No abstract

BackgroundImmunogenic cell death (ICD)-mediated immune response provides a strong rationale to overcome immune evasion in acute lymphoblastic leukemia (ALL). ICD will produce damage-associated molecular patterns (DAMPs) in tumor microenvironment. However, there are few studies on the application of DAMPs-related molecular subtypes in clinically predicting stage III of ALL prognosis. The current study is to identify the DAMPs-associated genes and their molecular subtypes in the stage III of ALL and construct a reliable risk model for prognosis as well as exploring the potential immune-related mechanism.Materials and methodsWe used Target and EBI database for differentially expressed genes (DEGs) analysis of the stage III pediatric ALL samples. Three clusters were identified based on a consistent clustering analysis. By using Cox regression and LASSO analysis, we determined DEGs that attribute to survival benefit. In addition, the Gene Set Enrichment Analysis (GSEA) was performed to identify potential molecular pathways regulated by the DAMPs-related gene signatures. ESTIMATE was employed for evaluating the composition of immune cell populations.ResultsA sum of 146 DAMPs-associated DEGs in ALL were determined and seven transcripts among them were selected to establish a risk model. The DAMPs-associated gene signature significantly contributed to worse prognosis in the high-risk group. We also found that the high-risk group exhibited low immune cell infiltration and high expression of immune checkpoints.ConclusionIn summary, our study showed that the DAMPs-related DEGs in the stage III of children ALL could be used to predict their prognosis. The risk model of DAMPs we established may be more sensitive to immunotherapy prediction.

No abstract

Objective: To determine the most influential data features and to develop machine learning approaches that best predict hospital readmissions among patients with diabetes. Methods: In this retrospective cohort study, we surveyed patient statistics and performed feature analysis to identify the most influential data features associated with readmissions. Classification of all-cause, 30-day readmission outcomes were modeled using logistic regression, artificial neural network, and EasyEnsemble. F1 statistic, sensitivity, and positive predictive value were used to evaluate the model performance. Results: We identified 14 most influential data features (4 numeric features and 10 categorical features) and evaluated 3 machine learning models with numerous sampling methods (oversampling, undersampling, and hybrid techniques). The deep learning model offered no improvement over traditional models (logistic regression and EasyEnsemble) for predicting readmission, whereas the other two algorithms led to much smaller differences between the training and testing datasets. Conclusions: Machine learning approaches to record electronic health data offer a promising method for improving readmission prediction in patients with diabetes. But more work is needed to construct datasets with more clinical variables beyond the standard risk factors and to fine-tune and optimize machine learning models.