To explore the advantages of traditional Chinese medicine （TCM） and integrative TCM-Western medicine approaches in the treatment of diabetic microvascular complications （DMC）， refine key pathophysiological insights and treatment principles， and promote academic innovation and strategic research planning in the prevention and treatment of DMC. The 38th session of the Expert Salon on Diseases Responding Specifically to Traditional Chinese Medicine， hosted by the China Association of Chinese Medicine， was held in Beijing， 2024. Experts in TCM， Western medicine， and interdisciplinary fields convened to conduct a systematic discussion on the pathogenesis， diagnostic and treatment challenges， and mechanism research related to DMC， ultimately forming a consensus on key directions. Four major research recommendations were proposed. The first is addressing clinical bottlenecks in the prevention and control of DMC by optimizing TCM-based evidence evaluation systems. The second is refining TCM core pathogenesis across DMC stages and establishing corresponding "disease-pattern-time" framework. The third is innovating mechanism research strategies to facilitate a shift from holistic regulation to targeted intervention in TCM. The fourth is advancing interdisciplinary collaboration to enhance the role of TCM in new drug development， research prioritization， and guideline formulation. TCM and integrative approaches offer distinct advantages in managing DMC. With a focus on the diseases responding specifically to TCM， strengthening evidence-based support and mechanism interpretation and promoting the integration of clinical care and research innovation will provide strong momentum for the modernization of TCM and the advancement of national health strategies.

ObjectiveTo investigate the mechanism of ferroptosis mediated by long-chain acyl-CoA synthetase 4 （ACSL4） signalling pathway in rats with coronary heart disease with blood stasis syndrome and the intervention effect of Xuefu Zhuyutang. MethodsSPF male SD rats were randomly divided into normal group， sham-operation group， model group， trimetazidine group （5.4 mg·kg^-1）， low-， medium-， and high-dose group （3.51， 7.02，14.04 g·kg^-1） of Xuefu Zhuyutang. The coronary artery left anterior descending ligation method was used to prepare a model of coronary heart disease with blood stasis syndrome， and continuous treatment for 7 d was conducted， while the sham-operation group was only threaded and not ligated. The general macroscopic symptoms of the rats were observed， and indicators such as electrocardiogram， echocardiography， and blood rheology were detected. The pathological morphology of myocardial tissue was observed by hematoxylin-eosin （HE） staining， and the changes in mitochondria in myocardial tissue were observed by transmission electron microscopy. The level of iron deposition in myocardial tissue was observed by Prussian blue staining. The levels of 12-hydroxyeicosatetraenoic acid （12-HETE） and 15-HETE were detected in serum by enzyme-linked immunosorbent assay. A biochemical colourimetric assay was used to detect the levels of Fe²⁺， lipid peroxidation （LPO）， glutathione （GSH）， and T-GSH/glutathione disulfide （GSSG） in myocardial tissue. DCFH-DA fluorescence quantitative assay was employed to detect the levels of reactive oxygen species （ROS）. Western blot and Real-time fluorescence quantitative polymerase chain reaction （Real-time PCR） was adopted to detect the protein and mRNA expressions of glutathione peroxidase 4 （GPX4）， ferritin heavy chain 1 （FTH1）， ACSL4， and ly-sophosphatidylcholine acyltransferase3 （LPCAT3） in myocardial tissue. ResultsCompared with those in the normal group， the rats in the model group were poor in general macroscopic symptoms. The electrocardiogram showed widened QRS wave amplitude and increased voltage， bow-back elevation of the ST segments， elevated T waves， J-point elevation， and accelerated heart rate. Echocardiography showed a significant reduction in left ventricular ejection fraction （LVEF） and left ventricular fraction shortening （LVFS） （P<0.01）. Blood rheology showed that the viscosity of the whole blood （low， medium， and high rate of shear） was significantly increased （P<0.01）. HE staining showed an abnormal structure of myocardial tissue. There was a large area of myocardial necrosis and inflammatory cell infiltration and a large number of connective tissue between myocardial fibers. Transmission electron microscopy showed that the mitochondria were severely atrophy or swelling. The cristae were reduced or even broken， and the matrix was flocculent or even vacuolated. Prussian blue staining showed that there were a large number of iron-containing particles， and the iron deposition was obvious. The content of 12-HETE and 15-HETE in the serum was significantly increased （P<0.01）. The content of Fe²⁺， LPO， and ROS in myocardial tissue was significantly increased （P<0.01）. The content of GSH was significantly decreased （P<0.01）， and T-GSH/GSSG was decreased （P<0.01）. The protein and mRNA expressions of GPX4 and FTH1 in myocardial tissue were both significantly decreased （P<0.05， P<0.01）， while those of ACSL4 and LPCAT3 increased significantly （P<0.01）. Compared with the model group， the general macroscopic symptoms and electrocardiogram results of rats in low-， medium- and high-dose groups of Xuefu Zhuyutang were alleviated， and the differences in LVEF/LVFS ratios were all significantly increased （P<0.05， P<0.01）. The differences in whole-blood viscosity （low， medium， and high rate of shear） were all significantly decreased （P<0.01）. The results of HE staining and transmission electron microscopy showed that the morphology， structure， and mitochondria of cardiomyocytes were improved. The content of 12-HETE and 15-HETE in serum was reduced to different degrees in low-， medium-， and high-dose groups of Xuefu Zhuyutang （P<0.05， P<0.01）. The content of Fe²⁺， LPO， and ROS was significantly reduced in the medium- and high-dose groups of Xuefu Zhuyutang （P<0.05， P<0.01）， and the content of GSH and T-GSH/GSSG was significantly increased （P<0.05， P<0.01）. The protein and mRNA expressions of GPX4 and FTH1 were significantly increased to varying degrees in the medium- and high-dose groups of Xuefu Zhuyutang （P<0.05， P<0.01）， and ACSL4 and LPCAT3 were decreased to different degrees in the low-， medium-， and high-dose groups of Xuefu Zhuyutang （P<0.05， P<0.01）. ConclusionXuefu Zhuyutang can regulate iron metabolism and anti-lipid oxidation reaction to mediate ferroptosis through the ACSL4 signalling pathway， thus exerting a protective effect on rats with coronary heart disease with blood stasis syndrome.

The accumulation of uremic toxins such as urea nitrogen, blood creatinine, and uric acid of patients with renal failure in vivo would lead to aggravated kidney damage. In this study, coated aldehyde oxy-starch (CAO) was used as an adsorbent to investigate its in vitro adsorption performance on renal failure indexes for urea, indoxyl sulfate (INS), monomethylamine (MMA), dimethylamine (DMA), uric acid (UA), and creatinine (Cr). The effects of variables such as pH, temperature, concentration, dosage, and time on the adsorption capacity of CAO were systematically investigated, employing analytical techniques of high-performance liquid chromatography (HPLC) and gas chromatography (GC). The results revealed that CAO exhibited a strong adsorption capacity for urea, INS, and MMA, alongside a moderate adsorption capacity for DMA, UA, and Cr. The adsorption kinetics and thermodynamic studies indicated that the adsorption of urea and UA by CAO were fitted in pseudo-first-order kinetics, and the adsorption isotherm aligned with the Freundlich adsorption model. The enthalpy change ΔH of urea in adsorption was in the range of 40 to 60 kJ·mol^-1, which demonstrated the presence of strong adsorption force due to the interactions of coordinating groups. The ΔH of UA was greater than 80 kJ·mol^-1, indicating the generation of chemical bonds during the adsorption. Both of them, Gibbs free energy ΔG was less than 0, within the range of -20 to 0 kJ·mol^-1, suggested that the adsorption of urea and UA by CAO occurred spontaneously as physical adsorption process. The adsorption entropy ΔS of urea and UA was > 0, which indicated an increase in entropy throughout the adsorption. The infrared spectroscopygram showed the formation of a chemical bond, specifically the imine bond, following the adsorption of urea by CAO, thereby indicating a chemical reaction during the adsorption. This study elucidates the adsorption mechanism of CAO on various indexes of renal failure, providing a scientific basis for its clinical usage.

ObjectiveTo investigate the influence of histone deacetylase 3 (HDAC3) on the occurrence, development of psoriasis-like inflammation in mice, and the relative immune mechanisms. MethodsHealthy C57BL/6 mice aged 6-8 weeks were selected and randomly divided into 3 groups: control group (Control), psoriasis model group (IMQ), and HDAC3 inhibitor RGFP966-treated psoriasis model group (IMQ+RGFP966). One day prior to the experiment, the back hair of the mice was shaved. After a one-day stabilization period, the mice in Control group was treated with an equal amount of vaseline, while the mice in IMQ group was treated with imiquimod (62.5 mg/d) applied topically on the back to establish a psoriasis-like inflammation model. The mice in IMQ+RGFP966 group received intervention with a high dose of the HDAC3-selective inhibitor RGFP966 (30 mg/kg) based on the psoriasis-like model. All groups were treated continuously for 5 d, during which psoriasis-like inflammation symptoms (scaling, erythema, skin thickness), body weight, and mental status were observed and recorded, with photographs taken for documentation. After euthanasia, hematoxylin-eosin (HE) staining was used to assess the effect of RGFP966 on the skin tissue structure of the mice, and skin thickness was measured. The mRNA and protein expression levels of HDAC3 in skin tissues were detected using reverse transcription real-time quantitative polymerase chain reaction (RT-qPCR) and Western blot (WB), respectively. Flow cytometry was employed to analyze neutrophils in peripheral blood and lymph nodes, CD4⁺ T lymphocytes, CD8⁺ T lymphocytes in peripheral blood, and IL-17A secretion by peripheral blood CD4⁺ T lymphocytes. Additionally, spleen CD4⁺ T lymphocyte expression of HDAC3, CCR6, CCR8, and IL-17A secretion levels were analyzed. Immunohistochemistry was used to detect the localization and expression levels of HDAC3, IL-17A, and IL-10 in skin tissues. ResultsCompared with the Control group, the IMQ group exhibited significant psoriasis-like inflammation, characterized by erythema, scaling, and skin wrinkling. Compared with the IMQ group, RGFP966 exacerbated psoriasis-like inflammatory symptoms, leading to increased hyperkeratosis. The psoriasis area and severity index (PASI) skin symptom scores were higher in the IMQ group than those in the Control group, and the scores were further elevated in the IMQ+RGFP966 group compared to the IMQ group. Skin thickness measurements showed a trend of IMQ+RGFP966>IMQ>Control. The numbers of neutrophils in the blood and lymph nodes increased sequentially in the Control, IMQ, and IMQ+RGFP966 groups, with a similar trend observed for CD4⁺ and CD8⁺ T lymphocytes in the blood. In skin tissues, compared with the Control group, the mRNA and protein levels of HDAC3 decreased in the IMQ group, but RGFP966 did not further reduce these expressions. HDAC3 was primarily located in the nucleus. Compared with the Control group, the nuclear HDAC3 content decreased in the skin tissues of the IMQ group, and RGFP966 further reduced nuclear HDAC3. Compared with the Control and IMQ groups, RGFP966 treatment decreased HDAC3 expression in splenic CD4⁺ and CD8⁺ T cells. RGFP966 treatment increased the expression of CCR6 and CCR8 in splenic CD4⁺ T cells and enhanced IL-17A secretion by peripheral blood and splenic CD4⁺ T lymphocytes. Additionally, compared with the IMQ group, RGFP966 reduced IL-10 protein levels and upregulated IL-17A expression in skin tissues. ConclusionRGFP966 exacerbates psoriatic-like inflammatory responses by inhibiting HDAC3, increasing the secretion of the cytokine IL-17A, and upregulating the expression of chemokines CCR8 and CCR6.

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 construct three-class （insufficient， normal， excessive） and two-class （insufficient， normal） models for predicting plasma concentration of valproic acid （VPA）， and compare the performance of these two models， with the aim of providing a reference for formulating clinical medication strategies. METHODS The clinical data of 480 patients who received VPA treatment and underwent blood concentration test at the Xi’an International Medical Center Hospital were collected from November 2022 to September 2024 （a total of 695 sets of data）. In this study， predictive models were constructed for target variables of three-class and two-class models. Feature ranking and selection were carried out using XGBoost scores. Twelve different machine learning algorithms were used for training and validation， and the performance of the models was evaluated using three indexes: accuracy， F1 score， and the area under the working characteristic curve of the subject （AUC）. RESULTS XGBoost feature importance scores revealed that in the three-class model， the importance ranking of kidney disease and electrolyte disorders was higher. However， in the two-class model， the importance ranking of these features significantly decreased， suggesting a close association with the excessive blood concentration of VPA. In the three-class model， Random Forest method performed best， with F1 score of 0.704 0 and AUC of 0.519 3 on the test set； while in the two-class model， CatBoost method performed optimally， with F1 score of 0.785 7 and AUC of 0.819 5 on the test set. CONCLUSIONS The constructed three-class model has the ability to predict excessive VPA blood concentration， but its prediction and model generalization abilities are poor； the constructed two-class model can only perform classification prediction for insufficient and normal blood concentration cases， but its model performance is stronger.

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.