BACKGROUND: Postoperative pulmonary complications (PPCs) are major adverse events in neurosurgical patients. This study aimed to develop and validate machine learning models predicting PPCs after neurosurgery. METHODS: PPCs were defined according to the European Perioperative Clinical Outcome standards as occurring within 7 postoperative days. Data of cases meeting inclusion/exclusion criteria were extracted from the anesthesia information management system to create three datasets: The development (data of Huashan Hospital, Fudan University from 2018 to 2020), temporal validation (data of Huashan Hospital, Fudan University in 2021) and external validation (data of other three hospitals in 2023) datasets. Machine learning models of six algorithms were trained using either 35 retrievable and plausible features or the 11 features selected by Lasso regression. Temporal validation was conducted for all models and the 11-feature models were also externally validated. Independent risk factors were identified and feature importance in top models was analyzed. RESULTS: PPCs occurred in 712 of 7533 (9.5%), 258 of 2824 (9.1%), and 207 of 2300 (9.0%) patients in the development, temporal validation and external validation datasets, respectively. During cross-validation training, all models except Bayes demonstrated good discrimination with an area under the receiver operating characteristic curve (AUC) of 0.840. In temporal validation of full-feature models, deep neural network (DNN) performed the best with an AUC of 0.835 (95% confidence interval [CI]: 0.805-0.858) and a Brier score of 0.069, followed by Logistic regression (LR), random forest and XGBoost. The 11-feature models performed comparable to full-feature models with very close but statistically significantly lower AUCs, with the top models of DNN and LR in temporal and external validations. An 11-feature nomogram was drawn based on the LR algorithm and it outperformed the minimally modified Assess respiratory RIsk in Surgical patients in CATalonia (ARISCAT) and Laparoscopic Surgery Video Educational Guidelines (LAS VEGAS) scores with a higher AUC (LR: 0.824, ARISCAT: 0.672, LAS: 0.663). Independent risk factors based on multivariate LR mostly overlapped with Lasso-selected features, but lacked consistency with the important features using the Shapley additive explanation (SHAP) method of the LR model. CONCLUSIONS: The developed models, especially the DNN model and the nomogram, had good discrimination and calibration, and could be used for predicting PPCs in neurosurgical patients. The establishment of machine learning models and the ascertainment of risk factors might assist clinical decision support for improving surgical outcomes. TRIAL REGISTRATION ChiCTR 2100047474; https://www.chictr.org.cn/showproj.html?proj=128279 .
Adult ; Aged ; Female ; Humans ; Male ; Middle Aged ; Algorithms ; Lung Diseases/etiology* ; Machine Learning ; Neurosurgical Procedures/adverse effects* ; Postoperative Complications/diagnosis* ; Risk Factors ; ROC Curve

ObjectiveBased on non-targeted metabolomics， to analyze the regulation of endogenous differential metabolites in serum of type 2 diabetes mellitus（T2DM） rats by Fuling Yunhua granules， and to clarify the metabolic pathways through which this granules exerted its effect on improving T2DM. MethodSeventy SD rats， half male and half female， were randomly divided into the control group， model group， and high， medium， low dose groups of Fuling Yunhua granules（20.70， 10.35， 5.18 g·kg^-1 in raw drug amount） and the positive drug group（pioglitazone hydrochloride tablets， 8.1 mg·kg^-1）. Except for the control group， other groups were fed with high-sugar and high-fat diet combined with intraperitoneal injection of streptozotocin（STZ） to establish a T2DM rat model. After successful modeling， the treatment groups were administered the corresponding drugs by gavage， and the control group and model group were treated with an equal volume of saline by gavage， once/d， for 28 d. Fasting blood glucose（FBG） and glycosylated hemoglobin A1c（GHbA1c） levels were measured in all groups of rats during the administration period， and hematoxylin-eosin（HE） staining was used to observe the pathomorphological changes in the pancreatic tissues of rats at the end of the administration period. The endogenous metabolite levels in rat serum were detected by ultra-performance liquid chromatography-linear ion trap-electrostatic field orbitrap high-resolution mass spectrometry（UPLC-LTQ-Orbitrap MS）， and the data were processed using principal component analysis（PCA） and orthogonal partial least squares-discriminant analysis（OPLS-DA）. Differential metabolites were identified by the Human Metabolome Database（HMDB） and the Kyoto Encyclopedia of Genes and Genomes（KEGG）， and screened for differential metabolites with variable importance in the projection（VIP） value>1， P<0.05， and fold change（FC）<0.6 or FC>1. And the metabolic pathway enrichment analysis of the screened differential metabolites was performed by MetaboAnalyst 5.0， then the screened differential metabolites were diagnosed and evaluated by the receiver operating characteristic（ROC） curves. ResultCompared with the control group， the FBG level of rats in the model group increased significantly（P<0.01）， the GHbA1c content tended to increase， but the difference was not statistically significant， and the pancreatic tissue of rats was obviously damaged， the number of pancreatic islets decreased， and the pancreatic β-cells were obviously reduced， atrophied and enlarged. Compared with the model group， the FBG levels of rats in the high dose group of Fuling Yunhua granules and the positive drug group were significantly reduced after 2 weeks of administration（P<0.05， P<0.01）， the GHbA1c content of rats in the high dose group of Fuling Yunhua granules was significantly reduced（P<0.05）， and the pancreatic tissue lesions of rats in the different dose groups of Fuling Yunhua granules were reduced. The results of non-targeted metabolomics showed that 46 differential metabolites were significantly changed in the model group compared with the blank group. Pathway enrichment analysis found that T2DM mainly affected biological processes including biosynthesis of primary bile acid， D-amino acid metabolism， steroid hormone biosynthesis， and glycerophospholipid metabolism in rats. Compared with the model group， the levels of 8 differential metabolites in the high dose group of Fuling Yunhua granules were significantly adjusted， and the pathway enrichment analysis found that D-amino acid metabolism， retinol metabolism， glycine， serine and threonine metabolism， tryptophan metabolism and other metabolic pathways were mainly involved. ROC curves further analysis revealed that the four characteristic differential markers of 11-cis-retinol， D-piperidinic acid， D-serine， and p-cresol sulfate had high diagnostic value for the treatment of T2DM with Fuling Yunhua granules. ConclusionFuling Yunhua granules can improve the symptoms of T2DM rats by regulating the amino acid metabolic and retinol metabolic pathways through the modulation of endogenous differential metabolites.

Kidney deficiency syndrome is a common clinical syndrome in traditional Chinese medicine （TCM）. With the progress of science and technology， clinical and animal experiments on kidney deficiency syndrome have made remarkable progress. Research on kidney deficiency and the nature of "kidney" involves a large number of physiological and pathological bases， which are closely related to physiological and pathological links in the human body， among which the neuroendocrine-immune network shares the closest relationship. However， there are still many challenges in modern research on kidney deficiency syndrome， such as expert consensus on clinical diagnostic criteria and evaluation indexes and optimization of animal experimental models. In the past decade， a large number of clinical and animal experiments have been reported in the literature on kidney deficiency syndrome， among which the literature focusing on the combination of disease and syndrome is predominant， and most of them focus on kidney Yang deficiency and kidney Yin deficiency， involving the exploration of many pathological mechanisms. Research on the mechanisms related to kidney deficiency syndrome encompasses multiple signaling pathways and various biochemical indicators， including the phosphatidylinositol 3-kinase/protein kinase B/nuclear factor-erythroid 2-relatedfactor-2（PI3K/Akt/Nrf2） signaling pathway， the Toll-like receptor 4/myeloid differentiation factor88/nuclear factor-κB（TLR4/MyD88/NF-κB） signaling pathway， the Janus kinase 2/signal transducer and activator of transcription 3（JAK2/STAT3） signaling pathway， Osteoprotectin/nuclear factor-κB receptor activator ligand/receptor activator of nuclear factor-κB （OPG/RANKL/RANK） signaling pathway. The biochemical indicators cover the cyclic adenosine monophosphate/cyclic guanosine monophosphate （cAMP/cGMP） ratio， Na⁺-K⁺-ATPase activity， Ca²⁺-Mg²⁺-ATPase activity， adrenocorticotropic hormone （ACTH）， polycorticosterone （CORT）， 17-OHCS， and other sex hormone indicators， providing crucial reference values for diagnosing kidney Yang deficiency or kidney Yin deficiency. The literature related to kidney deficiency syndrome over the past decade was collated and excavated， with a view to providing a reference for research on kidney deficiency syndrome.

ObjectiveTo clarify the therapeutic effect of Huashi Baidu prescription on pneumonia in mice caused by influenza A （H1N1） virus and explore its mechanism based on the transcriptome. MethodA mouse influenza viral pneumonia model was built by intranasal infection with influenza A virus， and mice were continuously administered the drug for five days， so as to investigate the general condition， lung index， viral load， pathological morphology of lung tissue， survival time， and prolongation rate of survival time of mice and clarify the therapeutic effect of Huashi Baidu prescription on influenza viral pneumonia. Transcriptome technology was used to detect the differentially expressed genes in the lung tissue of mice in the model group and the normal group， as well as the Huashi Baidu prescription group and the model group， and the potential core target of the Huashi Baidu prescription for the treatment of influenza viral pneumonia was screened. Real-time fluorescence quantitative polymerase chain reaction （Real-time PCR） was used to verify the effect of Huashi Baidu prescription on the mRNA expression level of core target genes. ResultCompared with the normal group， the lung index and viral load in the lung tissue of the model group were significantly increased （P<0.05， P<0.01）. Compared with the model group， the high-dose group of Huashi Baidu prescription significantly prolonged the survival time of mice infected with influenza A virus （P<0.05） and significantly reduced the lung index value of mice （P<0.05） and the viral load of lung tissue. The high-dose， medium-dose， and low-dose groups of Huashi Baidu prescription could significantly reduce lung tissue inflammation， blood stasis， swelling， and other pathological changes in mice （P<0.05， P<0.01）. Transcriptome analysis of lung tissue showed that core genes were mainly enriched in the nuclear transcription factor-κB （NF-κB） signaling pathway， interleukin-17 （IL-17） signaling pathway， cytokine-cytokine receptor interaction， and other pathways after the intervention of Huashi Baidu prescription. TRAF6， NFKBIA， CCL2， CCL7， and CXCL2 were the top five node genes with combined score values. Real-time PCR validation showed that Huashi Baidu prescription significantly downregulated the mRNA expression of key genes TRAF6 and NFKBIA in the NF-κB signaling pathway， as well as chemokines CCL2， CCL7， and CXCL2 （P<0.05， P<0.01）. ConclusionHuashi Baidu prescription has a therapeutic effect on influenza viral pneumonia， possibly by inhibiting the expression of key nodes TRAF6 and NFKBIA in the NF-κB signaling pathway and that of chemokines CCL2， CCL7， and CXCL2， reducing the recruitment of inflammatory cells and viral load， and exerting anti-influenza viral pneumonia effects.

Objective To observe the effect of Xin Guanxin Suhe Drop Pill(XGSDP) on isoprenaline (ISO)-induced acute myocardial ischemia (AMI).Methods Acute myocardial ischemia rats model were established by injection of ISO. Electrocardiogram (ECG) was observed, superoxide dismutase(SOD) activity, and the contents of malondialdehyde(MDA) and nitric oxide (NO) in the myocardial homogenate as well as serum and glutamate oxaloacetate transaminase (GOT) were observed after ISO injection. Results XGSDP could significantly increase NO content and SOD activity, decrease contents of MDA and GOT and reduce the positive rate of ECG( P