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.

In view of the few studies on the influence of Armillaria spp. infection on the content of the chemical components in different parts of Polyporus umbellatus sclerotia, this study determined the biomass of P. umbellatus sclerotia and the contents of ergosterol, polyporusterone A, polyporusterone B and polysaccharide in the separated cavity wall of the sclerotia and the uninfected part of the sclerotia in different harvesting years under the conditions of A. gallica and A. mellea infection respectively. According to the difference of content and dynamic changes of the polysaccharide and the steroid substances, the superior Armillaria sp. was screened to obtain the best harvest years of P. umbellatus. Using HPLC and UV-VIS spectrophotometry methods, the contents of ergosterol, polyporusterone A, polyporusterone B and polysaccharide in P. umbellatus sclerotia infected by the two Armillaria spp. in different years were determined. In addition, the differentially expressed genes related to P. umbellatus polysaccharide synthesis were screened according to the transcriptomic data of different parts of P. umbellatus after A. mellea infection. With the increase of years, the biomass of sclerotia infected by different Armillaria spp. had significant differences, and there were significant differences in the four components of sclerotia. The four components of the separated cavity wall of the sclerotia were significantly higher than those of the uninfected part. The best harvest time was the third year after cultivation. Transcriptomic analysis showed that the infection of Armillaria spp. could significantly promote polysaccharide synthesis, which provided a basis for polysaccharide content determination at the molecular level. The study clarified the influence of different Armillaria spp. infection on the accumulation of chemical components of P. umbellatus sclerotia, laying a foundation for exploring the symbiosis mechanism and provided a scientific clue for screening superior Armillaria sp. and guiding the artificial cultivation of P. umbellatus sclerotia.

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.

Aim To evaluate the effect of hyperoside/chitosan-nanoparticles(Hyp-NPs)on bone marrow mesenchymal stem cells(BMSCs)in vitro cell experi-ments and the underlying mechanism,and to conduct in vivo animal experiments to investigate the synergistic effect of Hyp-NPs and BMSCs on repairing endometrial damage in rats.Methods BMSCs were identified by flow cytometry.Hyp-NPs were prepared by ion crosslinking method,characterized and evaluated by laser particle size analyzer and transmission electron microscopy.The effects of different concentrations of Hyp-NPs on the migration of BMSCs were evaluated by scratch assay and immunofluorescence.NRF2 lentivir-us vector was constructed to explore the mechanism of Hyp-NPs on BMSCs.In animal experiments,Hyp-NPs loaded with BMSCs were co-transplanted into the uter-ine cavity of a rat model of endometrial injury.HE,Masson,IHC,TUNEL,and ELISA experiments were used to systematically evaluate the repair effect and pregnancy function of the composite formulation on rat endometrial injury from multiple aspects and angles,including general pathology,fibrosis,receptivity,cell proliferation,angiogenesis,stem cell recruitment,and inflammation of the endometrium.Results BMSCs were successfully isolated and cultured.Hyp-NPs with high stability and small particle size were successfully prepared.Scratch experiments indicated that Hyp-NPs could promote the migration of BMSCs.By successfully constructing a lentiviral NRF2 vector and oxidative damage model for BMSCs,immunofluorescence experi-ments showed that Hyp-NPs could regulate the biologi-cal axis of BMSCs by activating NRF2.Animal experi-ments showed that the synergistic administration of Hyp-NPs and BMSCs could increase endometrial thick-ness and glandular quantity,promote stem cell homing through anti-fibrotic,anti-apoptotic,and anti-inflam-matory effects,and restore pregnancy function in rats with endometrial injury.Conclusion The synergistic administration of Hyp-NPs and BMSCs could repair en-dometrial injury.

Aim To evaluate the effect of Pien Tze Huang(PTH)on influenza virus susceptibility in re-straint stress-induced H1N1 influenza susceptibility model in mice with emotional disorders and internal heat,guided by the theory of emotional pathogenesis.Methods Mice were infected with H1N1 influenza vi-rus following 18 h of restraint stress.The signs and weight changes of mice were recorded,and the morbid-ity of mice were analyzed.On the fourth day post viral infection,the lung tissue was collected.The pathologi-cal changes and inflammatory factors in lungs were de-tected by HE staining.The expression of NP was as-sessed by immunohistochemistry and Western blot.The level of lipid peroxidation end products was detected u-sing a commercial kit and western blot.Redox phos-pholipidomics was analyzed in lung tissue by HPLC-MS/MS.Results PTH significantly reduced the mor-tality of influenza-susceptible mice induced by emotion-al stress,inhibited the expression of NP and the re-lease of inflammatory factors,improved inflammation in lung tissue,and alleviated the accumulation of lipid peroxidation end products.Phospholipid oxidation a-nalysis revealed the elevated levels of oxidized phos-pholipid choline and phosphatidylethanolamine in lung tissue of influenza-susceptible mice,which were signif-icantly reduced by PTH administration.Conclusions PTH exhibits promising efficacy in ameliorating influ-enza virus susceptibility induced by internal heat,and its mechanism of action may be related to the regulation of phospholipid peroxidation.

Objective To investigate the effect of Polyetheretherketone(PEEK)rod semi-rigid pedicle screw fixation sys-tem in lumbar spine non-fusion surgery.Methods A total of 74 patients with tow-level lumbar degenerative diseases who un-derwent surgery from March 2017 to December 2019 were divided into PEEK rod group and titanium rod group.In the PEEK rod group,there were 34 patients,including 13 males and 21 females,aged from 51 to 79 years old with an average of(62.4±6.8)years old;There were 1 patient of L1-L3 segments,7 patients of L2-L4 segments,20 patients of L3-L5 segments and 6 pa-tients of L4-S1 segments.In the titanium rod group,there were 40 patients,including 17 males and 23 females,aged from 52 to 81 years old with an average of(65.2±7.3)years old;There were 3 patient of L1-L3 segments,11 patients of L2-L4 segments,19 patients of L3-L5 segments and 7 patients of L4-S1 segments.The general conditions of operation,such as operation time,intraoperative blood loss,postoperative drainage was recorded.The visual analogue scale(VAS)for low back pain and Os-westry disability index(ODI)were compared in preoperatively and postoperatively(3 months,12 months and last follow-up)between two groups.The change of range of motion(ROM)was observed by flexion and extension x-ray of lumbar Results All patients successfully completed the operation.The follow-up time ranged from 22 to 34 months with an average of(26.8±5.6)months.The operative time(142.2±44.7)min and intraoperative blood loss(166.5±67.4)ml in PEEK group were lower than those in titanium group[(160.7±57.3)min、(212.8±85.4)ml](P＜0.05).There was no significant differences in postoperative drainage between the two groups(P＞0.05).At the final follow-up visit,in PEEK group and titanium group VAS of low back pain[(0.8±0.4)points vs(1.0±0.5)points],VAS for leg pain[(0.7±0.4)points vs(0.8±0.5)points]and ODI[(9.8±1.6)％vs(12.1±1.5)％]were compared with preoperative[(5.8±1.1)points vs(6.0±1.1)points],[(7.2±1.7)points vs(7.0±1.6)points],[(68.5±8.9)％vs(66.3±8.2)％]were significantly different(P＜0.05).There was no significant difference in VAS scores between the two groups at each postoperative time point(P＞0.05).At 3 months after surgery,there was no difference in ODI between the two groups(P＞0.05).There were significant differences in ODI between PEEK group and titanium rod group at 12 months[(15.5±2.1)％vs(18.4±2.4)％]and at the last follow-up[(9.8±1.6)％vs(12.1±1.5)％](P＜0.05).The total range of motion(ROM)of lumbar decreased in both groups after surgery.At 12 months after surgery and the last follow-up,the PEEK group compared with the titanium rod group,the total range of motion of lumbar was statistically significant(P＜0.05).The range of motion(ROM)of the fixed segments decreased in both groups after surgery.The ROM of the fixed segments in PEEK group decreased from(9.5±4.6)° to(4.1±1.9)° at the last follow-up(P＜0.05),which in the titanium rod group was de-creased from(9.8±4.3)°to(0.9±0.5)° at the last follow-up(P＜0.05).The range of motion(ROM)of upper adjacent segment increased in both groups,there was statistical significance in the ROM of upper adjacent segment between the two groups at 12 months after surgery and the last follow-up,(P＜0.05).There was no screw loosening and broken rods in both groups during the follow-up period.Conclusion The PEEK rod semi-rigid pedicle screw internal fixation system used in lumbar non-fusion surgery can retain part of the mobility of the fixed segment,showing comparable short-term clinical efficacy to titanium rod fu-sion.PEEK rod semi-rigid pedicle screw internal fixation system is a feasible choice for the treatment of lumbar spine degener-ative diseases,and its long-term efficacy needs further follow-up observation.

Objective: To investigate the correlation between magnetic resonance imaging-based vertebral bone quality (VBQ) score and screw loosening after dynamic pedicle screw fixation with polyetheretherketone (PEEK) rods, and evaluate its predictive value. Methods: A retrospective analysis was conducted on the patients who underwent dynamic pedicle screw fixation with PEEK rods from March 2017 to June 2022. Data on age, sex, body mass index, hypertension, diabetes, hyperlipidemia history, long-term smoking, alcohol consumption, VBQ score, L1–4 average Hounsfield unit (HU) value, surgical fixation length, and the lowest instrumented vertebra were collected. Logistic regression analysis was employed to assess the relationship between VBQ score and pedicle screw loosening (PSL). Results: A total of 24 patients experienced PSL after surgery (20.5%). PSL group and non-PSL group showed statistical differences in age, number of fixed segments, fixation to the sacrum, L1–4 average HU value, and VBQ score (p < 0.05). The VBQ score in the PSL group was higher than that in the non-PSL group (3.56 ± 0.45 vs. 2.77 ± 0.31, p < 0.001). In logistic regression analysis, VBQ score (odds ratio, 3.425; 95% confidence interval, 1.552–8.279) were identified as independent risk factors for screw loosening. The area under the receiver operating characteristic curve for VBQ score predicting PSL was 0.819 (p < 0.05), with the optimal threshold of 3.15 (sensitivity, 83.1%; specificity, 80.5%). Conclusion The VBQ score can independently predict postoperative screw loosening in patients undergoing lumbar dynamic pedicle screw fixation with PEEK rods, and its predictive value is comparable to HU value.