A Comparison of XGBoost, Random Forest, and Nomograph for the Prediction of Disease Severity in Patients With COVID-19 Pneumonia: Implications of Cytokine and Immune Cell Profile

Abstract

Background and AimsThe aim of this study was to apply machine learning models and a nomogram to differentiate critically ill from non-critically ill COVID-19 pneumonia patients.MethodsClinical symptoms and signs, laboratory parameters, cytokine profile, and immune cellular data of 63 COVID-19 pneumonia patients were retrospectively reviewed. Outcomes were followed up until Mar 12, 2020. A logistic regression function (LR model), Random Forest, and XGBoost models were developed. The performance of these models was measured by area under receiver operating characteristic curve (AUC) analysis.ResultsUnivariate analysis revealed that there was a difference between critically and non-critically ill patients with respect to levels of interleukin-6, interleukin-10, T cells, CD4+ T, and CD8+ T cells. Interleukin-10 with an AUC of 0.86 was most useful predictor of critically ill patients with COVID-19 pneumonia. Ten variables (respiratory rate, neutrophil counts, aspartate transaminase, albumin, serum procalcitonin, D-dimer and B-type natriuretic peptide, CD4+ T cells, interleukin-6 and interleukin-10) were used as candidate predictors for LR model, Random Forest (RF) and XGBoost model application. The coefficients from LR model were utilized to build a nomogram. RF and XGBoost methods suggested that Interleukin-10 and interleukin-6 were the most important variables for severity of illness prediction. The mean AUC for LR, RF, and XGBoost model were 0.91, 0.89, and 0.93 respectively (in two-fold cross-validation). Individualized prediction by XGBoost model was explained by local interpretable model-agnostic explanations (LIME) plot.ConclusionsXGBoost exhibited the highest discriminatory performance for prediction of critically ill patients with COVID-19 pneumonia. It is inferred that the nomogram and visualized interpretation with LIME plot could be useful in the clinical setting. Additionally, interleukin-10 could serve as a useful predictor of critically ill patients with COVID-19 pneumonia.

Glutamine and amino acid metabolism as a prognostic signature and therapeutic target in endometrial cancer

“…Because when the amount of data is large with many missing values, and the number of independent variables is much larger than the sample size, the traditional Cox regression forward, backward, and stepwise method may not be applicable. In addition to these methods, the random forest, Support Vector Machine, principle component analysis, 33 deep learning, and extreme gradient boosting of machine learning are also becoming much popular 32,34,35 . Machine learning are more robust and can outfit imbalanced datasets 35 .…”

Application of Machine Learning to Improve Appropriateness of Treatment in an Orthopaedic Setting of Personalized Medicine

“…In addition, XGBoost employs a sparsity-aware algorithm [ 25 ] that automatically handles missing data values [ 29 ], including hyperparameters that provide tweaking for unbalanced datasets [ 30 ]. XGBoost outperforms other algorithms across a wide range of feature sets and in various settings [ 31 ], including orthopedics (e.g., Li and Zhang; Bugarin et al [ 32 , 33 ]).…”

Potential predictors of severe course and outcome of community-acquired pneumonia

“…По результатам исследования H.Han et al с использованием ряда маркеров воспаления у пациентов с COVID-19 (n = 102) также подтвердилось значение IL-6 и IL-10 как прогностических показателей для быстрой диагностики высокого риска ухудшения течения заболевания [23]. При применении искусственного интеллекта (моделей машинного обучения) для дифференциации пациентов с SARS-CoV-2-ассоциированной пневмонией, находящихся в критическом и некритическом состоянии, выявлено, что IL-10 и IL-6 являются наиболее важными показателями для прогнозирования тяжести заболевания [24]. Отмечено также последовательное увеличение уровней IL-6, IL-10 и моноцитарного хемоаттрактантного белка-1 по мере утяжеления состояния пациентов от легкого до тяжелого и критического [25].…”