MagicalRsq: Machine-learning-based genotype imputation quality calibration.

Whole-genome sequencing (WGS) is the gold standard for fully characterizing genetic variation but is still prohibitively expensive for large samples. To reduce costs, many studies sequence only a subset of individuals or genomic regions, and genotype imputation is used to infer genotypes for the remaining individuals or regions without sequencing data. However, not all variants can be well imputed, and the current state-of-the-art imputation quality metric, denoted as standard Rsq, is poorly calibrated for lower-frequency variants. Here, we propose MagicalRsq, a machine-learning-based method that integrates variant-level imputation and population genetics statistics, to provide a better calibrated imputation quality metric. Leveraging WGS data from the Cystic Fibrosis Genome Project (CFGP), and whole-exome sequence data from UK BioBank (UKB), we performed comprehensive experiments to evaluate the performance of MagicalRsq compared to standard Rsq for partially sequenced studies. We found that MagicalRsq aligns better with true R2 than standard Rsq in almost every situation evaluated, for both European and African ancestry samples. For example, when applying models trained from 1,992 CFGP sequenced samples to an independent 3,103 samples with no sequencing but TOPMed imputation from array genotypes, MagicalRsq, compared to standard Rsq, achieved net gains of 1.4 million rare, 117k low-frequency, and 18k common variants, where net gains were gained numbers of correctly distinguished variants by MagicalRsq over standard Rsq. MagicalRsq can serve as an improved post-imputation quality metric and will benefit downstream analysis by better distinguishing well-imputed variants from those poorly imputed. MagicalRsq is freely available on GitHub.

CrossFuse-XGBoost: accurate prediction of the maximum recommended daily dose through multi-feature fusion, cross-validation screening and extreme gradient boosting.

In the drug development process, approximately 30% of failures are attributed to drug safety issues. In particular, the first-in-human (FIH) trial of a new drug represents one of the highest safety risks, and initial dose selection is crucial for ensuring safety in clinical trials. With traditional dose estimation methods, which extrapolate data from animals to humans, catastrophic events have occurred during Phase I clinical trials due to interspecies differences in compound sensitivity and unknown molecular mechanisms. To address this issue, this study proposes a CrossFuse-extreme gradient boosting (XGBoost) method that can directly predict the maximum recommended daily dose of a compound based on existing human research data, providing a reference for FIH dose selection. This method not only integrates multiple features, including molecular representations, physicochemical properties and compound-protein interactions, but also improves feature selection based on cross-validation. The results demonstrate that the CrossFuse-XGBoost method not only improves prediction accuracy compared to that of existing local weighted methods [k-nearest neighbor (k-NN) and variable k-NN (v-NN)] but also solves the low prediction coverage issue of v-NN, achieving full coverage of the external validation set and enabling more reliable predictions. Furthermore, this study offers a high level of interpretability by identifying the importance of different features in model construction. The 241 features with the most significant impact on the maximum recommended daily dose were selected, providing references for optimizing the structure of new compounds and guiding experimental research. The datasets and source code are freely available at https://github.com/cqmu-lq/CrossFuse-XGBoost.

KbhbXG: A Machine learning architecture based on XGBoost for prediction of lysine ß-Hydroxybutyrylation (Kbhb) modification sites.

Lysine ß-hydroxybutyrylation is an important post-translational modification (PTM) involved in various physiological and biological processes. In this research, we introduce a novel predictor KbhbXG, which utilizes XGBoost to identify ß-hydroxybutyrylation modification sites based on protein sequence information. The traditional experimental methods employed for the identification of ß-hydroxybutyrylated sites using proteomic techniques are both costly and time-consuming. Thus, the development of computational methods and predictors can play a crucial role in facilitating the rapid identification of ß-hydroxybutyrylation sites. Our proposed KbhbXG model first utilizes machine learning algorithm XGBoost to predict ß-hydroxybutyrylation modification sites. On the independent test set, KbhbXG achieves an accuracy of 0.7457, specificity of 0.7771, and an impressive area under the curve (AUC) score of 0.8172. The high AUC score achieved by our method demonstrates its potential for effectively identifying novel ß-hydroxybutyrylation sites, thereby facilitating further research and exploration of the ß-hydroxybutyrylation process. Also, functional analyses have revealed that different organisms preferentially engage in distinct biological processes and pathways, which can provide valuable insights for understanding the mechanism of ß-hydroxybutyrylation and guide experimental verification. To promote transparency and reproducibility, we have made both the codes and dataset of KbhbXG publicly available. Researchers interested in utilizing our proposed model can access these resources at https://github.com/Lab-Xu/KbhbXG.

Multi-label classification with XGBoost for metabolic pathway prediction.

BACKGROUND: Metabolic pathway prediction is one possible approach to address the problem in system biology of reconstructing an organism's metabolic network from its genome sequence. Recently there have been developments in machine learning-based pathway prediction methods that conclude that machine learning-based approaches are similar in performance to the most used method, PathoLogic which is a rule-based method. One issue is that previous studies evaluated PathoLogic without taxonomic pruning which decreases its performance. RESULTS: In this study, we update the evaluation results from previous studies to demonstrate that PathoLogic with taxonomic pruning outperforms previous machine learning-based approaches and that further improvements in performance need to be made for them to be competitive. Furthermore, we introduce mlXGPR, a XGBoost-based metabolic pathway prediction method based on the multi-label classification pathway prediction framework introduced from mlLGPR. We also improve on this multi-label framework by utilizing correlations between labels using classifier chains. We propose a ranking method that determines the order of the chain so that lower performing classifiers are placed later in the chain to utilize the correlations between labels more. We evaluate mlXGPR with and without classifier chains on single-organism and multi-organism benchmarks. Our results indicate that mlXGPR outperform other previous pathway prediction methods including PathoLogic with taxonomic pruning in terms of hamming loss, precision and F1 score on single organism benchmarks. CONCLUSIONS: The results from our study indicate that the performance of machine learning-based pathway prediction methods can be substantially improved and can even outperform PathoLogic with taxonomic pruning.

Lipidomic-Based Algorithms Can Enhance Prediction of Obstructive Coronary Artery Disease.

Lipidomics emerges as a promising research field with the potential to help in personalized risk stratification and improve our understanding on the functional role of individual lipid species in the metabolic perturbations occurring in coronary artery disease (CAD). This study aimed to utilize a machine learning approach to provide a lipid panel able to identify patients with obstructive CAD. In this posthoc analysis of the prospective CorLipid trial, we investigated the lipid profiles of 146 patients with suspected CAD, divided into two categories based on the existence of obstructive CAD. In total, 517 lipid species were identified, from which 288 lipid species were finally quantified, including glycerophospholipids, glycerolipids, and sphingolipids. Univariate and multivariate statistical analyses have shown significant discrimination between the serum lipidomes of patients with obstructive CAD. Finally, the XGBoost algorithm identified a panel of 17 serum biomarkers (5 sphingolipids, 7 glycerophospholipids, a triacylglycerol, galectin-3, glucose, LDL, and LDH) as totally sensitive (100% sensitivity, 62.1% specificity, 100% negative predictive value) for the prediction of obstructive CAD. Our findings shed light on dysregulated lipid metabolism's role in CAD, validating existing evidence and suggesting promise for novel therapies and improved risk stratification.

Identification of crosstalk genes relating to ECM-receptor interaction genes in MASH and DN using bioinformatics and machine learning.

This study aimed to identify genes shared by metabolic dysfunction-associated fatty liver disease (MASH) and diabetic nephropathy (DN) and the effect of extracellular matrix (ECM) receptor interaction genes on them. Datasets with MASH and DN were downloaded from the Gene Expression Omnibus (GEO) database. Pearson's coefficients assessed the correlation between ECM-receptor interaction genes and cross talk genes. The coexpression network of co-expression pairs (CP) genes was integrated with its protein-protein interaction (PPI) network, and machine learning was employed to identify essential disease-representing genes. Finally, immuno-penetration analysis was performed on the MASH and DN gene datasets using the CIBERSORT algorithm to evaluate the plausibility of these genes in diseases. We found 19 key CP genes. Fos proto-oncogene (FOS), belonging to the IL-17 signalling pathway, showed greater centrality PPI network; Hyaluronan Mediated Motility Receptor (HMMR), belonging to ECM-receptor interaction genes, showed most critical in the co-expression network map of 19 CP genes; Forkhead Box C1 (FOXC1), like FOS, showed a high ability to predict disease in XGBoost analysis. Further immune infiltration showed a clear positive correlation between FOS/FOXC1 and mast cells that secrete IL-17 during inflammation. Combining the results of previous studies, we suggest a FOS/FOXC1/HMMR regulatory axis in MASH and DN may be associated with mast cells in the acting IL-17 signalling pathway. Extracellular HMMR may regulate the IL-17 pathway represented by FOS through the Mitogen-Activated Protein Kinase 1 (ERK) or PI3K-Akt-mTOR pathway. HMMR may serve as a signalling carrier between MASH and DN and could be targeted for therapeutic development.

A machine learning approach to predict daptomycin exposure from two concentrations based on Monte Carlo simulations.

Daptomycin is a concentration-dependent lipopeptide antibiotic for which exposure/effect relationships have been shown. Machine learning (ML) algorithms, developed to predict the individual exposure to drugs, have shown very good performances in comparison to maximum a posteriori Bayesian estimation (MAP-BE). The aim of this work was to predict the area under the blood concentration curve (AUC) of daptomycin from two samples and a few covariates using XGBoost ML algorithm trained on Monte Carlo simulations. Five thousand one hundred fifty patients were simulated from two literature population pharmacokinetics models. Data from the first model were split into a training set (75%) and a testing set (25%). Four ML algorithms were built to learn AUC based on daptomycin blood concentration samples at pre-dose and 1 h post-dose. The XGBoost model (best ML algorithm) with the lowest root mean square error (RMSE) in a 10-fold cross-validation experiment was evaluated in both the test set and the simulations from the second population pharmacokinetic model (validation). The ML model based on the two concentrations, the differences between these concentrations, and five other covariates (sex, weight, daptomycin dose, creatinine clearance, and body temperature) yielded very good AUC estimation in the test (relative bias/RMSE = 0.43/7.69%) and validation sets (relative bias/RMSE = 4.61/6.63%). The XGBoost ML model developed allowed accurate estimation of daptomycin AUC using C0, C1h, and a few covariates and could be used for exposure estimation and dose adjustment. This ML approach can facilitate the conduct of future therapeutic drug monitoring (TDM) studies.

iRice-MS: An integrated XGBoost model for detecting multitype post-translational modification sites in rice.

Post-translational modification (PTM) refers to the covalent and enzymatic modification of proteins after protein biosynthesis, which orchestrates a variety of biological processes. Detecting PTM sites in proteome scale is one of the key steps to in-depth understanding their regulation mechanisms. In this study, we presented an integrated method based on eXtreme Gradient Boosting (XGBoost), called iRice-MS, to identify 2-hydroxyisobutyrylation, crotonylation, malonylation, ubiquitination, succinylation and acetylation in rice. For each PTM-specific model, we adopted eight feature encoding schemes, including sequence-based features, physicochemical property-based features and spatial mapping information-based features. The optimal feature set was identified from each encoding, and their respective models were established. Extensive experimental results show that iRice-MS always display excellent performance on 5-fold cross-validation and independent dataset test. In addition, our novel approach provides the superiority to other existing tools in terms of AUC value. Based on the proposed model, a web server named iRice-MS was established and is freely accessible at http://lin-group.cn/server/iRice-MS.

Development of a machine learning-based model to predict prognosis of alpha-fetoprotein-positive hepatocellular carcinoma.

BACKGROUND: Patients with alpha-fetoprotein (AFP)-positive hepatocellular carcinoma (HCC) have aggressive biological behavior and poor prognosis. Therefore, survival time is one of the greatest concerns for patients with AFP-positive HCC. This study aimed to demonstrate the utilization of six machine learning (ML)-based prognostic models to predict overall survival of patients with AFP-positive HCC. METHODS: Data on patients with AFP-positive HCC were extracted from the Surveillance, Epidemiology, and End Results database. Six ML algorithms (extreme gradient boosting [XGBoost], logistic regression [LR], support vector machine [SVM], random forest [RF], K-nearest neighbor [KNN], and decision tree [ID3]) were used to develop the prognostic models of patients with AFP-positive HCC at one year, three years, and five years. Area under the receiver operating characteristic curve (AUC), confusion matrix, calibration curves, and decision curve analysis (DCA) were used to evaluate the model. RESULTS: A total of 2,038 patients with AFP-positive HCC were included for analysis. The 1-, 3-, and 5-year overall survival rates were 60.7%, 28.9%, and 14.3%, respectively. Seventeen features regarding demographics and clinicopathology were included in six ML algorithms to generate a prognostic model. The XGBoost model showed the best performance in predicting survival at 1-year (train set: AUC = 0.771; test set: AUC = 0.782), 3-year (train set: AUC = 0.763; test set: AUC = 0.749) and 5-year (train set: AUC = 0.807; test set: AUC = 0.740). Furthermore, for 1-, 3-, and 5-year survival prediction, the accuracy in the training and test sets was 0.709 and 0.726, 0.721 and 0.726, and 0.778 and 0.784 for the XGBoost model, respectively. Calibration curves and DCA exhibited good predictive performance as well. CONCLUSIONS: The XGBoost model exhibited good predictive performance, which may provide physicians with an effective tool for early medical intervention and improve the survival of patients.

Characterizing the antigenic evolution of pandemic influenza A (H1N1) pdm09 from 2009 to 2023.

The H1N1pdm09 virus has been a persistent threat to public health since the 2009 pandemic. Particularly, since the relaxation of COVID-19 pandemic mitigation measures, the influenza virus and SARS-CoV-2 have been concurrently prevalent worldwide. To determine the antigenic evolution pattern of H1N1pdm09 and develop preventive countermeasures, we collected influenza sequence data and immunological data to establish a new antigenic evolution analysis framework. A machine learning model (XGBoost, accuracy = 0.86, area under the receiver operating characteristic curve = 0.89) was constructed using epitopes, physicochemical properties, receptor binding sites, and glycosylation sites as features to predict the antigenic similarity relationships between influenza strains. An antigenic correlation network was constructed, and the Markov clustering algorithm was used to identify antigenic clusters. Subsequently, the antigenic evolution pattern of H1N1pdm09 was analyzed at the global and regional scales across three continents. We found that H1N1pdm09 evolved into around five antigenic clusters between 2009 and 2023 and that their antigenic evolution trajectories were characterized by cocirculation of multiple clusters, low-level persistence of former dominant clusters, and local heterogeneity of cluster circulations. Furthermore, compared with the seasonal H1N1 virus, the potential cluster-transition determining sites of H1N1pdm09 were restricted to epitopes Sa and Sb. This study demonstrated the effectiveness of machine learning methods for characterizing antigenic evolution of viruses, developed a specific model to rapidly identify H1N1pdm09 antigenic variants, and elucidated their evolutionary patterns. Our findings may provide valuable support for the implementation of effective surveillance strategies and targeted prevention efforts to mitigate the impact of H1N1pdm09.

Proteomic analysis of human aqueous humor from fuchs uveitis syndrome.

Fuchs uveitis syndrome (FUS) is a commonly misdiagnosed uveitis syndrome often presenting as an asymptomatic mild inflammatory condition until complications arise. The diagnosis of this disease remains clinical because of the lack of specific laboratory tests. The aqueous humor (AH) is a complex fluid containing nutrients and metabolic wastes from the eye. Changes in the AH protein provide important information for diagnosing intraocular diseases. This study aimed to analyze the proteomic profile of AH in individuals diagnosed with FUS and to identify potential biomarkers of the disease. We used liquid chromatography-tandem mass spectrometry-based proteomic methods to evaluate the AH protein profiles of all 37 samples, comprising 15 patients with FUS, six patients with Posner-Schlossman syndrome (PSS), and 16 patients with age-related cataract. A total of 538 proteins were identified from a comprehensive spectral library of 634 proteins. Subsequent differential expression analysis, enrichment analysis, and construction of key sub-networks revealed that the inflammatory response, complement activation and hypoxia might be crucial in mediating the process of FUS. The hypoxia inducible factor-1 may serve as a key regulator and therapeutic target. Additionally, the innate and adaptive immune responses are considered dominant in the patients with FUS. A diagnostic model was constructed using machine-learning algorithm to classify FUS, PSS, and normal controls. Two proteins, complement C1q subcomponent subunit B and secretogranin-1, were found to have the highest scores by the Extreme Gradient Boosting, suggesting their potential utility as a biomarker panel. Furthermore, these two proteins as biomarkers were validated in a cohort of 18 patients using high resolution multiple reaction monitoring assays. Therefore, this study contributes to advancing of the current knowledge of FUS pathogenesis and promotes the development of effective diagnostic strategies.

LightGBM outperforms other machine learning techniques in predicting graft failure after liver transplantation: Creation of a predictive model through large-scale analysis.

BACKGROUND: The incidence of graft failure following liver transplantation (LTx) is consistent. While traditional risk scores for LTx have limited accuracy, the potential of machine learning (ML) in this area remains uncertain, despite its promise in other transplant domains. This study aims to determine ML's predictive limitations in LTx by replicating methods used in previous heart transplant research. METHODS: This study utilized the UNOS STAR database, selecting 64,384 adult patients who underwent LTx between 2010 and 2020. Gradient boosting models (XGBoost and LightGBM) were used to predict 14, 30, and 90-day graft failure compared to conventional logistic regression model. Models were evaluated using both shuffled and rolling cross-validation (CV) methodologies. Model performance was assessed using the AUC across validation iterations. RESULTS: In a study comparing predictive models for 14-day, 30-day and 90-day graft survival, LightGBM consistently outperformed other models, achieving the highest AUC of.740,.722, and.700 in shuffled CV methods. However, in rolling CV the accuracy of the model declined across every ML algorithm. The analysis revealed influential factors for graft survival prediction across all models, including total bilirubin, medical condition, recipient age, and donor AST, among others. Several features like donor age and recipient diabetes history were important in two out of three models. CONCLUSIONS: LightGBM enhances short-term graft survival predictions post-LTx. However, due to changing medical practices and selection criteria, continuous model evaluation is essential. Future studies should focus on temporal variations, clinical implications, and ensure model transparency for broader medical utility.

Country-specific determinants for COVID-19 case fatality rate and response strategies from a global perspective: an interpretable machine learning framework.

BACKGROUND: There are significant geographic inequities in COVID-19 case fatality rates (CFRs), and comprehensive understanding its country-level determinants in a global perspective is necessary. This study aims to quantify the country-specific risk of COVID-19 CFR and propose tailored response strategies, including vaccination strategies, in 156 countries. METHODS: Cross-temporal and cross-country variations in COVID-19 CFR was identified using extreme gradient boosting (XGBoost) including 35 factors from seven dimensions in 156 countries from 28 January, 2020 to 31 January, 2022. SHapley Additive exPlanations (SHAP) was used to further clarify the clustering of countries by the key factors driving CFR and the effect of concurrent risk factors for each country. Increases in vaccination rates was simulated to illustrate the reduction of CFR in different classes of countries. FINDINGS: Overall COVID-19 CFRs varied across countries from 28 Jan 2020 to 31 Jan 31 2022, ranging from 68 to 6373 per 100,000 population. During the COVID-19 pandemic, the determinants of CFRs first changed from health conditions to universal health coverage, and then to a multifactorial mixed effect dominated by vaccination. In the Omicron period, countries were divided into five classes according to risk determinants. Low vaccination-driven class (70 countries) mainly distributed in sub-Saharan Africa and Latin America, and include the majority of low-income countries (95.7%) with many concurrent risk factors. Aging-driven class (26 countries) mainly distributed in high-income European countries. High disease burden-driven class (32 countries) mainly distributed in Asia and North America. Low GDP-driven class (14 countries) are scattered across continents. Simulating a 5% increase in vaccination rate resulted in CFR reductions of 31.2% and 15.0% for the low vaccination-driven class and the high disease burden-driven class, respectively, with greater CFR reductions for countries with high overall risk (SHAP value > 0.1), but only 3.1% for the ageing-driven class. CONCLUSIONS: Evidence from this study suggests that geographic inequities in COVID-19 CFR is jointly determined by key and concurrent risks, and achieving a decreasing COVID-19 CFR requires more than increasing vaccination coverage, but rather targeted intervention strategies based on country-specific risks.

Utilizing nanotechnology and advanced machine learning for early detection of gastric cancer surgery.

Nanotechnology has emerged as a promising frontier in revolutionizing the early diagnosis and surgical management of gastric cancers. The primary factors influencing curative efficacy in GIC patients are drug inefficacy and high surgical and pharmacological therapy recurrence rates. Due to its unique optical features, good biocompatibility, surface effects, and small size effects, nanotechnology is a developing and advanced area of study for detecting and treating cancer. Considering the limitations of GIC MRI and endoscopy and the complexity of gastric surgery, the early diagnosis and prompt treatment of gastric illnesses by nanotechnology has been a promising development. Nanoparticles directly target tumor cells, allowing their detection and removal. It also can be engineered to carry specific payloads, such as drugs or contrast agents, and enhance the efficacy and precision of cancer treatment. In this research, the boosting technique of machine learning was utilized to capture nonlinear interactions between a large number of input variables and outputs by using XGBoost and RNN-CNN as a classification method. The research sample included 350 patients, comprising 200 males and 150 females. The patients' mean ± SD was 50.34 ± 13.04 with a mean age of 50.34 ± 13.04. High-risk behaviors (P = 0.070), age at diagnosis (P = 0.034), distant metastasis (P = 0.004), and tumor stage (P = 0.014) were shown to have a statistically significant link with GC patient survival. AUC was 93.54%, Accuracy 93.54%, F1-score 93.57%, Precision 93.65%, and Recall 93.87% when analyzing stomach pictures. Integrating nanotechnology with advanced machine learning techniques holds promise for improving the diagnosis and treatment of gastric cancer, providing new avenues for precision medicine and better patient outcomes.

Ligand-based pharmacophore modeling and machine learning for the discovery of potent aurora A kinase inhibitory leads of novel chemotypes.

Aurora-A (AURKA) is serine/threonine protein kinase involved in the regulation of numerous processes of cell division. Numerous studies have demonstrated strong association between AURKA and cancer. AURKA is overexpressed in many cancers, such as colon, breast and prostate cancers. Consequently, AURKA has emerged as promising target for therapeutic intervention in cancer management. Herein, we describe a computational workflow for the discovery of novel anti-AURKA inhibitory leads starting with ligand-based assessment of the pharmacophoric space of six diverse sets of inhibitors. Subsequently, machine learning/QSAR modeling was coupled with genetic function algorithm to search for the best possible combination of machine learner, ligand-based pharmacophore(s) and molecular descriptors capable of explaining variation in anti-AURKA bioactivities within a collected list of inhibitors. Two learners succeeded in achieving acceptable structure/activity correlations, namely, random forests and extreme gradient boosting (XGBoost). Three pharmacophores emerged in the successful ML models. These were then used as 3D search queries to mine the National Cancer Institute database for novel anti-AURKA leads. Top-ranking 38 hits were assessed in vitro for their anti-AURKA bioactivities. Among them, three compounds exhibited promising dose-response curves, demonstrating experimental IC50 values ranging from sub-micromolar to low micromolar values. Remarkably, two of these compounds are of novel chemotypes.

Relationship between triglyceride-glucose index and blood eosinophils among asthmatic individuals in the USA.

BACKGROUND: Presently, the majority of investigations primarily evaluate the correlation between triglyceride-glucose index (TyGI) with lung diseases, such as asthma. However, they did not delve into the correlation between TyGI and inflammatory responses related to the disease. Few studies have explored the association between TyGI and blood eosinophil count (BEOC). Thus, National Health and Nutrition Examination Survey (NHANES) data were used in this study to evaluate the correlation between TyGI and BEOC in individuals with asthma. METHODS: This study investigated 3902 individuals with asthma. Linear regression analysis was performed to investigate the association between TyGI and BEOC in patients with asthma. Subsequently, the GAM and threshold effect models were used to validate the presence of either a nonlinear or linear association between TyGI and BEOC. Finally, stratified analyses were conducted to ascertain the correlations between different subgroups. RESULTS: Four linear regression models confirmed a positive linear correlation between TyGI and BEOC in patients with asthma. In Model D, which controlled for all covariates, BEOC increased by 12.44 cells/uL for every extra unit of TyGI. The GAM and threshold effect models further verified the positive linear correlation between TyGI and BEOC. The XGBoost model indicated that the six most significant variables influencing BEOC, in order of relative importance, were age, cholesterol level, body mass index (BMI), poverty-to-income ratio (PIR), BNEUC, and TyGI. CONCLUSIONS: In patients with asthma, the study discovered a linear positive correlation between TyGI and BEOC. This indicates a potential connection between TyGI and alterations in the immune status of individuals with asthma, which may help detect abnormalities in a timely manner and provide a reference for clinical decision-making. This study offers fresh insights for the future exploration of the management and treatment of asthma.

An Explainable Artificial Intelligence Model to Predict Malignant Cerebral Edema after Acute Anterior Circulating Large-Hemisphere Infarction.

INTRODUCTION: Malignant cerebral edema (MCE) is a serious complication and the main cause of poor prognosis in patients with large-hemisphere infarction (LHI). Therefore, the rapid and accurate identification of potential patients with MCE is essential for timely therapy. This study utilized an artificial intelligence-based machine learning approach to establish an interpretable model for predicting MCE in patients with LHI. METHODS: This study included 314 patients with LHI not undergoing recanalization therapy. The patients were divided into MCE and non-MCE groups, and the eXtreme Gradient Boosting (XGBoost) model was developed. A confusion matrix was used to measure the prediction performance of the XGBoost model. We also utilized the SHapley Additive exPlanations (SHAP) method to explain the XGBoost model. Decision curve and receiver operating characteristic curve analyses were performed to evaluate the net benefits of the model. RESULTS: MCE was observed in 121 (38.5%) of the 314 patients with LHI. The model showed excellent predictive performance, with an area under the curve of 0.916. The SHAP method revealed the top 10 predictive variables of the MCE such as ASPECTS score, NIHSS score, CS score, APACHE II score, HbA1c, AF, NLR, PLT, GCS, and age based on their importance ranking. CONCLUSION: An interpretable predictive model can increase transparency and help doctors accurately predict the occurrence of MCE in LHI patients not undergoing recanalization therapy within 48 h of onset, providing patients with better treatment strategies and enabling optimal resource allocation.

Data-driven shortened Insomnia Severity Index (ISI): a machine learning approach.

BACKGROUND: The Insomnia Severity Index (ISI) is a widely used questionnaire with seven items for identifying the risk of insomnia disorder. Although the ISI is still short, more shortened versions are emerging for repeated monitoring in routine clinical settings. In this study, we aimed to develop a data-driven shortened version of the ISI that accurately predicts the severity level of insomnia disorder. METHODS: We collected a sample of 800 responses from the EMBRAIN survey system. Based on the responses, seven items were grouped based on the similarity of their response using exploratory factor analysis (EFA). The most representative item within each group was selected by using eXtreme Gradient Boosting (XGBoost). RESULTS: Based on the selected three key items, maintenance of sleep, interference with daily function, and concerns about sleep problems, we developed a data-driven shortened questionnaire of ISI, ISI-3 m (machine learning). ISI-3 m achieved the highest coefficient of determination ( R 2 = 0.910 ) for the ISI score prediction task and the accuracy of 0.965, precision of 0.841, and recall of 0.838 for the multiclass-classification task, outperforming four previous versions of the shortened ISI. CONCLUSION: As ISI-3 m is a highly accurate shortened version of the ISI, it allows clinicians to efficiently screen for insomnia and observe variations in the condition throughout the treatment process. Furthermore, the framework based on the combination of EFA and XGBoost developed in this study can be utilized to develop data-driven shortened versions of the other questionnaires.

Interpretable machine learning models for predicting the incidence of antibiotic- associated diarrhea in elderly ICU patients.

BACKGROUND: Antibiotic-associated diarrhea (AAD) can prolong hospitalization, increase medical costs, and even lead to higher mortality rates. Therefore, it is essential to predict the incidence of AAD in elderly intensive care unit(ICU) patients. The objective of this study was to create a prediction model that is both interpretable and generalizable for predicting the incidence of AAD in elderly ICU patients. METHODS: We retrospectively analyzed data from the First Medical Center of the People's Liberation Army General Hospital (PLAGH) in China. We utilized the machine learning model Extreme Gradient Boosting (XGBoost) and Shapley's additive interpretation method to predict the incidence of AAD in elderly ICU patients in an interpretable manner. RESULTS: A total of 848 adult ICU patients were eligible for this study. The XGBoost model predicted the incidence of AAD with an area under the receiver operating characteristic curve (ROC) of 0.917, sensitivity of 0.889, specificity of 0.806, accuracy of 0.870, and an F1 score of 0.780. The XGBoost model outperformed the other models, including logistic regression, support vector machine (AUC = 0.809), K-nearest neighbor algorithm (AUC = 0.872), and plain Bayes (AUC = 0.774). CONCLUSIONS: While the XGBoost model may not excel in absolute performance, it demonstrates superior predictive capabilities compared to other models in forecasting the incidence of AAD in elderly ICU patients categorized based on their characteristics.

Using machine-learning models to predict extubation failure in neonates with bronchopulmonary dysplasia.

AIM: To develop a decision-support tool for predicting extubation failure (EF) in neonates with bronchopulmonary dysplasia (BPD) using a set of machine-learning algorithms. METHODS: A dataset of 284 BPD neonates on mechanical ventilation was used to develop predictive models via machine-learning algorithms, including extreme gradient boosting (XGBoost), random forest, support vector machine, naïve Bayes, logistic regression, and k-nearest neighbor. The top three models were assessed by the area under the receiver operating characteristic curve (AUC), and their performance was tested by decision curve analysis (DCA). Confusion matrix was used to show the high performance of the best model. The importance matrix plot and SHapley Additive exPlanations values were calculated to evaluate the feature importance and visualize the results. The nomogram and clinical impact curves were used to validate the final model. RESULTS: According to the AUC values and DCA results, the XGboost model performed best (AUC = 0.873, sensitivity = 0.896, specificity = 0.838). The nomogram and clinical impact curve verified that the XGBoost model possessed a significant predictive value. The following were predictive factors for EF: pO2, hemoglobin, mechanical ventilation (MV) rate, pH, Apgar score at 5 min, FiO2, C-reactive protein, Apgar score at 1 min, red blood cell count, PIP, gestational age, highest FiO2 at the first 24 h, heart rate, birth weight, pCO2. Further, pO2, hemoglobin, and MV rate were the three most important factors for predicting EF. CONCLUSIONS: The present study indicated that the XGBoost model was significant in predicting EF in BPD neonates with mechanical ventilation, which is helpful in determining the right extubation time among neonates with BPD to reduce the occurrence of complications.