An Overview of Supervised Machine Learning Methods and Data Analysis for COVID-19 Detection

Abstract

Background and Objective. To mitigate the spread of the virus responsible for COVID-19, known as SARS-CoV-2, there is an urgent need for massive population testing. Due to the constant shortage of PCR (polymerase chain reaction) test reagents, which are the tests for COVID-19 by excellence, several medical centers have opted for immunological tests to look for the presence of antibodies produced against this virus. However, these tests have a high rate of false positives (positive but actually negative test results) and false negatives (negative but actually positive test results) and are therefore not always reliable. In this paper, we proposed a solution based on Data Analysis and Machine Learning to detect COVID-19 infections. Methods. Our analysis and machine learning algorithm is based on most cited two clinical datasets from the literature: one from San Raffaele Hospital Milan Italia and the other from Hospital Israelita Albert Einstein São Paulo Brasilia. The datasets were processed to select the best features that most influence the target, and it turned out that almost all of them are blood parameters. EDA (Exploratory Data Analysis) methods were applied to the datasets, and a comparative study of supervised machine learning models was done, after which the support vector machine (SVM) was selected as the one with the best performance. Results. SVM being the best performant is used as our proposed supervised machine learning algorithm. An accuracy of 99.29%, sensitivity of 92.79%, and specificity of 100% were obtained with the dataset from Kaggle (https://www.kaggle.com/einsteindata4u/covid19) after applying optimization to SVM. The same procedure and work were performed with the dataset taken from San Raffaele Hospital (https://zenodo.org/record/3886927#.YIluB5AzbMV). Once more, the SVM presented the best performance among other machine learning algorithms, and 92.86%, 93.55%, and 90.91% for accuracy, sensitivity, and specificity, respectively, were obtained. Conclusion. The obtained results, when compared with others from the literature based on these same datasets, are superior, leading us to conclude that our proposed solution is reliable for the COVID-19 diagnosis.

An automated machine learning model for diagnosing coronavirus disease 2019 (COVID-19) infection

“…The best pipeline has the highest cross-validation score in the run at 95.94%. [32] This study predicted COVID-19 via data analysis (DA) and ML models based on patient clinical data. The authors presented five models to be trained and evaluated according to well-defined evaluation criteria, aiming to select the best model.…”

Using Artificial Intelligence for COVID-19 Detection in Blood Exams: A Comparative Analysis

“…Several works [31], [32], [33], [34], [35], [36], [37], [38], [39] have also focused on datasets containing different types of routine clinical information, including data extracted from blood tests [40], [41], [42], [43]. These datasets, often acquired under emergency conditions, are highly varied in terms of the features considered as well as the specific targets of the analysis.…”

“…These datasets, often acquired under emergency conditions, are highly varied in terms of the features considered as well as the specific targets of the analysis. In some cases, the focus was on the most influential hematological features for the identification of COVID-19 positive patients [31], [32], [33], [34], [35], [36], [38]. Other works concentrated on early detection models to distinguish hospital admissions due to COVID-19 and possible entry into emergency department [37], or to distinguish between COVID-19 and influenza [39].…”

“…Within this frame of reference, several traditional machine learning algorithms were used, from Decision Trees and Random Forest [31], [32], [35] to Bayes Network [33]. Also, SVM-based strategies [36], eXtreme Gradient Boosting [37], [38], [39], and hierarchical classification systems [34] were proposed.…”

A Social Network Analysis Approach to Evaluate the Relationship Between the Mobility Network Metrics and the COVID-19 Outbreak

“…Machine learning techniques mainly include classification methods such as random forest, neural networks, Bayesian networks, and regression tree (cart). 6 Researchers used neural network and Bayesian network techniques, respectively to predict the prevalence of dengue infection, 7,8 and similarly, using neural network and genetic algorithms the Oyster norovirus outbreak has been estimated. 9,10 Besides, various approaches have been proposed to deal with the transmission of new coronavirus and predict the future prevalence by researchers in different countries so far.…”