[This corrects the article DOI: 10.1016/j.jpha.2023.08.006.].

OBJECTIVE: To construct machine learning prediction model for sepsis-associated encephalopathy (SAE), and analyze the application value of the model on early identification of SAE risk in elderly septic patients. METHODS: Patients aged over 60 years with a primary diagnosis of sepsis admitted to intensive care unit (ICU) from 2008 to 2023 were selected from Medical Information Mart for Intensive Care-IV 2.2 (MIMIC-IV 2.2). Demographic variables, disease severity scores, comorbidities, interventions, laboratory indicators, and hospitalization details were collected. Key factors associated with SAE were identified using univariate Logistic regression analysis. The data were randomly divided into training and validation sets in a 7 : 3 ratio. Multivariable Logistic regression analysis was conducted in the training set and visualized using a nomogram model for prediction of SAE. The discrimination of the model was evaluated in the validation set using the receiver operator characteristic curve (ROC curve), and its calibration was assessed using calibration curve. Furthermore, multiple machine learning algorithms, including multi-layer perceptron (MLP), support vector machine (SVM), naive bayes (NB), gradient boosting machine (GBM), random forest (RF), and extreme gradient boosting (XGB), were constructed in the training set. Their predictive performance was subsequently evaluated on the validation set. Taking the XGB model as an example, the interpretability of the model through the SHapley Additive exPlanations (SHAP) algorithm was enhanced to identify the key predictive factors and their contributions. RESULTS: A total of 2 204 septic patients were finally enrolled, of whom 840 developed SAE (38.1%). A total of 21 variables associated with SAE were screened through univariate Logistic regression analysis. Multivariable Logistic regression analysis showed that endotracheal intubation [odds ratio (OR) = 0.40, 95% confidence interval (95%CI) was 0.19-0.88, P < 0.001], oxygen therapy (OR = 0.76, 95%CI was 0.53-0.95, P = 0.023), tracheotomy (OR = 0.20, 95%CI was 0.07-0.53, P < 0.001), continuous renal replacement therapy (CRRT; OR = 0.32, 95%CI was 0.15-0.70, P < 0.001), cerebrovascular disease (OR = 0.31, 95%CI was 0.16-0.60, P < 0.001), rheumatic disease (OR = 0.44, 95%CI was 0.19-0.99, P < 0.001), male (OR = 0.68, 95%CI was 0.54-0.86, P = 0.001), and maximum anion gap (AG; OR = 0.95, 95%CI was 0.93-0.97, P < 0.001) were associated with an decreased probability of SAE, and age (OR = 1.05, 95%CI was 1.03-1.06, P < 0.001), acute physiology score III (APSIII; OR = 1.02, 95%CI was 1.01-1.02, P < 0.001), Oxford acute severity of illness score (OASIS; OR = 1.04, 95%CI was 1.03-1.06, P < 0.001), and length of hospital stay (OR = 1.01, 95%CI was 1.01-1.02, P < 0.001) were associated with an increased probability of SAE. A nomogram model was constructed based on these variables. In the validation set, ROC curve analysis showed that the model achieved an area under the ROC curve (AUC) of 0.723, and the calibration curve showed good consistency between the predicted probability of the model and the observed probability. Among the machine learning algorithms, including MLP, SVM, NB, GBM, RF, and XGB, the SVM model and RF model demonstrated relatively good predictive performance, with AUC of 0.748 and 0.739, respectively, and the sensitivity was both exceeding 85%. The predictive performance of the XGB model was explained through SHAP analysis, and the results indicated that APSIII score (SHAP value was 0.871), age (SHAP value was 0.521), and OASIS score (SHAP value was 0.443) were important factors affecting the predictive performance of the model. CONCLUSIONS The machine learning-based SAE prediction model exhibits good predictive capability and holds significant application value for the early identification of SAE risk in elderly septic patients.
Humans ; Machine Learning ; Aged ; Sepsis-Associated Encephalopathy ; Sepsis/complications* ; Intensive Care Units ; Logistic Models ; Middle Aged ; Male ; ROC Curve ; Female ; Bayes Theorem ; Nomograms ; Support Vector Machine ; Algorithms

ObjectiveTo study whether Chaihu Longgu Mulitang can inhibit hypothalamic inflammation， mitigate anxiety-like behavior， and alleviate anxiety symptoms by regulating the p38 mitogen-activated protein kinase/nuclear factor-κB （p38 MAPK/NF-κB） signaling pathway in the rat model of generalized anxiety disorder （GAD）. MethodTwelve out of 74 Wistar rats were randomly selected as the blank group， and the remaining rats were subjected to chronic restraint stress for the modeling of GAD. The open field test （OFT） and elevated Porteus maze test （PMT） were conducted 14 days after modeling to detect the anxiety-like behaviors. Sixty successfully modeled rats were selected and randomized into model， low-， medium-， and high-dose （6， 12， and 24 g·kg^-1， respectively） Chaihu Longgu Mulitang， and diazepam （1 mg·kg^-1） groups （n=12） and administrated with corresponding drugs for 14 consecutive days. OFT and PMT were then carried out to examine the anxiety-like behaviors of the rats. The levels of tumor necrosis factor-α （TNF-α）， interleukin-6 （IL-6）， and interleukin-1β （IL-1β） in the hypothalamus and serum of rats were determined by the enzyme-linked immunosorbent assay （ELISA）. Real-time fluorescence quantitative polymerase chain reaction （Real-time PCR）was conducted to determine the mRNA levels of p38 MAPK， NF-κB p65， nuclear factor κB inhibitor α （IκBα）， and ionized calcium binding adaptor molecule 1 （Iba-1）. The protein levels of p38 MAPK， phosphorylated （p）-p38 MAPK， NF-κB p65， p-NF-κB p65， and IκBα in the hypothalamus of rats were determined by Western blot. The expression of Iba-1 in the hypothalamic microglia was detected by immunofluorescence assay. ResultCompared with the blank group， the model group had decreased body weight， scattered dark yellow fur， increased irritability， and preference to hibernation in the corner. In addition， the modeled rats showed increased edge movement distance and time in OFT （P<0.01） and decreased movement distance and time and the number of entries in the open arm in PMT （P<0.01）. The modeling increased the fluorescence intensity of Iba-1 in paraventricular nucleus of hypothalamus （P<0.01）， elevated the levels of IL-1β， IL-6， and TNF-α in the serum and hypothalamus （P<0.01）， up-regulated the protein and mRNA levels of p38 MAPK， NF-κB p65， p-p38 MAPK， p-NF-κB p65， and Iba-1 （P<0.05， P<0.01）， and down-regulated the protein and mRNA levels of IκBα （P<0.01） in the hypothalamus. Compared with the model group， medium- and high-dose Chaihu Longgu Mulitang and diazepam increased the body weight， improved the fur and behaviors， decreased the edge movement distance and time in OFT （P<0.05， P<0.01）， and increased the movement distance and time in the open arm in PMT （P<0.05， P<0.01）. Furthermore， they decreased the fluorescence intensity of Iba-1 in hypothalamic microglia （P<0.05， P<0.01）， lowered the levels of IL-1β， IL-6， and TNF-α in the serum and hypothalamic tissue （P<0.05， P<0.01）， down-regulated the mRNA and protein levels of p38 MAPK， NF-κB p65， p-p38 MAPK， p-NF-κB p65， and Iba-1 （P<0.05， P<0.01）， and up-regulated the mRNA and protein levels of IκBα （P<0.05， P<0.01） in the hypothalamus. ConclusionChaihu Longgu Mulitang can mitigate anxiety-like behaviors and relieve anxiety in GAD rats by inhibiting the p38 MAPK/NF-κB signaling pathway and reducing the activation of microglia and the levels of pro-inflammatory cytokines in the hypothalamus.

Organoids are in vitro three-dimensional(3D)multicellular cultures that are generated through deploying the self-renewal and self-organizing capacities of stem cells.They recapitulate key structural and functional features of corresponding organs or tissues,providing an ideal in vitro model and research platform for the study of developmental biology,regenerative medicine,disease modeling and drug development.The conventional organoid culture system mainly relies on manual operations with lengthy and complicated procedures,which generate organoid cultures of individual variations and batch differences,limiting their translational applications.Therefore,to engineer the organoid culture system by introducing microfluidic chip technology to enhance the throughput and automation level,is of great significance for achieving large-scale,homogeneous,and standardized organoid cultures.This article reviews the current research progress of high-throughput and automated organoid chips and discusses the main limitations and potential challenges for the future study.

BACKGROUND:The distribution of horizontal pressure in adolescent idiopathic scoliosis can be used to evaluate the coronal imbalance.Currently,there are no reports on the characteristics of coronal pressure distribution and its correlation with coronal imbalance. OBJECTIVE:To explore the correlation between coronal pressure variation and coronal imbalance in adolescent idiopathic scoliosis patients. METHODS:A total of 39 adolescent idiopathic scoliosis patients who met the inclusion and exclusion criteria in Lianyungang First People's Hospital from March 2021 to June 2022 were selected as the adolescent idiopathic scoliosis group,and 30 matched healthy volunteers were recruited from the outpatient department as the control group.The horizontal position pressure,folding position pressure,coefficient of variation,and global and trunk pressure variation were measured by the TBED Postural Couch evaluation system.In the adolescent idiopathic scoliosis group,full-length spine radiographs were taken to measure Cobb angle,coronal balance,apical vertebral deviation and Nash-Moe rotation.The characteristics of coronal pressure variation and its correlation with coronal imbalance were analyzed. RESULTS AND CONCLUSION:(1)Compared with the control group,there was no statistically significant difference in the horizontal position pressure and folding position pressure in the adolescent idiopathic scoliosis group(P>0.05),but the global pressure variation and coefficient of variation were significantly increased(all P<0.05).(2)There were differences in both sides of the trunk of the adolescent idiopathic scoliosis group and the control group(P<0.05),and the convex side pressure variation in the adolescent idiopathic scoliosis group was higher than the concave side pressure variation in the left and right side pressure variation in the control group(all P<0.05).(3)The variation of the convex side pressure of the trunk in adolescent idiopathic scoliosis patients was positively correlated with coronal balance intensity(r=0.692,P<0.05),moderately positively correlated with Cobb angle and apical vertebral deviation(r=0.499,0.595,all P<0.05),and weakly correlated with Nash-Moe grade(r=0.377,P<0.05).The variation of the concave side pressure of the trunk was moderately positively correlated with coronal balance(r=0.410,P<0.05),and the rest was weakly correlated or not correlated(P>0.05).(4)These findings indicate that pressure variation may be used as an adjoint assessment tool in patients with mild to moderate scoliosis.

A major impedance to neuronal regeneration after peripheral nerve injury(PNI)is the activation of various programmed cell death mechanisms in the dorsal root ganglion.Ferroptosis is a form of pro-grammed cell death distinguished by imbalance in iron and thiol metabolism,leading to lethal lipid peroxidation.However,the molecular mechanisms of ferroptosis in the context of PNI and nerve regeneration remain unclear.Ferroportin(Fpn),the only known mammalian nonheme iron export protein,plays a pivotal part in inhibiting ferroptosis by maintaining intracellular iron homeostasis.Here,we explored in vitro and in vivo the involvement of Fpn in neuronal ferroptosis.We first delineated that reactive oxygen species at the injury site induces neuronal ferroptosis by increasing intracellular iron via accelerated UBA52-driven ubiquitination and degradation of Fpn,and stimulation of lipid peroxidation.Early administration of the potent arterial vasodilator,hydralazine(HYD),decreases the ubiquitination of Fpn after PNI by binding to UBA52,leading to suppression of neuronal cell death and significant ac-celeration of axon regeneration and motor function recovery.HYD targeting of ferroptosis is a promising strategy for clinical management of PNI.

Objective:Abnormal immune system activation and inflammation are crucial in causing Parkinson's disease.However,we still don't fully understand how certain immune-related genes contribute to the disease's development and progression.This study aims to screen key immune-related gene in Parkinson's disease based on weighted gene co-expression network analysis(WGCNA)and machine learning. Methods:This study downloaded the gene chip data from the Gene Expression Omnibus(GEO)database,and used WGCNA to screen out important gene modules related to Parkinson's disease.Genes from important modules were exported and a Venn diagram of important Parkinson's disease-related genes and immune-related genes was drawn to screen out immune related genes of Parkinson's disease.Gene ontology(GO)analysis and Kyoto Encyclopedia of Genes and Genomes(KEGG)were used to analyze the the functions of immune-related genes and signaling pathways involved.Immune cell infiltration analysis was performed using the CIBERSORT package of R language.Using bioinformatics method and 3 machine learning methods[least absolute shrinkage and selection operator(LASSO)regression,random forest(RF),and support vector machine(SVM)],the immune-related genes of Parkinson's disease were further screened.A Venn diagram of differentially expressed genes screened using the 4 methods was drawn with the intersection gene being hub nodes(hub)gene.The downstream proteins of the Parkinson's disease hub gene was identified through the STRING database and a protein-protein interaction network diagram was drawn. Results:A total of 218 immune genes related to Parkinson's disease were identified,including 45 upregulated genes and 50 downregulated genes.Enrichment analysis showed that the 218 genes were mainly enriched in immune system response to foreign substances and viral infection pathways.The results of immune infiltration analysis showed that the infiltration percentages of CD4+ T cells,NK cells,CD8+ T cells,and B cells were higher in the samples of Parkinson's disease patients,while resting NK cells and resting CD4+ T cells were significantly infiltrated in the samples of Parkinson's disease patients.ANK1 was screened out as the hub gene.The analysis of the protein-protein interaction network showed that the ANK1 translated and expressed 11 proteins which mainly participated in functions such as signal transduction,iron homeostasis regulation,and immune system activation. Conclusion:This study identifies the Parkinson's disease immune-related key gene ANK1 via WGCNA and machine learning methods,suggesting its potential as a candidate therapeutic target for Parkinson's disease.

Objective:To screen the main characteristic variables affecting the incidence of cardiovascular and cerebrovascular diseases,and to construct the Bayesian network model of cardiovascular and cerebrovascular disease incidence risk based on the top 10 characteristic variables,and to provide the reference for predicting the risk of cardiovascular and cerebrovascular disease incidence.Methods:From the UK Biobank Database,315 896 participants and related variables were included.The feature selection was performed by categorical boosting(CatBoost)algorithm,and the participants were randomly divided into training set and test set in the ratio of 7∶3.A Bayesian network model was constructed based on the max-min hill-climbing(MMHC)algorithm.Results:The prevalence of cardiovascular and cerebrovascular diseases in this study was 28.8％.The top 10 variables selected by the CatBoost algorithm were age,body mass index(BMI),low-density lipoprotein cholesterol(LDL-C),total cholesterol(TC),the triglyceride-glucose(TyG)index,family history,apolipoprotein A/B ratio,high-density lipoprotein cholesterol(HDL-C),smoking status,and gender.The area under the receiver operating characteristic(ROC)curve(AUC)for the CatBoost training set model was 0.770,and the model accuracy was 0.764;the AUC of validation set model was 0.759 and the model accuracy was 0.763.The clinical efficacy analysis results showed that the threshold range for the training set was 0.06-0.85 and the threshold range for the validation set was 0.09-0.81.The Bayesian network model analysis results indicated that age,gender,smoking status,family history,BMI,and apolipoprotein A/B ratio were directly related to the incidence of cardiovascular and cerebrovascular diseases and they were the significant risk factors.TyG index,HDL-C,LDL-C,and TC indirectly affect the risk of cardiovascular and cerebrovascular diseases through their impact on BMI and apolipoprotein A/B ratio.Conclusion:Controlling BMI,apolipoprotein A/B ratio,and smoking behavior can reduce the incidence risk of cardiovascular and cerebrovascular diseases.The Bayesian network model can be used to predict the risk of cardiovascular and cerebrovascular disease incidence.