Identification of prognostic signatures in remnant gastric cancer through an interpretable risk model based on machine learning: a multicenter cohort study.

OBJECTIVE: The purpose of this study was to develop an individual survival prediction model based on multiple machine learning (ML) algorithms to predict survival probability for remnant gastric cancer (RGC). METHODS: Clinicopathologic data of 286 patients with RGC undergoing operation (radical resection and palliative resection) from a multi-institution database were enrolled and analyzed retrospectively. These individuals were split into training (80%) and test cohort (20%) by using random allocation. Nine commonly used ML methods were employed to construct survival prediction models. Algorithm performance was estimated by analyzing accuracy, precision, recall, F1-score, area under the receiver operating characteristic curve (AUC), confusion matrices, five-fold cross-validation, decision curve analysis (DCA), and calibration curve. The best model was selected through appropriate verification and validation and was suitably explained by the SHapley Additive exPlanations (SHAP) approach. RESULTS: Compared with the traditional methods, the RGC survival prediction models employing ML exhibited good performance. Except for the decision tree model, all other models performed well, with a mean ROC AUC above 0.7. The DCA findings suggest that the developed models have the potential to enhance clinical decision-making processes, thereby improving patient outcomes. The calibration curve reveals that all models except the decision tree model displayed commendable predictive performance. Through CatBoost-based modeling and SHAP analysis, the five-year survival probability is significantly influenced by several factors: the lymph node ratio (LNR), T stage, tumor size, resection margins, perineural invasion, and distant metastasis. CONCLUSIONS: This study established predictive models for survival probability at five years in RGC patients based on ML algorithms which showed high accuracy and applicative value.

Evaluate prognostic accuracy of SOFA component score for mortality among adults with sepsis by machine learning method.

INTRODUCTION: Sepsis has the characteristics of high incidence, high mortality of ICU patients. Early assessment of disease severity and risk stratification of death in patients with sepsis, and further targeted intervention are very important. The purpose of this study was to develop machine learning models based on sequential organ failure assessment (SOFA) components to early predict in-hospital mortality in ICU patients with sepsis and evaluate model performance. METHODS: Patients admitted to ICU with sepsis diagnosis were extracted from MIMIC-IV database for retrospective analysis, and were randomly divided into training set and test set in accordance with 2:1. Six variables were included in this study, all of which were from the scores of 6 organ systems in SOFA score. The machine learning model was trained in the training set and evaluated in the validation set. Six machine learning methods including linear regression analysis, least absolute shrinkage and selection operator (LASSO), Logistic regression analysis (LR), Gaussian Naive Bayes (GNB) and support vector machines (SVM) were used to construct the death risk prediction models, and the accuracy, area under the receiver operating characteristic curve (AUROC), Decision Curve Analysis (DCA) and K-fold cross-validation were used to evaluate the prediction performance of developed models. RESULT: A total of 23,889 patients with sepsis were enrolled, of whom 3659 died in hospital. Three feature variables including renal system score, central nervous system score and cardio vascular system score were used to establish prediction models. The accuracy of the LR, GNB, SVM were 0.851, 0.844 and 0.862, respectively, which were better than linear regression analysis (0.123) and LASSO (0.130). The AUROCs of LR, GNB and SVM were 0.76, 0.76 and 0.67, respectively. K-fold cross validation showed that the average AUROCs of LR, GNB and SVM were 0.757 ± 0.005, 0.762 ± 0.006, 0.630 ± 0.013, respectively. For the probability threshold of 5-50%, LY and GNB models both showed positive net benefits. CONCLUSION: The two machine learning-based models (LR and GNB models) based on SOFA components can be used to predict in-hospital mortality of septic patients admitted to ICU.

miR-6071 inhibits hepatocellular carcinoma progression via targeting PTPN11.

Hepatocellular carcinoma (HCC) is a deadly malignancy. Liver cancer stem cells (LCSCs) participated in HCC progression and caused failure of chemotherapy. However, the underlying mechanism for the LCSCs regulation was unclear. In this study, we found that miR-6071 expression was decreased in LCSCs. Gain-of-function assays showed that miR-6071 overexpression repressed LCSCs self-renewal and tumorigenesis and inhibited HCC cells proliferation and migration. In mechanism, bioinformatics and luciferase reporter assay demonstrated that miR-6071 targeted 3'UTR of PTPN11 mRNA. Pearson analysis revealed a negative correlation between miR-6071 expression and PTPN11 levels in HCC tissue samples. Further study showed that PTPN11 interference and specific inhibitors IACS-13909 abrogated the discrepancy of self-renewal ability, proliferation, migration and tumorigenicity capacity between miR-6071 overexpression HCC cells and control cells. Moreover, upregulation of miR-6071 sensitized HCC cells to lenvatinib treatment. Clinical cohort analysis revealed that HCC patients with high miR-6071 expression got more survival benefit from postoperative lenvatinib treatment than patients with low miR-6071 levels. In conclusion, our study demonstrated a regulation mechanism of LCSCs, a target against LSCSs, and a biomarker for postoperative lenvatinib treatment.

Identification of early biomarkers of transcriptomics in alveolar macrophage for the prognosis of intubated ARDS patients.

BACKGROUND: Currently, the rate of morbidity and mortality in acute respiratory distress syndrome (ARDS) remains high. One of the potential reasons for the poor and ineffective therapies is the lack of early and credible indicator of risk prediction that would help specific treatment of severely affected ARDS patients. Nevertheless, assessment of the clinical outcomes with transcriptomics of ARDS by alveolar macrophage has not been performed. METHODS: The expression data GSE116560 was obtained from the Gene Expression Omnibus databases (GEO) in NCBI. This dataset consists of 68 BAL samples from 35 subjects that were collected within 48 h of ARDS. Differentially expressed genes (DEGs) of different outcomes were analyzed using R software. The top 10 DEGs that were up- or down-regulated were analyzed using receiver operating characteristic (ROC) analysis. Kaplan-Meier survival analysis within two categories according to cut-off and the value of prediction of the clinical outcomes via DEGs was verified. GO enrichment, KEGG pathway analysis, and protein-protein interaction were also used for functional annotation of key genes. RESULTS: 24,526 genes were obtained, including 235 up-regulated and 292 down-regulated DEGs. The gene ADORA3 was chosen as the most obvious value to predict the outcome according to the ROC and survival analysis. For functional annotation, ADORA3 was significantly augmented in sphingolipid signaling pathway, cGMP-PKG signaling pathway, and neuroactive ligand-receptor interaction. Four genes (ADORA3, GNB1, NTS, and RHO), with 4 nodes and 6 edges, had the highest score in these clusters in the protein-protein interaction network. CONCLUSIONS: Our results show that the prognostic prediction of early biomarkers of transcriptomics as identified in alveolar macrophage in ARDS can be extended for mechanically ventilated critically ill patients. In the long term, generalizing the concept of biomarkers of transcriptomics in alveolar macrophage could add to improving precision-based strategies in the ICU patients and may also lead to identifying improved strategy for critically ill patients.

Identification of transcriptomics biomarkers for the early prediction of the prognosis of septic shock from pneumopathies.

BACKGROUND: Identifying the biological subclasses of septic shock might provide specific targeted therapies for the treatment and prognosis of septic shock. It might be possible to find biological markers for the early prediction of septic shock prognosis. METHODS: The data were obtained from the Gene Expression Omnibus databases (GEO) in NCBI. GO enrichment and KEGG pathway analyses were performed to investigate the functional annotation of up- and downregulated DEGs. ROC curves were drawn, and their areas under the curves (AUCs) were determined to evaluate the predictive value of the key genes. RESULTS: 117 DEGs were obtained, including 36 up- and 81 downregulated DEGs. The AUC for the MME gene was 0.879, as a key gene with the most obvious upregulation in septic shock. The AUC for the THBS1 gene was 0.889, as a key downregulated gene with the most obvious downregulation in septic shock. CONCLUSIONS: The upregulation of MME via the renin-angiotensin system pathway and the downregulation of THBS1 through the PI3K-Akt signaling pathway might have implications for the early prediction of prognosis of septic shock in patients with pneumopathies.

The predictive value of procalcitonin for early detection of infection in elderly type 2 diabetes mellitus.

OBJECTIVE: Aimed to investigate the predictive value of procalcitonin (PCT) in early detection of infections in elderly patients with type 2 diabetes, and to discover the optimum cut-off points of PCT. METHODS: A retrospective study was conducted with type 2 diabetic patients (≥65 years) with lung infection (LI), urinary tract infection (UTI) or skin and soft tissue infection (SSTI). The receiver operating characteristic (ROC) curves of the 3 markers (PCT, WBC count, and CRP) were constructed and compared to assess their accuracies in diagnosing. RESULTS: Among the three different groups with LI, UTI or SSTI, the area under the ROC curve (AUC) of PCT was 0.98 (95% confidence interval (CI): 0.96-0.99, p < 0.05) for the LI group, 0.98 (95% CI: 0.96-0.99, p < 0.05) for the UTI group, and 0.97 (95% CI: 0.94-1.00, p < 0.05) for the SSTI group. The optimum cut-off point of PCT level was 0.73 ng/mL (Sn 89.7%, Sp 97.7%) for the LI group, 1.48 ng/mL (Sn 88.9%, Sp 100%) for the UTI group, and 0.73 ng/mL (Sn 85.7%, Sp 97.7%) for the SSTI group. CONCLUSION: PCT demonstrated the strongest correlation with each of the infection types, indicating significant diagnostic value. Optimum cut-off points of PCT levels in elderly diabetes were higher.

Identification of influence factors in overweight population through an interpretable risk model based on machine learning: a large retrospective cohort.

BACKGROUND: The identification of associated overweight risk factors is crucial to future health risk predictions and behavioral interventions. Several consensus problems remain in machine learning, such as cross-validation, and the resulting model may suffer from overfitting or poor interpretability. METHODS: This study employed nine commonly used machine learning methods to construct overweight risk models. The general community are the target of this study, and a total of 10,905 Chinese subjects from Ningde City in Fujian province, southeast China, participated. The best model was selected through appropriate verification and validation and was suitably explained. RESULTS: The overweight risk models employing machine learning exhibited good performance. It was concluded that CatBoost, which is used in the construction of clinical risk models, may surpass previous machine learning methods. The visual display of the Shapley additive explanation value for the machine model variables accurately represented the influence of each variable in the model. CONCLUSIONS: The construction of an overweight risk model using machine learning may currently be the best approach. Moreover, CatBoost may be the best machine learning method. Furthermore, combining Shapley's additive explanation and machine learning methods can be effective in identifying disease risk factors for prevention and control.

Predicting risk of obesity in overweight adults using interpretable machine learning algorithms.

Objective: To screen for predictive obesity factors in overweight populations using an optimal and interpretable machine learning algorithm. Methods: This cross-sectional study was conducted between June 2011 and January 2012. The participants were randomly selected using a simple random sampling technique. Seven commonly used machine learning methods were employed to construct obesity risk prediction models. A total of 5,236 Chinese participants from Ningde City, Fujian Province, Southeast China, participated in this study. The best model was selected through appropriate verification and validation and suitably explained. Subsequently, a minimal set of significant predictors was identified. The Shapley additive explanation force plot was used to illustrate the model at the individual level. Results: Machine learning models for predicting obesity have demonstrated strong performance, with CatBoost emerging as the most effective in both model validity and net clinical benefit. Specifically, the CatBoost algorithm yielded the highest scores, registering 0.91 in the training set and an impressive 0.83 in the test set. This was further corroborated by the area under the curve (AUC) metrics, where CatBoost achieved 0.95 for the training set and 0.87 for the test set. In a rigorous five-fold cross-validation, the AUC for the CatBoost model ranged between 0.84 and 0.91, with an average AUC of ROC at 0.87 ± 0.022. Key predictors identified within these models included waist circumference, hip circumference, female gender, and systolic blood pressure. Conclusion: CatBoost may be the best machine learning method for prediction. Combining Shapley's additive explanation and machine learning methods can be effective in identifying disease risk factors for prevention and control.

USP5 promotes lipopolysaccharide-induced apoptosis and inflammatory response by stabilizing the TXNIP protein.

BACKGROUND: The role of thioredoxin-interacting protein (TXNIP) in lipopolysaccharide-induced liver injury in mice has been reported, but the underlying mechanisms are poorly understood. METHODS: We overexpressed deubiquitinase in cells overexpressing TXNIP and then detected the level of TXNIP to screen out the deubiquitinase regulating TXNIP; the interaction between TXNIP and deubiquitinase was verified by coimmunoprecipitation. After knockdown of a deubiquitinase and overexpression of TXNIP in Huh7 and HepG2 cells, lipopolysaccharide was used to establish a cellular inflammatory model to explore the role of deubiquitinase and TXNIP in hepatocyte inflammation. RESULTS: In this study, we discovered that ubiquitin-specific protease 5 (USP5) interacts with TXNIP and stabilizes it through deubiquitylation in Huh-7 and HepG2 cells after treatment with lipopolysaccharide. In lipopolysaccharide-treated Huh-7 and HepG2 cells, USP5 knockdown increased cell viability, reduced apoptosis, and decreased the expression of inflammatory factors, including NLRP3, IL-1ß, IL-18, ASC, and procaspase-1. Overexpression of TXNIP reversed the phenotype induced by knockdown USP5. CONCLUSIONS: In summary, USP5 promotes lipopolysaccharide-induced apoptosis and inflammatory response by stabilizing the TXNIP protein.

Development of a Risk Model for Predicting Microalbuminuria in the Chinese Population Using Machine Learning Algorithms.

OBJECTIVE: Microalbuminuria (MAU) occurs due to universal endothelial damage, which is strongly associated with kidney disease, stroke, myocardial infarction, and coronary artery disease. Screening patients at high risk for MAU may aid in the early identification of individuals with an increased risk of cardiovascular events and mortality. Hence, the present study aimed to establish a risk model for MAU by applying machine learning algorithms. METHODS: This cross-sectional study included 3,294 participants ranging in age from 16 to 93 years. R software was used to analyze missing values and to perform multiple imputation. The observed population was divided into a training set and a validation set according to a ratio of 7:3. The first risk model was constructed using the prepared data, following which variables with P <0.1 were extracted to build the second risk model. The second-stage model was then analyzed using a chi-square test, in which a P ≥ 0.05 was considered to indicate no difference in the fit of the models. Variables with P <0.05 in the second-stage model were considered important features related to the prevalence of MAU. A confusion matrix and calibration curve were used to evaluate the validity and reliability of the model. A series of risk prediction scores were established based on machine learning algorithms. RESULTS: Systolic blood pressure (SBP), diastolic blood pressure (DBP), fasting blood glucose (FBG), triglyceride (TG) levels, sex, age, and smoking were identified as predictors of MAU prevalence. Verification using a chi-square test, confusion matrix, and calibration curve indicated that the risk of MAU could be predicted based on the risk score. CONCLUSION: Based on the ability of our machine learning algorithm to establish an effective risk score, we propose that comprehensive assessments of SBP, DBP, FBG, TG, gender, age, and smoking should be included in the screening process for MAU.

An application based on bioinformatics and machine learning for risk prediction of sepsis at first clinical presentation using transcriptomic data.

Background: Linking genotypic changes to phenotypic traits based on machine learning methods has various challenges. In this study, we developed a workflow based on bioinformatics and machine learning methods using transcriptomic data for sepsis obtained at the first clinical presentation for predicting the risk of sepsis. By combining bioinformatics with machine learning methods, we have attempted to overcome current challenges in predicting disease risk using transcriptomic data. Methods: High-throughput sequencing transcriptomic data processing and gene annotation were performed using R software. Machine learning models were constructed, and model performance was evaluated by machine learning methods in Python. The models were visualized and interpreted using the Shapley Additive explanation (SHAP) method. Results: Based on the preset parameters and using recursive feature elimination implemented via machine learning, the top 10 optimal genes were screened for the establishment of the machine learning models. In a comparison of model performance, CatBoost was selected as the optimal model. We explored the significance of each gene in the model and the interaction between each gene through SHAP analysis. Conclusion: The combination of CatBoost and SHAP may serve as the best-performing machine learning model for predicting transcriptomic and sepsis risks. The workflow outlined may provide a new approach and direction in exploring the mechanisms associated with genes and sepsis risk.

Significance of Early Postoperative Arterial Lactic Acid, Inferior Vena Cava Variability, and Central Venous Pressure in Hypovolemic Shock.

INTRODUCTION: Up to one-third of patients admitted to the ICU are in circulatory shock, and early recognition of the condition is vital if subsequent tissue injuries are to be avoided. We would like to know what role the arterial lactic acid, inferior vena cava variability, and CVP (central venous pressure) play in the early stages of shock. METHODS: This is a retrospective study of patients who underwent surgical resuscitation in the Department of Critical Care Medicine. We use the ROC (receiver-operating characteristic) curve to evaluate the significance of each indicator in the diagnosis. For correlation analysis between groups, we first use linear regression for processing and then analysis with correlation. RESULTS: The ROC curve analysis shows that the area under the curve of the lactic acid group was 0.9272, the area under the curve of the inferior vena cava variability group was 0.8652, and the area under the curve of the CVP group was 0.633. Correlation analysis shows that the inferior vena cava variability and arterial lactic acid Pearson's r = 0.2863 and CVP and arterial lactic acid Pearson's r = 0.0729. CONCLUSION: The diagnostic value of arterial lactate is still very high and can still be used as an early warning indicator to help clinicians be alert to the microcirculatory disorders that have emerged quietly. The degree of inferior vena cava variability is linearly related to arterial lactic acid and can also be used as a reference indicator for early evaluation of shock. The diagnostic value of CVP is obviously lower.

Sepsis leads to thyroid impairment and dysfunction in rat model.

Sepsis was a systemic response to a local infection. Apoptosis was observed in the experimental sepsis. In this study, cecal ligation and puncture (CLP)-induced sepsis was established in rats. We found that sepsis decreased thyroid hormone levels, including triiodothyronine (T3), thyroxine (T4), free T3 (fT3), and free T4 (fT4). Besides, we detected the increasing expression level of Caspase-3 and increasing ratio of TUNEL positive cells in the thyroid after sepsis. Furthermore, a series of pathological ultrastructural changes were observed in thyroid follicular epithelial cells by CLP-induced sepsis. This study established a sepsis animal model and provided the cellular and molecular basis for decoding the pathological mechanism in thyroid with the occurrence of sepsis.