Deep-learning Based Approach to Identify Covid-19

Abstract

COVID-19 is a large-scale contagious respiratory disease that has spread across the world in 2020. Therefore, a low-cost, fast, and easily available solution is needed to provide a COVID-19 diagnosis to curb the outbreak. According to recent studies, one of the main symptoms of COVID-19 is coughing. The goal of this research effort is to develop a method for the automatic diagnosis of COVID-19 by detecting cough during recorded conversations. The method is composed of five main modules: sound extraction, sound feature extraction, cough detection, cough classification, and COVID-19 diagnosis. The method extracts relevant features from the audio signal and then uses machine learning and deep learning models, like SVM, KNN, and RNN, to classify them, which can successfully diagnose COVID-19 from audio recordings. Our method has relatively high accuracy when dealing with completely unfamiliar cough samples. When the training set and the test set are from two different databases, it still achieves an accuracy of 81.25% (AUC of 0.79). As more data sets are collected, the model can be further developed and improved to create a machine learning solution based on cough analysis for COVID-19 detection, which may be promoted as a non-clinical selfinspection solution.

Machine learning for detecting COVID-19 from cough sounds: An ensemble-based MCDM method

“… k-Nearest Neighbor (k-NN) is a well-known classifier that appears in large-scale ML applications. As we have seen from previous studies, researchers used k-NN in non-COVID-19 applications such as night coughing and sniffing [ 50 ] and used k-NN to detect COVID-19 in cough samples [ 32 , 38 , 47 , 51 ]. Histogram-based Gradient Boosting (HGBoost) is a highly desirable ML technology, where the application needs to get better quality performance in less inference time.…”

“…k-Nearest Neighbor (k-NN) is a well-known classifier that appears in large-scale ML applications. As we have seen from previous studies, researchers used k-NN in non-COVID-19 applications such as night coughing and sniffing [ 50 ] and used k-NN to detect COVID-19 in cough samples [ 32 , 38 , 47 , 51 ].…”

Multi-modal Point-of-Care Diagnostics for COVID-19 Based On Acoustics and Symptoms

“…Agbley et al [23] demonstrated a specificity of 0.81 at a sensitivity of 0.43 on a subset of COUGHVID dataset. Feng et al [24] used a subset of cough sounds from Coswara dataset and reported a performance of 0.90 AUC. Laguarte et al [20] obtained AUC greater than 0.90 on samples from the COVID-19 Cough data set.…”

“…Laguarte et al [20] obtained AUC greater than 0.90 on samples from the COVID-19 Cough data set. These studies use acoustic feature representations of cough sounds such as Mel frequency cepstral coefficients (MFCCs) [18], Mel-spectrogram [20], [22], or scalograms [23], while the classifier models are deep learning based neural networks such as convolutional neural networks (CNNs) [23], recurrent neural networks (RNNs) [24], CNN based feature embeddings in support vector machines (SVM) [18] or with CNN based residual networks [20], [22]. There are also attempts at creating more controlled COVID-19 cough sound dataset from individuals in hospitals [25], [26].…”

Diagnosis of COVID-19 and non-COVID-19 patients by classifying only a single cough sound

“…The authors used the MFCCs method in the feature extraction stage, and the accuracy rate of the system was calculated as 92.85%. In a study [38] on Coswara and Virufy data sets, features extracted from the frequency and time domain were classified using machine learning methods. The results were compared with the recurrent neural network (RNN) method, which is common in deep learning methods.…”