Opening the black box: interpretable machine learning for predictor finding of metabolic syndrome

Abstract

Objective The internal workings ofmachine learning algorithms are complex and considered as low-interpretation "black box" models, making it difficult for domain experts to understand and trust these complex models. The study uses metabolic syndrome (MetS) as the entry point to analyze and evaluate the application value of model interpretability methods in dealing with difficult interpretation of predictive models. Methods The study collects data from a chain of health examination institution in Urumqi from 2017 ~ 2019, and performs 39,134 remaining data after preprocessing such as deletion and filling. RFE is used for feature selection to reduce redundancy; MetS risk prediction models (logistic, random forest, XGBoost) are built based on a feature subset, and accuracy, sensitivity, specificity, Youden index, and AUROC value are used to evaluate the model classification performance; post-hoc model-agnostic interpretation methods (variable importance, LIME) are used to interpret the results of the predictive model. Results Eighteen physical examination indicators are screened out by RFE, which can effectively solve the problem of physical examination data redundancy. Random forest and XGBoost models have higher accuracy, sensitivity, specificity, Youden index, and AUROC values compared with logistic regression. XGBoost models have higher sensitivity, Youden index, and AUROC values compared with random forest. The study uses variable importance, LIME and PDP for global and local interpretation of the optimal MetS risk prediction model (XGBoost), and different interpretation methods have different insights into the interpretation of model results, which are more flexible in model selection and can visualize the process and reasons for the model to make decisions. The interpretable risk prediction model in this study can help to identify risk factors associated with MetS, and the results showed that in addition to the traditional risk factors such as overweight and obesity, hyperglycemia, hypertension, and dyslipidemia, MetS was also associated with other factors, including age, creatinine, uric acid, and alkaline phosphatase. Conclusion The model interpretability methods are applied to the black box model, which can not only realize the flexibility of model application, but also make up for the uninterpretable defects of the model. Model interpretability methods can be used as a novel means of identifying variables that are more likely to be good predictors.

Risk prediction of gestational diabetes mellitus with four machine learning models

“…Still, these models also have several shortcomings. For one, they are all black box algorithms [52][53][54] . Although RF can be used to evaluate the relative importance of each variable, it is di cult to solve the relationship between variables in the model.…”

Interpretation of Drop Size Predictions from a Random Forest Model Using Local Interpretable Model-Agnostic Explanations (LIME) in a Rotating Disc Contactor

“…Random forest regression is a powerful and flexible black-box model. Its ability to handle nonlinear relationships, interactions between features, and noisy data makes it a popular choice among machine learning practitioners. , Random forest is an ensemble learning method that evolved from decision trees. Decision tree regression works by recursively partitioning the input space into regions, followed by assigning a constant value to each region based on the average or median value of the target variable in that region.…”

Machine Learning Approach for Metabolic Syndrome Diagnosis Using Explainable Data-Augmentation-Based Classification

“…Furthermore, waist-to-hip ratio and BMI were shown to be the most important indicators for metabolic prediction. The study in [ 21 ] collected a dataset of 39,134 Chinese metabolic syndrome patients. The data set contained information on 19 different diagnostic tests, including alkaline phosphatase, prior diabetes, uric acid, and eosinophil percentage.…”