Predictive value of machine learning models for lymph node metastasis in gastric cancer: A two-center study

BACKGROUND Gastric cancer is one of the most common malignant tumors in the digestive system,ranking sixth in incidence and fourth in mortality worldwide.Since 42.5%of metastatic lymph nodes in gastric cancer belong to nodule type and peripheral type,the application of imaging diagnosis is restricted.AIM To establish models for predicting the risk of lymph node metastasis in gastric cancer patients using machine learning(ML)algorithms and to evaluate their pre-dictive performance in clinical practice.METHODS Data of a total of 369 patients who underwent radical gastrectomy at the Depart-ment of General Surgery of Affiliated Hospital of Xuzhou Medical University(Xuzhou,China)from March 2016 to November 2019 were collected and retro-spectively analyzed as the training group.In addition,data of 123 patients who underwent radical gastrectomy at the Department of General Surgery of Jining First People’s Hospital(Jining,China)were collected and analyzed as the verifi-cation group.Seven ML models,including decision tree,random forest,support vector machine(SVM),gradient boosting machine,naive Bayes,neural network,and logistic regression,were developed to evaluate the occurrence of lymph node metastasis in patients with gastric cancer.The ML models were established fo-llowing ten cross-validation iterations using the training dataset,and subsequently,each model was assessed using the test dataset.The models’performance was evaluated by comparing the area under the receiver operating characteristic curve of each model.RESULTS Among the seven ML models,except for SVM,the other ones exhibited higher accuracy and reliability,and the influences of various risk factors on the models are intuitive.CONCLUSION The ML models developed exhibit strong predictive capabilities for lymph node metastasis in gastric cancer,which can aid in personalized clinical diagnosis and treatment.

On the energy conservation and critical velocities for the propagation of a "steady-shock" wave in a bar made of cellular material

The propagation of shock waves in a cellular bar is systematically studied in the framework of continuum solids by adopting two idealized material models, viz. the dynamic rigid, perfectly plastic, locking （D-R-PP-L） model and the dynamic rigid, linear hardening plastic, locking （D-R-LHP-L） model, both considering the effects of strain-rate on the material properties. The shock wave speed relevant to these two models is derived. Consider the case of a bar made of one of such material with initial length L 0 and initial velocity v i impinging onto a rigid target. The variations of the stress, strain, particle velocity, specific internal energy across the shock wave and the cease distance of shock wave are all determined analytically. In particular the ＂energy conservation condition＂ and the ＂kinematic existence condition＂ as proposed by Tan et al. （2005） is re-examined, showing that the ＂energy conservation condition＂ and the consequent ＂critical velocity＂, i.e. the shock can only be generated and sustained in R-PP-L bars when the impact velocity is above this critical velocity, is incorrect. Instead, with elastic deformation, strain-hardening and strain-rate sensitivity of the cellular materials being considered, it is appropriate to redefine a first and a second critical impact velocity for the existence and propagation of shock waves in cellular solids. Starting from the basic relations for shock wave propagating in D-R-LHP-L cellular materials, a new method for inversely determining the dynamic stress-strain curve for cellular materials is proposed. By using e.g. a combination of Taylor bar and Hopkinson pressure bar impact experimental technique, the dynamic stress-strain curve of aluminum foam could bedetermined. Finally, it is demonstrated that this new formulation of shock theory in this one-dimensional stress state can be generalized to shocks in a one-dimensional strain state, i.e. for the case of plate impact on cellular materials, by simply making proper replacements of the elastic and plastic constants.

Mechanism of Xuebijing injection on hepatic ischemia-reperfusion injury based on network pharmacology

Objective:Using network pharmacology to predict the main active ingredients,targets and signaling pathways of Xuebijing injection in the treatment of hepatic ischemia-reperfusion injury and explore its potential mechanism of action.Methods:Screen the active ingredients and their targets of Danshen,Honghua,Chishao,Chuanxiong,and Danggui in Xuebijing injection through Traditional Chinese Medicine Systems Pharmacology(TCMSP)database and the Hepatic ischemia-reperfusion injury related targets through Online Mendelian Inheritance in Man(OMIM)and GeneCards Suite(The Human Gene Database)database.And acquire drug-disease intersection targets at the same time.The STRING database was used to construct a protein-protein interaction(PPI)network and topologically screen the central targets.Use the R language online search Bioconductor platform to perform GO function enrichment on the target;Database for Annotation,Visualization and Integrated Discovery(DAVID)database was used to perform KEGG channel enrichment analysis on the target.Use Cytoscape 3.7.2 to construct a"ingredient-target-pathway"network diagram and perform a topology analysis.Results:A total of 115 active ingredients were selected from Xuebijing injection,including Quercetin,Luteolin,Kaempferol,Beta-carotene,and Tanshinone IIa,etc.It corresponds to 217 targets.There are 1057 disease-related targets,and 114 drug-disease common targets.PPI topologically screened out 17 target proteins.Topological analysis of the network graph obtained 15 target genes.Thire intersection contains key targets such as JUN,PPARG,PTGS2,AKT1 and MAPK1.A total of 137 related signaling pathways were obtained by GO enrichment analysis.A total of 8 signaling pathways were obtained through KEGG enrichment(P<0.05,FDR<0.05),among which signaling pathways such as Toll-like receptors,T cell receptors,NOD-like receptors,VEGF,and ErbB played an important role in immune regulation,anti-apoptosis,anti-inflammatory,anti-oxidation,and promoting angiogenesis in the treatment.Conclusion:Xuebijing injection can treat hepatic ischemia-reperfusion injury through multiple components,multiple targets and multiple pathways.

Mechanism of Aidi injection on hepatocellular carcinoma based on network pharmacology

Objective:To explore the active ingredients and potential mechanism of Aidi injection in the treatment of hepatocellular carcinoma by network pharmacology.Methods:Traditional Chinese Medicine Systems Pharmacology Database and Analysis Platform(TCMSP)and Traditional Chinese Medicines Integrated Database(TCMID)were used to screen the active ingredients of four traditional Chinese medicines of Renshen,Huangqi,Ciwujia,and Banmao and corresponding potential targets.Screening of hepatocellular carcinoma-related targets through the Online Mendelian Inheritance in Man(OMIM)and GeneCards Suite(The Human Gene Database)database platforms.The drug and disease targets are merged to obtain the intersection,and the information is imported into Cytoscape 3.7.2 to construct a network diagram of the active ingredients of Aidi injection and related targets of hepatocellular carcinoma,and the topology analysis is performed.A protein-protein interaction(PPI)network was constructed and analyzed using the STRING online analysis platform.Uses the Database for Annotation,Visualization and Integrated Discovery(DAVID)to perform GO function enrichment analysis of targets and enrichment of KEGG pathways analysis.Results:A total of 33 potential active ingredients were screened from Aidi injection for treating hepatocellular carcinoma,including quercetin,kaempferol,beta-sitosterol,isorhamnetin and other important active ingredients.There are 106 potential targets for active ingredient action,6,677 disease-related targets,and 89 drug-disease common targets.Through the network diagram,it was found that the highest degree of target is PTGS1.In the PPI graph,a total of 87 nodes.Among them,the higher degree values include IL6,CASP3,VEGFA,MAPK8,JUN,EGFR,MYC,PTGS2 and FOS.A total of 60 related signal pathways were obtained by GO enrichment analysis.It mainly involves biological processes such as inhibiting abnormal proliferation and differentiation of hepatocellular carcinoma cells,inhibiting angiogenesis of hepatocellular carcinoma,regulating cell cycle and promoting apoptosis.KEGG pathway enrichment analysis screened a total of eight significantly different signal pathways.Among them,P53,VEGF,MAPK,Toll-like receptor,ErbB signaling pathways play an important role in treatment.Conclusion:This study initially revealed the potential mechanism of multi-component,multi-target and multi-pathway treatment of Aidi injection for hepatocellular carcinoma,and provided ideas for the subsequent verification of the molecular mechanism of Aidi injection for hepatocellular carcinoma.

Electroacupuncture alleviates postoperative pain through inhibiting neuroinflammation via stimulator of interferon genes/type-1 interferon pathway

Objective This work explores the impact of electroacupuncture(EA)on acute postoperative pain(APP)and the role of stimulator of interferon genes/type-1 interferon(STING/IFN-1)signaling pathway modulation in the analgesic effect of EA in APP rats.Methods The APP rat model was initiated through abdominal surgery and the animals received two 30 min sessions of EA at bilateral ST36(Zusanli)and SP6(Sanyinjiao)acupoints.Mechanical,thermal and cold sensitivity tests were performed to measure the pain threshold,and electroencephalograms were recorded in the primary somatosensory cortex to identify the effects of EA treatment on APP.Western blotting and immunofluorescence were used to examine the expression and distribution of proteins in the STING/IFN-1 pathway as well as neuroinflammation.A STING inhibitor(C-176)was administered intrathecally to verify its role in EA.Results APP rats displayed mechanical and thermal hypersensitivities compared to the control group(P<0.05).APP significantly reduced the amplitude ofθ,αandγoscillations compared to their baseline values(P<0.05).Interestingly,expression levels of proteins in the STING/IFN-1 pathway were downregulated after inducing APP(P<0.05).Further,APP increased pro-inflammatory factors,including interleukin-6,tumor necrosis factor-αand inducible nitric oxide synthase,and downregulated anti-inflammatory factors,including interleukin-10 and arginase-1(P<0.05).EA effectively attenuated APP-induced painful hypersensitivities(P<0.05)and restored theθ,αandγpower in APP rats(P<0.05).Meanwhile,EA distinctly activated the STING/IFN-1 pathway and mitigated the neuroinflammatory response(P<0.05).Furthermore,STING/IFN-1 was predominantly expressed in isolectin-B4-or calcitonin-gene-related-peptide-labeled dorsal root ganglion neurons and superficial laminae of the spinal dorsal horn.Inhibition of the STING/IFN-1 pathway by intrathecal injection of C-176 weakened the analgesic and anti-inflammatory effects of EA on APP(P<0.05).Conclusion EA can generate robust analgesic and anti-inflammatory effects on APP,and these effects may be linked to activating the STING/IFN-1 pathway,suggesting that STING/IFN-1 may be a target for relieving APP.