The abnormal accumulation of neutrophil extracellular traps （NETs） can promote the initiation and progression of pancreatic cancer， which is considered a potential therapeutic target for this disease. The Miraculous Pivot·Inquiry About Statement （《灵枢·口问》） have recorded the concept of "defensive qi stagnation". Based on the recognition that the function of defensive qi is similar to the immune function of neutrophils， and combining traditional Chinese medicine theory with clinical practice， it is proposed that the abnormal accumulation of NETs may be a pathological product of "defensive qi stagnation"， with the spleen being the critical site of pathology. Further exploring the application strategy of cleaning spleen and restoring defensive qi method in pancreatic cancer treatment， it is proposed to employ three approaches such as dredging method to eliminate spleen stagnation and inhibit pancreatic cancer proliferation， cleaning method to remove spleen dampness and suppress the inflammatory micro-environment， and tonifying method to strengthen Weiqi and to improve the immune microenvironment， which aims to provide new insights for the clinical treatment of pancreatic cancer with traditional Chinese medicine.

This article combined the pathogenic characteristics of "latent wind" with the theory of collateral diseases to clarify the pathological features of tumor blood vessels， including their active proliferation， high permeabi-lity， and promotion of metastasis. The theory framework of "latent wind in collaterals" as the tumor mechanism was proposed， which suggests that at the site of tumor lesions， the collaterals inherit the nature of latent wind to grow excessively， adopt an open and discharge nature to leak essence， and tumor toxins， characterized by their rapid movement and frequent changes， spread and metastasize， driving the progression of malignant tumors. Focusing on the fundamental pathogenesis of "latent wind in collaterals"， specific clinical treatment principles and methods centered on treating wind are proposed， including regulating qi and dispelling wind， clearing heat and extinguishing wind， unblocking collaterals and expelling wind， and reinforcing healthy qi to calm wind， so as to provide references for enhancing the precision of traditional Chinese medicine in treating malignant tumors.

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 order to explore the possible role and molecular mechanism of the combined action of leech and bear bile in liver and gallbladder diseases, this study first used network pharmacology methods to screen the components and targets of leech and bear bile, as well as the related target genes of liver and gallbladder diseases. The selected key genes were subjected to interaction network and GO/KEGG enrichment analysis. Then, using sodium oleate induced HepG2 cell lipid deposition model and DL-ethionine induced mouse fatty liver model, the activity of leech and bear bile alone and in combination in reducing liver fat was evaluated in vitro and in vivo, and the expression of key pathway related proteins suggested by network pharmacology was detected by Western blot. The results of network pharmacology analysis showed that the active ingredients of leech and bear bile have 295 intersecting targets with liver and gallbladder related diseases, involving more than 200 signaling pathways, including the PI3K/Akt signaling pathway and phospholipase D signaling pathway closely related to glucose and lipid metabolism. The results of in vitro validation experiments showed that both leech and bear bile, alone and in combination, can significantly inhibit the lipid deposition induced by sodium oleate in human liver cells, reduce the triacylglycerol level in cell culture supernatant, and inhibit the lipid content in liver cells. The observation results of Nile red staining confocal microscopy showed that the combination of leech and bear bile had better activity in reducing lipid deposition in liver cells compared to using them alone. In a mouse fatty liver model, the combination of leech and bear bile can better reduce elevated organ indices, blood lipids, and liver lipid levels, as well as lower the levels of serum liver injury biomarkers. The animals used in this experiment and related disposal meet the requirements of animal welfare. Before the experiment, it was reviewed and approved by IACUC, Institute of Materia Medica, Chinese Academy of Medical Sciences. The Western blot experiment results showed that the combination of leech and bear bile can significantly upregulate the expression levels of p-PI3K and p-Akt proteins, and increase the p-PI3K/PI3K and p-Akt/Akt ratios, which is consistent with the predicted results of network pharmacology. The combination of leech and bear bile has great potential for treating fatty liver disease, and activating the PI3K/Akt pathway may be one of the important mechanisms for reducing lipid deposition in liver cells.

ObjectiveTo observe the clinical efficacy and safety of Modified Guomin Decoction （加味过敏煎， MGD） in patients with mild to moderate atopic dermatitis （AD） of the traditional Chinese medicine （TCM） pattern of heart fire and spleen deficiency， and to explore its possible mechanisms. MethodsIn this randomized， double-blind， placebo-controlled study， 72 patients with mild to moderate AD and the TCM pattern of heart fire and spleen deficiency were randomly divided into a treatment group and a control group， with 36 cases in each group. The treatment group received oral MGD granules combined with topical vitamin E emulsion， while the control group received oral placebo granules combined with topical vitamin E treatment. Both groups were treated twice daily for 4 weeks. Clinical efficacy， TCM syndrome scores， Visual Analogue Scale （VAS） for pruritus， Dermatology Life Quality Index （DLQI） scores， Scoring Atopic Dermatitis （SCORAD） and serum biomarkers， including interleukin-33 （IL-33）， interleukin-1β （IL-1β）， immunoglobulin E （IgE）， and tumor necrosis factor-α （TNF-α） were compared before and after treatment. Safety indexes was also assessed. ResultsThe total clinical effective rates were 77.78% （28/36） in the treatment group and 38.89% （14/36） in the control group， with cure rates of 19.44% （7/36） and 2.78% （1/36）， respectively. The treatment group showed significantly better clinical outcomes compared to the control group （P＜0.05）. The treatment group exhibited significant reductions in total TCM syndrome scores， including erythema， edema， papules， scaling， lichenification， pruritus， irritability， insomnia， abdominal distension， and fatigue scores， as well as reductions in VAS， DLQI， SCORAD， and serum IgE and IL-33 levels （P＜0.05 or P＜0.01）. Compared to the control group， the treatment group had significantly better improvements in all indicators except for insomnia （P＜0.05）. No adverse events occurred in either group. ConclusionMGD is effective and safe in treating mild to moderate AD patients with heart fire and spleen deficiency pattern. It significantly alleviates pruritus， improves TCM syndromes and quality of life， and enhances clinical efficacy， possibly through modulation of immune responses.

Retinal displacement refers to the strong fluorescent lines parallel to the retinal vessels that are detected through autofluorescence examination after rhegmatogenous retinal detachment(RRD)surgery. Actually, even if patients with RRD achieve macroscopic structural reattachment after the operation, the visual function of some patients remains suboptimal. This is associated with the incomplete recovery of retinal function, and retinal displacement is one of the critical influencing factors. This paper reviews the related concepts of retinal displacement and systematically summarizes the incidence of retinal displacement after RRD surgery and its impact on function, the possible mechanisms of retinal displacement, and the influence of various factors on the occurrence of retinal displacement reported in the recent 5 a. It is conducive to enabling surgeons to conduct better design and planning for retinal reattachment surgeries, then achieve higher integrity of retinal function recovery, and enable patients to obtain better postoperative visual function.

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.