MSCCov19Net: multi-branch deep learning model for COVID-19 detection from cough sounds

Abstract

Coronavirus has an impact on millions of lives and has been added to the important pandemics that continue to affect with its variants. Since it is transmitted through the respiratory tract, it has had significant effects on public health and social relations. Isolating people who are COVID positive can minimize the transmission, therefore several exams are proposed to detect the virus such as reverse transcription-polymerase chain reaction (RT-PCR), chest X-Ray, and computed tomography (CT). However, these methods suffer from either a low detection rate or high radiation dosage, along with being expensive. In this study, deep neural network–based model capable of detecting coronavirus from only coughing sound, which is fast, remotely operable and has no harmful side effects, has been proposed. The proposed multi-branch model takes M el Frequency Cepstral Coefficients (MFCC), S pectrogram, and C hromagram as inputs and is abbreviated as MSCCov19Net . The system is trained on publicly available crowdsourced datasets, and tested on two unseen (used only for testing) clinical and non-clinical datasets. Experimental outcomes represent that the proposed system outperforms the 6 popular deep learning architectures on four datasets by representing a better generalization ability. The proposed system has reached an accuracy of 61.5 % in Virufy and 90.4 % in NoCoCoDa for unseen test datasets. The Architecture of the Proposed MSCCov19Net Model for Covid-19 Detection

Deep Learning-Based Classification of Chest Diseases Using X-rays, CT Scans, and Cough Sound Images

“…This demonstrates that the suggested model was statistically distinct from the other contributing models since it combined more information from the base classifiers and produced better predictions. In this section, we evaluate the suggested DCDD_Net model with previous research [82][83][84][85][86][87]. In comparison to prior SOTA studies, Table 8 provides an in-depth analysis of the proposed DCDD_Net model in the context of numerous performance assessment criteria, including accuracy, recall, and F1-score.…”

“…Cross-sectional images are produced using a CT scan, which combines several X-ray images collected at various angles. Scalograms represent the actual frequencies of a wave's continuous wavelet transform (CWT) factors [82][83][84][85][86][87]. Cough signals utilize CWT to convey data from the time domain to the frequency domain, as demonstrated in Figure 3.…”

“…The success rate for the classification of the proposed DCDD_Net with SOTA classifiers is shown in Table 8. Ibrahim et al [82] used the AlexNet model for the classification of five chest diseases using CXR images. They obtained the images from different public databases.…”

Multi-modal deep learning methods for classification of chest diseases using different medical imaging and cough sounds

“…According to the findings of their research, binary classifiers have the potential to achieve an AUC of 0.964% and an accuracy of 96%. Using the methodology presented in [ 88 ] sounds that do not belong to the COVID-19 family may be discriminated against from COVID-19 sounds. For training and evaluation, they used a total of 50 groups, with each group including 3,597 noises that were unrelated to coughing and 1,838 coughs.…”

Accumulated bispectral image-based respiratory sound signal classification using deep learning

“…In [ 16 ], the author has used a multi-branch input convolution network with MFCC, spectrogram, and chromagram features. The VIRUFY dataset achieved 61% accuracy and 90.4% accuracy in the NoCoCoDa dataset.…”