Development of a Smartphone-Based Expert System for COVID-19 Risk Prediction at Early Stage

Abstract

COVID-19 has imposed many challenges and barriers on traditional healthcare systems due to the high risk of being infected by the coronavirus. Modern electronic devices like smartphones with information technology can play an essential role in handling the current pandemic by contributing to different telemedical services. This study has focused on determining the presence of this virus by employing smartphone technology, as it is available to a large number of people. A publicly available COVID-19 dataset consisting of 33 features has been utilized to develop the aimed model, which can be collected from an in-house facility. The chosen dataset has 2.82% positive and 97.18% negative samples, demonstrating a high imbalance of class populations. The Adaptive Synthetic (ADASYN) has been applied to overcome the class imbalance problem with imbalanced data. Ten optimal features are chosen from the given 33 features, employing two different feature selection algorithms, such as K Best and recursive feature elimination methods. Mainly, three classification schemes, Random Forest (RF), eXtreme Gradient Boosting (XGB), and Support Vector Machine (SVM), have been applied for the ablation studies, where the accuracy from the XGB, RF, and SVM classifiers achieved 97.91%, 97.81%, and 73.37%, respectively. As the XGB algorithm confers the best results, it has been implemented in designing the Android operating system base and web applications. By analyzing 10 users' questionnaires, the developed expert system can predict the presence of COVID-19 in the human body of the primary suspect. The preprocessed data and codes are available on the GitHub repository.

Keywords: Android or web-based user applications; COVID-19 prediction; adaptive synthetic sampling; feature selection methods; machine learning classifiers.

Figure 1

The recommended workflow for the identification of COVID-19 in conjunction with the proposed user application.

Figure 2

Heat–map representation of the features, dispensing the correlations between them.

Figure 3

Flow chart of the working mechanism of the XGB algorithm in the XGB classifier. The highest accuracy was achieved by applying the XGB algorithm to the dataset.

Figure 4

Confusion matrices of (a) XGB, (b) RF, (c) SVM, (d) Twin-SVM with linear kernel, and (e) Twin-SVM with RBF kernel for evaluating their class-wise performance.

Figure 5

(a) ROC of RF, XGB, and SVM; and (b) bootstrap ROC analysis of XGB. The XGB classifier outperformed all the other algorithms.

Figure 6

(a) Box and Whisker plot of 10-fold CV of the main three ML classifiers and (b) a multi-comparison test of the three trained models in this article.

Figure 7

Feature importance and cumulative importance using XGB. The most important feature is loss_of_test, which is a noteworthy discovery from the dataset investigation. Then, (a) showed the selected 10 critical features, and (b) visualizes the cumulative feature importance curve, which also indicates that it grows to 95% above.

Figure 8

Violin plot of data distribution in terms of fever and cough.

Figure 9

Association rules (a) scatter plot, (b) support and lift, and (c) items in the left hand side (LHS) and the right hand side (RHS) group.

Figure 10

Screenshot of the designed web application for COVID-19 prognostication, deploying the proposed framework.

Figure 11

Several screenshots, such as (a) queries from users, (b) queries from users, and (c) predicted results of the implemented mobile application for COVID-19 prognostication, deploying the proposed framework.