An enhanced diabetes prediction amidst COVID-19 using ensemble models

Abstract

In the contemporary landscape of healthcare, the early and accurate prediction of diabetes has garnered paramount importance, especially in the wake of the COVID-19 pandemic where individuals with diabetes exhibit increased vulnerability. This research embarked on a mission to enhance diabetes prediction by employing state-of-the-art machine learning techniques. Initial evaluations highlighted the Support Vector Machines (SVM) classifier as a promising candidate with an accuracy of 76.62%. To further optimize predictions, the study delved into advanced feature engineering techniques, generating interaction and polynomial features that unearthed hidden patterns in the data. Subsequent correlation analyses, visualized through heatmaps, revealed significant correlations, especially with attributes like Glucose. By integrating the strengths of Decision Trees, Gradient Boosting, and SVM in an ensemble model, we achieved an accuracy of 93.2%, showcasing the potential of harmonizing diverse algorithms. This research offers a robust blueprint for diabetes prediction, holding profound implications for early diagnosis, personalized treatments, and preventive care in the context of global health challenges and with the goal of increasing life expectancy.

Keywords: COVID-19; classification; correlation analysis; diabetes; ensemble models; feature engineering; interaction; polynomial.

Figure 1

The word cloud.

Figure 2

Various features original and median imputed distribution.

Figure 3

Box plot for all the features.

Figure 4

Corrected box plots for all the features.

Figure 5

Cumulative and individual explained variance by different principal components, 2D PCA of diabetes data, ROC curves and PR curves of four different classifiers—Logistic Regression, Random Forest, Gradient Boosting, and Support Vector Machines.

Figure 6

Confusion matrix for the classifiers used in this work.

Figure 7

Correlation matrix.

Figure 8

Visual representation of the model's performance across various metrics, trade-off between precision and recall at various thresholds and receiver operating characteristic (ROC) curve provides insights into the model's ability to discriminate between the classes.