Insulinaemia and insulin resistance in Caucasian general paediatric population aged 2 to 10 years: Associated risk factors

Simple Method to Predict Insulin Resistance in Children Aged 6-12 Years by Using Machine Learning

BACKGROUND

Due to the increasing insulin resistance (IR) in childhood, rates of diabetes and cardiovascular disease may rise in the future and seriously threaten the healthy development of children. Finding an easy way to predict IR in children can help pediatricians to identify these children in time and intervene appropriately, which is particularly important for practitioners in primary health care.

PATIENTS AND METHODS

Seventeen features from 503 children 6-12 years old were collected. We defined IR by HOMA-IR greater than 3.0, thus classifying children with IR and those without IR. Data were preprocessed by multivariate imputation and oversampling to resolve missing values and data imbalances; then, recursive feature elimination was applied to further select features of interest, and 5 machine learning methods-namely, logistic regression (LR), support vector machine (SVM), random forest (RF), extreme gradient boosting (XGBoost), and gradient boosting with categorical features support (CatBoost)-were used for model training. We tested the trained models on an external test set containing information from 133 children, from which performance metrics were extracted and the optimal model was selected.

RESULTS

After feature selection, the numbers of chosen features for the LR, SVM, RF, XGBoost, and CatBoost models were 6, 9, 10, 14, and 6, respectively. Among them, glucose, waist circumference, and age were chosen as predictors by most of the models. Finally, all 5 models achieved good performance on the external test set. Both XGBoost and CatBoost had the same AUC (0.85), which was highest among those of all models. Their accuracy, sensitivity, precision, and F1 scores were also close, but the specificity of XGBoost reached 0.79, which was significantly higher than that of CatBoost, so XGBoost was chosen as the optimal model.

CONCLUSION

The model developed herein has a good predictive ability for IR in children 6-12 years old and can be clinically applied to help pediatricians identify children with IR in a simple and inexpensive way.

Variants rs1544410 and rs2228570 of the vitamin D receptor gene and glycemic levels in adolescents from Northeast Brazil

Introduction

Objective: to verify the association of serum concentrations of 25-hydroxyvitamin D and glycemic levels with the genetic variants rs1544410 and rs2228570 of the VDR gene in adolescents from the Northeast region of Brazil. Materials and methods: a cross-sectional epidemiological study with 208 adolescents from public schools in the city of João Pessoa (Paraíba, Brazil) between 15 and 19 years of age. Blood samples were collected for DNA extraction and analysis of polymorphisms rs1544410 and rs2228570, as well as biochemical analyses (25-hydroxyvitamin D, parathyroid hormone, calcium and glycemia). Results: the mean age was 17.7 (± 1.14) years. Half of adolescents had sufficient serum levels of 25-hydroxyvitamin D and the other half had insufficient/deficient vitamin. The most frequent genotypic distribution was bb and Ff and of lesser frequency BB and ff. There was a significant relationship between the genotypes of rs1544410 and glycemia values (p = 0.049) in the relationships between the genotypes BBxbb (p = 0.012) and Bbxbb (p = 0.037); (p = 0.036, OR = 2.15, 95% CI = 1.05-4.41), and in the BB+Bb group analysis when compared to the bb (p = 0.025, OR = 1.89, 95% CI = 1.08-3.29) presented higher risk of glycemia above the median. On the other hand, when Bb+bb was analyzed in relation to BB, adolescents had a greater chance of blood glucose below the median (p = 0.025, OR = 0.66, CI = 0.47-0.95). Conclusion: this study showed a significant relation of glycemia with the distribution of rs1544410 polymorphism genotypes.