Skip to content
KPI Science News
ISSN 2617-5509 · e‑ISSN 2663-7472 Open Access · CC BY 4.0
Vol. · Issue 3 · 2021 Dec 15, 2021 article

COMPARATIVE ANALYSIS OF DEEP NEURAL NETWORK MODELS FOR PREDICTION OF COVID-19 DIAGNOSIS

NN
Nadezhda I. Nedashkovskaya Corresponding Igor Sikorsky Kyiv Polytechnic Institute, Ukraine Ukraine
OS
Oleksandr O. Sapelnikov Igor Sikorsky Kyiv Polytechnic Institute, Ukraine Ukraine
PublishedDec 15, 2021
PDF (Ukrainian)
KPISN 3 VOL · 3
VOL · ISS 3 · 2021 View issue

Abstract

Background . The situation with the coronavirus COVID-19 in mid-2021 is causing concern around the world again due to the emergence of a new more dangerous strain “delta”. Attempts are being made to build mathematical models for describing the disease spread. China, the United States and other countries are developing artificial intelligence tools to help predict the COVID-19 diagnosis and are deploying them in hospitals. Objective. The purpose of the paper is to compare how different deep convolutional neural network models deal with COVID-19 disease recognition based on chest X-rays. Methods. The models are based on the VGG16, ResNet50 and SqueezeNet network architectures with the addition of layers to regularize these networks. The models were trained using the transfer learning technology. The quality of the models was assessed based on the confusion matrix, precision, recall, specificity and F-scores on the validation set of X-ray images. Results. Several neural network models have been built to classify an X-ray image of the human chest into two classes: person with COVID-19 (class 1) or healthy person (class 2). The ResNet50 model achieved a fairly high precision of 96 % for class 1 and an overall specificity of 96.14 %. Recall disease detection based on this model was 88%. The VGG-16 model correctly classified 100% (Experiment 1) / 96 % (Experiment 2) of COVID-19 patients in the test sample. The patient class precision and overall specificity values based on this model, however, were lower at 89 % / 92 % and 88.46 % / 92.31 %, respectively, for the two experiments. The F-score values for the ResNet50 and VGG-16 models were quite significant and equal to 92 and 94 %. Conclusions. Deep convolutional neural network models have shown promising results in predicting the COVID-19 diagnosis and require research. Predicting the diagnosis using the developed software based on these models takes up to one minute – this is faster than RT-PCR (reverse transcription PCR) tests, which are used to confirm the disease.

Keywords

#convolutional neural networks #deep learning #image classification #predicting the COVID-19 diagnosis

References

  1. World Health Organization, Coronavirus disease (COVID-19) outbreak situation. Available: https://www.who.int . (Accessed: 20.09.2021).
  2. World Health Organization. Coronavirus disease (COVID-2019) situation reports. Available: https://www.who.int/emergencies/diseases/novel-coronavirus-2019/situation-reports (Accessed: 20.09.2021).
  3. Coronavirus COVID-19 global cases by the Center for systems science and engineering at Johns Hopkins University. Available: https://coronavirus.jhu.edu/map.html . (Accessed: 20.09.2021).
  4. Coronavirus in Ukraine. Available: https://index.minfin.com.ua/ua/reference/coronavirus/ukraine/ (Accessed: 20.09.2021).
  5. R. A. Middelburg and F. R. Rosendaal, “COVID-19: How to make between-country comparison”, Int. J. Infect. Diseases, vol. 96, pp. 477–481, Jul. 2020, doi: 10.1016/j.ijid.2020.05.066.
  6. M. Yousaf, S. Zahir, M. Riaz, S. M. Hussain, and K. Shah, “Statistical analysis of forecasting COVID-19 for upcoming month in Pakistan”, Chaos Solit. Fractals, vol. 138, p. 109926, Sep. 2020, doi: 10.1016/j.chaos.2020.109926.
  7. S. Singh, K. S. Parmar, J. Kumar, and S. J. S. Makkhan, “Development of new hybrid model of discrete wavelet decomposition and autoregressive integrated moving average (ARIMA) models in application to one month forecast the casualties cases of COVID-19”, Chaos Solit. Fractals, vol. 135, p. 109866, Jun. 2020, doi: 10.1016/j.chaos.2020.109866.
  8. T. Chakraborty and I. Ghosh, “Real-time forecasts and risk assessment of novel coronavirus (COVID-19) cases: A data-driven analysis”, Chaos Solit. Fractals, vol. 135, p. 109850, Jun. 2020, doi: 10.1016/j.chaos.2020.109850.
  9. D. Fanelli and F. Piazza, “Analysis and forecast of COVID-19 spreading in China, Italy and France”, Chaos Solit. Fractals, vol. 134, p. 109761, May 2020, doi: 10.1016/j.chaos.2020.109761.
  10. G. Rainisch, E. A. Undurraga, and G. Chowell, “A dynamic modeling tool for estimating healthcare demand from the COVID19 epidemic and evaluating population-wide interventions”, Int. J. Infect. Diseases, vol. 96, pp. 376–383, Jul. 2020, doi: 10.1016/j.ijid.2020.05.043.
  11. K. Sarkar, S. Khajanchi, and J. J. Nieto, “Modeling and forecasting the COVID-19 pandemic in India”, Chaos Solit. Fractals, vol. 139, p. 110049, Oct. 2020, doi: 10.1016/j.chaos.2020.110049.
  12. R. Salgotra, M. Gandomi, A. H. Gandomi, “Time series analysis and forecast of the covid-19 pandemic in India using genetic programming”, Chaos Solit. Fractals, vol. 138, p. 109945, Sep. 2020, doi: 10.1016/j.chaos.2020.109945.
  13. N. Chintalapudi, G. Battineni, G. G. Sagaro, and F. Amenta, “COVID-19 outbreak reproduction number estimations and forecasting in Marche, Italy”, Int. J. Infect. Diseases, vol. 96, pp. 327–333, Jul. 2020, doi: 10.1016/j.ijid.2020.05.029.
  14. E. Aviv-Sharon and A. Aharoni, “Generalized logistic growth modeling of the COVID-19 pandemic in Asia”, Infect. Disease Modelling, vol. 5, pp. 505–509, 2020, doi: 10.1016/j.idm.2020.07.003.
  15. Wired 2020 Chinese hospitals deploy AI to help diagnose Covid-19. Avaliable: https://www.wired.com/story/ chinese-hospitals-deploy-ai-help-diagnose-covid-19. (Accessed: 20.09.2021)
  16. Y. Fang et al., “Sensitivity of chest CT for COVID-19: Comparison to RT-PCR”, Radiology, vol. 296, no. 2, pp. 115–117, Aug. 2020, doi: 10.1148/radiol.2020200432.
  17. T. Ai et al., “Correlation of Chest CT and RT-PCR Testing for coronavirus disease 2019 (COVID-19) in China: A report of 1014 Cases”, Radiology, vol. 296, no. 2, pp. E32–E40, Aug. 2020, doi: 10.1148/radiol.2020200642.
  18. M.-Y. Ng et al., “Imaging Profile of the COVID-19 Infection: radiologic findings and literature review”, Radiology: Cardiothoracic Imaging, vol. 2, no. 1, p. e200034, Feb. 2020, doi: 10.1148/ryct.2020200034.
  19. C. Huang et al., “Clinical features of patients infected with 2019 novel coronavirus in Wuhan, China”, Lancet, vol. 395, no.10223, pp. 497–506, Feb. 2020, doi: 10.1016/S0140-6736(20)30183-5.
  20. M. Cellina, M. Orsi, T. Toluian, C. Valenti Pittino, and G. Oliva, “False negative chest X-Rays in patients affected by COVID-19 pneumonia and corresponding chest CT findings”, Radiography, vol. 26, no. 3, pp. e189–e194, Aug. 2020, doi: 10.1016/j.radi.2020.04.017.
  21. A. England, E. Littler, S. Romani, and P. Cosson, “Modifications to mobile chest radiography technique during the COVID-19 pandemic – implications of X-raying through side room windows”, Radiography, Available online 3 August 2020. vol. 27, no. 1, pp. 193–199, Feb. 2021, doi: 10.1016/j.radi.2020.07.015.
  22. S. Minaee, R. Kafieh, M. Sonka, S. Yazdani, and G. J. Soufi, “Deep-COVID: Predicting COVID-19 from chest X-ray images using deep transfer learning”, Med. Image Anal., vol. 65, p. 101794, Oct. 2020, doi: 10.1016/j.media.2020.101794.
  23. T. Mahmud, M.A. Rahman, S.A. Fattah, “CovXNet: A multi-dilation convolutional neural network for automatic COVID-19 and other pneumonia detection from chest X-ray images with transferable multi-receptive feature optimization”, Comput. in Biol. and Medicine, vol. 122, p. 103869, Jul. 2020, doi: 10.1016/j.compbiomed.2020.103869.
  24. H. Panwar, P. K. Gupta, M. K. Siddiqui, R. Morales-Menendez, P. Bhardwaj, and V. Singh, “a deep learning and Grad-CAM based color visualization approach for fast detection of COVID-19 cases using chest X-ray and CT-scan images”, Chaos Solit. Fractals, vol. 140, p. 110190, Nov. 2020, doi: 10.1016/j.chaos.2020.110190.
  25. J. P. Cohen and P. Morrison and L. Dao, “COVID-19 image data collection”, arXiv:2003.11597, 2020, Available: https://github.com/ieee8023/covid-chestxray-dataset . (Accessed: 20.09.2021)
  26. J. P. Cohen et al., “COVID-19 image data collection: prospective predictions are the future”, arXiv:2006.11988, 2020. Available: https://github.com/ieee8023/covid-chestxray-dataset . (Accessed: 20.09.2021)
  27. CheXpert: A large chest X-Ray dataset and competition. Available: https://stanfordmlgroup.github.io/competitions/chexpert/ (Accessed: 20.09.2021)
  28. NIH. Open access biomedical image search engine. Available: https://openi.nlm.nih.gov/ (Accessed: 20.09.2021)
  29. Y. LeCun et al., “Backpropagation applied to handwritten zip code recognition”, Neural Comput., vol. 1, no. 4, pp. 541–551, Dec. 1989, doi:10.1162/neco.1989.1.4.541.
  30. Neural network architectures. Available: https://habr.com/ru/company/nix/blog/430524/ (Accessed: 20.09.2021)
  31. Convolutional network for image feature extraction VGG16. Available: https://neurohive.io/ru/vidy-nejrosetej/vgg16-model/ . (Accessed: 20.09.2021)
  32. K. He, X. Zhang, S. Ren, and J. Sun, “Deep residual learning for image recognition”, arXiv:1512.03385, 2015.
  33. K. He, X. Zhang, S. Ren, and J. Sun, “Deep residual learning for image recognition”, Proceedings of the 2016 IEEE Conf. on Comput. Vision and Pattern Recognition, pp. 770-778, Jun. 2016, doi:10.1109/CVPR.2016.90.
  34. ResNet (34,50,101): “Residual” CNN for image classification. Available: https://neurohive.io/ru/vidy-nejrosetej/ resnet-34-50-101/ (Accessed: 20.09.2021)
  35. A. Veit, M. Wilber, and S. Belongie, “Residual networks behave like ensembles of relatively shallow networks”, NIPS’16: Proceedings of the 30th International Conference on Neural Information Processing Systems, pp. 550–558, 2016.
  36. I. Goodfellow, Y. Bengio, and A. Courville, Deep Learning. MIT Press, Cambridge, MA, USA, 2017.
  37. B. Ramsundar and R. B. Zadeh, TensorFlow for deep learning: from linear regression to reinforcement learning. O’Reilly Media, Sebastopol, CA, USA, 2018.
  38. T. Hope, Y. S. Resheff, and I. Lieder, Learning Tensorflow: A guide to building deep learning systems. O’Reilly Media, Sebastopol, CA, USA, 2017.
  39. A. Gulli, A. Kapoor, and S. Pal, Deep learning with TensorFlow 2 and Keras, 2nd ed. Packt Publishing, Birmingham – Mumbai, 2019.
  40. N. Shukla, Machine learning with TensorFlow. Manning Publications, NY, USA, 2018.
  41. A. Geron, Hands-on machine learning with Scikit-Learn and TensorFlow. O’Reilly Media, Sebastopol, CA, USA, 2017.
  42. TensorFlow documentation. Available: www.tensorflow.org (Accessed: 20.09.2021).
  43. Keras documentation. Available: https://keras.io/guides/ (Accessed: 20.09.2021).
§ 06 — Related

Similar articles in this journal

Related peer-reviewed studies published in this journal.
View all issues

AN APPROACH TO CONSTRUCTING AN INFORMATIVE FEATURE SPACE OF TECHNICAL INDICATORS FOR FINANCIAL PROCESSES BASED ON NEURAL NETWORK MODELS

Roman Tertychnyi, Denys Chuprin, Petro Bidyuk

Background. Financial time series are characterized by nonlinearity, non-stationarity, and a high level of noise, which significantly complicates their modeling and…

#informative feature space construction approach #technical indicators #Parkinson volatility #neural networks
10.20535/kpisn.2026.3.359610

Most read articles by the same author(s)

Similar Articles

1-10 of 53

You may also start an advanced similarity search for this article.