Development of a Machine Learning Based Web Application for Early Diagnosis of COVID-19 Based on Symptoms

Abstract

Detecting the presence of a disease requires laboratory tests, testing kits, and devices; however, these were not always available on hand. This study proposes a new approach in disease detection using machine learning algorithms by analyzing symptoms experienced by a person without requiring laboratory tests. Six supervised machine learning algorithms such as J48 decision tree, random forest, support vector machine, k-nearest neighbors, naïve Bayes algorithms, and artificial neural networks were applied in the “COVID-19 Symptoms and Presence Dataset” from Kaggle. Through hyperparameter optimization and 10-fold cross validation, we attained the highest possible performance of each algorithm. A comparative analysis was performed according to accuracy, sensitivity, specificity, and area under the ROC curve. Results show that random forest, support vector machine, k-nearest neighbors, and artificial neural networks outweighed other algorithms by attaining 98.84% accuracy, 100% sensitivity, 98.79% specificity, and 98.84% area under the ROC curve. Finally, we developed a web application that will allow users to select symptoms currently being experienced, and use it to predict the presence of COVID-19 through the developed prediction model. Based on this mechanism, the proposed method can effectively predict the presence or absence of COVID-19 in a person immediately without using laboratory tests, kits, and devices in a real-time manner.

A Semi-Supervised Machine Learning Approach in Predicting High-Risk Pregnancies in the Philippines

“…Moreover, mothers can personally use this model to know their risk level, should they be able to access this platform. In future studies, the authors plan to develop an easy-to-use web application [ 30 ], not just for health workers, but also for mothers to eventually use but with medical guidance. This application will then tell the mothers their next steps and will alert health workers to their risk level and condition.…”

Graph data science and machine learning for the detection of COVID-19 infection from symptoms

“…Two studies by Villavicencio et al (2021 , 2022) analyzed the same COVID-19 symptom dataset. One utilized five supervised ML techniques, namely the support vector machine (SVM), random forest (RF), NB algorithms, k-nearest neighbor, and J48 decision tree, using WEKA machine learning software ( Villavicencio et al, 2021 ).…”

“…They showed that the SVM outperformed other ML methods in terms of accuracy and MAE. The study by Villavicencio et al (2022) used the same ML algorithms as that in 2021 along with an ANN for 18 selected features. Furthermore, the performance measures in Villavicencio et al (2022) compared the accuracy, specificity, sensitivity, and AUROC of the six models.…”

“…The study by Villavicencio et al (2022) used the same ML algorithms as that in 2021 along with an ANN for 18 selected features. Furthermore, the performance measures in Villavicencio et al (2022) compared the accuracy, specificity, sensitivity, and AUROC of the six models. This study used a web application that allows users to manage their COVID-19 health status in real time without undergoing laboratory tests.…”

“…Table 1 presents a comparative analysis for the main advantages and limitations between these different studies and the proposed method presented in this article. Moreover, we compared our results with those of the state-of-the-art study ( Villavicencio et al, 2022 ), which achieved the highest accuracy of 98.84% and used the same COVID-19 dataset.…”