Compare the classification performances of convolutional neural networks and capsule networks on the Coswara dataset

Öz

Since the onset of the COVID-19 pandemic, numerous machine learning models have been developed to classify and distinguish COVID-19 positive sounds from egative ones. The aim of this study is to compare the classification performances of convolutional neural networks and capsule networks on the Coswara dataset. The dataset was collected using a website application and contains 1404 and 522 healthy COVID-19 positive subjects. Each subject contains nine different types of sounds. After feature extraction, the dataset was preprocessed by applying oversampling (SOMTE) and normalization (MinMax Scaler) techniques. K-fold cross-validation was used to train and evaluate the models. The CNN classifiers achieved an (AUC) of 90%, while the CapsNet classifiers achieved an (AUC) of 86%. Finally, when leave-one-out cross-validation was used, the CNN classifier achieved an (AUC) of 99%. In addition, we also compared the performance of the CNN and CapsNet networks without applying any preprocessing techniques to the Coswara dataset. The CNN classifiers achieved an AUC of 88%, while the CapsNet classifiers achieved an AUC of 50% without applying oversampling techniques. Moreover, the CNN classifiers achieved an AUC of 81%, while the CapsNet classifiers achieved an AUC of 55% without applying normalization techniques.

Anahtar Kelimeler

• COVID-19
• CNN
• CapsNet
• k-fold
• leave-one-out

Compare the classification performances of convolutional neural networks and capsule networks on the Coswara dataset

Abstract

Since the beginning of the COVID-19 pandemic, researchers have developed numerous machine learning models to distinguish between positive and negative COVID-19 sounds. The aim of this study is to compare the classification performances of convolutional neural networks (CNN) and capsule networks (CapsNet) on the Coswara dataset, which includes 1404 healthy subjects and 522 COVID-19 positive subjects, each containing nine different types of sounds. The dataset was preprocessed by using oversampling and normalization techniques after feature extraction. k-fold cross-validation was used (where k=10) to train and evaluate the models. The CNN classifiers achieved a 94% ACC, while the CapsNet classifiers achieved an 90% ACC. Furthermore, when using leave-one-out cross-validation, the CNN classifier achieved an ACC of 99%. we also compared the performance of the CNN and CapsNet networks on the Coswara dataset without preprocessing. Without oversampling techniques, the CNN classifiers achieved an 93% ACC, compared to 54% for the CapsNet classifiers. When normalization techniques were not applied, the CNN classifiers achieved an 86% ACC, while the CapsNet classifiers achieved a 26% ACC.

Keywords

• COVID-19
• CNN
• CapsNet
• k-fold
• leave-one-out

References

1. [1] WHO, https://www.who.int/health-topics/coronavirus.
2. [2] D. Wang, B. Hu, C. Hu, F. Zhu, X. Liu, J. Zhang, B. Wang, H. Xiang, Z. Cheng, Y. Xiong et al, “Clinical characteristics of 138 hospitalized patients with 2019 novel coronavirus–infected pneumonia in Wuhan, China”, JAMA, vol. 323, no. 11, pp. 1061– 1069, 2020.
3. [3] A. I. Khan , J. L.Shah , M. M. Bhat, “CoroNet: A deep neural network for detection and diagnosis of COVID-19 from chest x-ray images”,2020.
4. [4] S.Walvekar,D. Shinde, “Detection of COVID-19 from CT Images Using resnet50”, 2020.
5. [5] P.Aggarwal , N. K. Mishra, B.Fatimah , P. Singh , A. Gupta , S. D. Joshi ,”COVID-19 image classification using deep learning: Advances, challenges and opportunities”, 2022, 105350.
6. [6] P.Bagad,A.Dalmia, J. Doshi, A. Nagrani, P. Bhamare, A.Mahale,S.Rane, N. Agarwal, R.Panicker, “Cough Against COVID: Evidence of COVID-19 Signature in Cough Sounds |”,2020.
7. [7] M.Pahar, M.Klopper, R. Warren, T.Niesler, “COVID-19 Cough : Classification using Machine Learning and Global Smartphone Recordings” ,2021,104572.
8. [8] M.Aly, K.H. Rahouma, S. M. Ramzy,” Pay attention to the speech: COVID-19 diagnosis using machine learning and crowdsourced respiratory and speech recordings”, pp 3487-3500, 2022.

9. [9] https://www.kaggle.com/datasets/janashreeananthan/coswara .
10. [10] https://coswara.iisc.ac.in/?locale=en-US.
11. [11] https://imbalanced-learn.org/stable.
12. [12] https://scikit- learn.org/stable/modules/generated/sklearn.preprocessing.MinMaxScaler.html.
13. [13] https://librosa.org/doc/main/generated/librosa.feature. mfcc.html.
14. [14] R. Yamashita & M. Nishio & R.Gian Do & K. Togashi,” Convolutional neural networks: an overview and application in radiology”, pp.611–629, 2018.
15. [15] M. Patrick, A. Adekoya a , A.Mighty , B.Edward : “Capsule Networks – A survey”, pp1295-1310, 2022.
16. [16] https://github.com/XifengGuo/CapsNet-Keras.
17. [17] O. Abayomi-Alli, R.Damaševičius ,A.Abbasi ,R.Maskeliūnas s ,” Detection of COVID-19 from Deep Breathing Sounds Using Sound Spectrum with Image Augmentation and Deep Learning Techniques”, 2022.
18. [18] G. Chaudhari, X.Jiang, A.Fakhry, A. Han, J. Xiao, S. Shen, A. Khanzada,” Virufy: Global Applicability of Crowdsourced and Clinical Datasets for AI Detection of COVID-19 from Cough”,2020.
19. [19] L. Orlandic, T. Teijeiro, D. Atienza, The COUGHVID crowdsourcing dataset: A corpus for the study of large-scale cough analysis algorithms, 2020.
20. [20] L.VERDE, G.DE PIETRO, A. GHONEIM, M. ALRASHOUD , K. N. AL-MUTIB , G. SANNINO ,” Exploring the Use of Artificial Intelligence Techniques to Detect the Presence of Coronavirus Covid-19 Through Speech and Voice Analysis”, 2021.3075571.