the effect of ECT2 on p33ING1 protein.Results ECT2 expression increased and p33ING1 expression decreased in esophageal squamous cell carcinoma tissues.Overexpression of ECT2 significantly increased the growth,colony formation,migration and invasion abilities of KYSE140 cells,and decreased the apoptosis rate and p33ING1 expression of KYSE140 cells.In addition,inhibition of ECT2 expression could reverse the above changes.Conclusion The high expression of ECT2 can promote the growth and metastasis of esophageal squamous cell carcinoma KYSE140 cells and inhibit their apoptosis.The mechanism may be related to the inhibition of p33ING1 expression by ECT2.

Objective:To study the current status of longitudinal extrauterine growth restriction (EUGR) in extremely preterm infants (EPIs) and to develop a prediction model based on clinical data from multiple NICUs.Methods:From January 2017 to December 2018, EPIs admitted to 32 NICUs in North China were retrospectively studied. Their general conditions, nutritional support, complications during hospitalization and weight changes were reviewed. Weight loss between birth and discharge > 1SD was defined as longitudinal EUGR. The EPIs were assigned into longitudinal EUGR group and non-EUGR group and their nutritional support and weight changes were compared. The EPIs were randomly assigned into the training dataset and the validation dataset with a ratio of 7∶3. Univariate Cox regression analysis and multiple regression analysis were used in the training dataset to select the independent predictive factors. The best-fitting Nomogram model predicting longitudinal EUGR was established based on Akaike Information Criterion. The model was evaluated for discrimination efficacy, calibration and clinical decision curve analysis.Results:A total of 436 EPIs were included in this study, with a mean gestational age of (26.9±0.9) weeks and a birth weight of (989±171) g. The incidence of longitudinal EUGR was 82.3%(359/436). Seven variables (birth weight Z-score, weight loss, weight growth velocity, the proportion of breast milk ≥75% within 3 d before discharge, invasive mechanical ventilation ≥7 d, maternal antenatal corticosteroids use and bronchopulmonary dysplasia) were selected to establish the prediction model. The area under the receiver operating characteristic curve of the training dataset and the validation dataset were 0.870 (95% CI 0.820-0.920) and 0.879 (95% CI 0.815-0.942), suggesting good discrimination efficacy. The calibration curve indicated a good fit of the model ( P>0.05). The decision curve analysis showed positive net benefits at all thresholds. Conclusions:Currently, EPIs have a high incidence of longitudinal EUGR. The prediction model is helpful for early identification and intervention for EPIs with higher risks of longitudinal EUGR. It is necessary to expand the sample size and conduct prospective studies to optimize and validate the prediction model in the future.

【Objective】 To objectively evaluate the quality control level of blood testing process in blood banks through quantitative monitoring and trend analysis, and to promote the homogenization level and standardized management of blood testing laboratories in blood banks. 【Methods】 A quality monitoring indicator system covering the whole process of blood collection and supply, including blood donation service, blood component preparation, blood testing, blood supply and quality control was established. The questionnaire Quality Monitoring Indicators for Blood Collection and Supply Process with clear definition of indicators and calculation formulas was distributed to 17 blood banks in Shandong province. Quality monitoring indicators of each blood bank from January to December 2022 were collected, and 31 indicators in terms of blood testing were analyzed using SPSS25.0 software. 【Results】 The proportion of unqualified serological tests in 17 blood bank laboratories was 55.84% for ALT, 13.63% for HBsAg, 5.08% for anti HCV, 5.62% for anti HIV, 18.18% for anti TP, and 1.65% for other factors (mainly sample quality). The detection unqualified rate and median were (1.23±0.57)% and 1.11%, respectively. The ALT unqualified rate and median were (0.74±0.53)% and 0.60%, respectively. The detection unqualified rate was positively correlated with ALT unqualified rate (r=0.974, P<0.05). The unqualified rate of HBsAg, anti HCV, anti HIV and anti TP was (0.15±0.09)%, (0.05±0.04)%, (0.06±0.03)% and (0.20±0.05)% respectively. The average unqualified rate, average hemolysis rate, average insufficient volume rate and the abnormal hematocrit rate of samples in 17 blood bank laboratories was 0.21‰, 0.08‰, 0.01‰ and 0.02‰ respectively. There were differences in the retest concordance rates of four HBsAg, anti HCV and anti HIV reagents, and three anti TP reagents among 17 blood bank laboratories (P<0.05). The usage rate of ELISA reagents was (114.56±3.30)%, the outage rate of ELISA was (10.23±7.05) ‰, and the out of range rate of ELISA was (0.90±1.17) ‰. There was no correlation between the out of range rate, outrage rate and usage rate (all P>0.05), while the outrage rate was positively correlated with the usage rate (r=0.592, P<0.05). A total of 443 HBV DNA positive samples were detected in all blood banks, with an unqualified rate of 3.78/10 000; 15 HCV RNA positive samples were detected, with an unqualified rate of 0.13/10 000; 5 HIV RNA positive samples were detected, with an unqualified rate of 0.04/10 000. The unqualified rate of NAT was (0.72±0.04)‰, the single NAT reaction rate [(0.39±0.02)‰] was positively correlated with the single HBV DNA reaction rate [ (0.36±0.02) ‰] (r=0.886, P<0.05). There was a difference in the discriminated reactive rate by individual NAT among three blood bank laboratories (C, F, H) (P<0.05). The median resolution rate of 17 blood station laboratories by minipool test was 36.36%, the median rate of invalid batch of NAT was 0.67%, and the median rate of invalid result of NAT was 0.07‰. The consistency rate of ELISA dual reagent detection results was (99.63±0.24)%, and the median length of equipment failure was 14 days. The error rate of blood type testing in blood collection department was 0.14‰. 【Conclusion】 The quality monitoring indicator system for blood testing process in Shandong can monitor potential risks before, during and after the experiment, and has good applicability, feasibility, and effectiveness, and can facilitate the continuous improvement of laboratory quality control level. The application of blood testing quality monitoring indicators will promote the homogenization and standardization of blood quality management in Shandong, and lay the foundation for future comprehensive evaluations of blood banks.

【Objective】 To establish an effective quality monitoring indicator system for blood quality control in blood banks, in order to analyze the quality control indicators for blood collection and supply, and evaluate blood quality control process, thus promoting continuous improvement and standardizing management of blood quality control in blood banks. 【Methods】 A quality monitoring indicator system covering the whole process of blood collection and supply, including blood donation services, component preparation, blood testing, blood supply and quality control was established. The Questionnaire of Quality Monitoring Indicators for Blood Collection and Supply Process was distributed to 17 blood banks in Shandong, which clarified the definition and calculation formula of indicators. The quality monitoring indicator data from January to December 2022 in each blood bank were collected, and 20 quality control indicators data were analyzed by SPSS25.0 software. 【Results】 The average pass rate of key equipment monitoring, environment monitoring, key material monitoring, and blood testing item monitoring of 17 blood banks were 99.47%, 99.51%, 99.95% and 98.99%, respectively. Significant difference was noticed in the pass rate of environment monitoring among blood banks of varied scales(P<0.05), and the Pearson correlation coefficient (r) between the total number of blood quality testing items and the total amount of blood component preparation was 0.645 (P<0.05). The average discarding rates of blood testing or non-blood testing were 1.14% and 3.36% respectively, showing significant difference among blood banks of varied scales (P<0.05). The average discarding rate of lipemic blood was 3.07%, which had a positive correlation with the discarding rate of non testing (r=0.981 3, P<0.05). There was a statistically significant difference in the discarding rate of lipemic blood between blood banks with lipemic blood control measures and those without (P<0.05). The average discarding rate of abnormal color, non-standard volume, blood bag damage, hemolysis, blood protein precipitation and blood clotting were 0.20%, 0.14%, 0.06%, 0.06%, 0.02% and 0.02% respectively, showing statistically significant differences among large, medium and small blood banks(P<0.05).The average discarding rates of expired blood, other factors, confidential unit exclusion and unqualified samples were 0.02%, 0.05%, 0.003% and 0.004%, respectively. The discarding rate of blood with air bubbles was 0.015%, while that of blood with foreign body and unqualified label were 0. 【Conclusion】 The quality control indicator system of blood banks in Shandong can monitor weak points in process management, with good applicability, feasibility, and effectiveness. It is conducive to evaluate different blood banks, continuously improve the quality control level of blood collection and supply, promote the homogenization and standardization of blood quality management, and lay the foundation for comprehensive evaluation of blood banks in Shandong.

【Objective】 To establish an effective quality indicator monitoring system, scientifically and objectively evaluate the quality management level of blood banks, and achieve continuous improvement of quality management in blood bank. 【Methods】 A quality monitoring indicator system that covers the whole process of blood collection and supply was established, the questionnaire of Quality Monitoring Indicators for Blood Collection and Supply Process with clear definition of indicators and calculation formulas was distributed to 17 blood banks in Shandong. Statistical analysis of 21 quality monitoring indicators in terms of blood donation service (10 indicators), blood component preparation (7 indicators ), and blood supply (4 indicators) from each blood bank from January to December 2022 were conducted using SPSS25.0 software The differences in quality monitoring indicators of blood banks of different scales were analyzed. 【Results】 The average values of quality monitoring indicators for blood donation service process of 17 blood banks were as follows: 44.66% (2 233/5 000) of regular donors proportion, 0.22% (11/50) of adverse reactions incidence, 0.46% (23/5 000) of non-standard whole blood collection rate, 0.052% (13/25 000) of missed HBsAg screening rate, 99.42% (4 971/5 000) of first, puncture successful rate, 86.49% (173/200) of double platelet collection rate, 66.50% (133/200) of 400 mL whole blood collection rate, 99.25% (397/400) of donor satisfaction rate, 82.68% (2 067/2 500) of use rate of whole blood collection bags with bypass system with sample tube, and 1 case of occupational exposure in blood collection.There was a strong positive correlation between the proportion of regular blood donors and the collection rate of 400 mL whole blood (P<0.05). The platelet collection rate, incidence of adverse reactions to blood donation, and non-standard whole blood collection rate in large blood banks were significantly lower than those in medium and small blood banks (P<0.05). The average quality monitoring indicators for blood component preparation process of 17 blood banks were as follows: the leakage rate of blood component preparation bags was 0.03% (3/10 000), the discarding rate of lipemic blood was 3.05% (61/2 000), the discarding rate of hemolysis blood was 0.13%(13/10 000). 0.06 case had labeling errors, 8 bags had blood catheter leaks, 2.76 bags had blood puncture/connection leaks, and 0.59 cases had non-conforming consumables. The discarding rate of hemolysis blood of large blood banks was significantly lower than that of medium and small blood banks (P<0.05), and the discarding rate of lipemic blood of large and medium blood banks was significantly lower than that of small blood banks (P<0.05). The average values of quality monitoring indicators for blood supply process of 17 blood banks were as follows: the discarding rate of expired blood was 0.023% (23/100 000), the leakage rate during storage and distribution was of 0.009%(9/100 000), the discarding rate of returned blood was 0.106% (53/50 000), the service satisfaction of hospitals was 99.16% (2 479/2 500). The leakage rate of blood components during storage and distribution was statistically different with that of blood component preparation bags between different blood banks (P<0.05). There were statistically significant differences in the proportion of regular blood donors, incidence of adverse reactions, non-standard whole blood collection rate, 400 mL whole blood collection rate, double platelet collection rate, the blood bag leakage rate during preparation process, the blood components leakage rate during storage and distribution as well as the discarding rate of lipemic blood, hemolysis blood, expired blood and returned blood among large, medium and small blood banks (all P<0.05). 【Conclusion】 The establishment of a quality monitoring indicator system for blood donation services, blood component preparation and blood supply processes in Shandong has good applicability, feasibility and effectiveness. It can objectively evaluate the quality management level, facilitate the continuous improvement of the quality management system, promote the homogenization of blood management in the province and lay the foundation for future comprehensive evaluation of blood banks.