NeMoc-nCoV: A Comprehensive Study on a Customised Self-Attention Deep Convolutional Neural Network for the Identification of Diverse Morbidity Patterns
DOI:
https://doi.org/10.52783/jns.v14.2661Keywords:
N/AAbstract
The global outbreak of the highly contagious coronavirus disease, known as COVID-19, has had a profound impact on a significant portion of the global population, affecting millions of individuals across various regions. The rapid growth and increasing numbers of illnesses pose significant challenges for medical experts in promptly detecting and containing the spread of the disease. Medical image analysis is a burgeoning field of study that offers a promising avenue for addressing the aforementioned issue with greater precision and rigor. This research work presents the implementation of an image processing system that leverages deep learning techniques and neural networks for the purpose of predicting the 2019-nCoV virus, bacterial pneumonia, viral pneumonia, and pleurisy using chest radiograph images. This research paper proposes the utilization of convolutional neural networks, deep learning, and machine learning techniques for the purpose of distinguishing between COVID-19 positive, bacterial pneumonia, viral pneumonia, pleurisy and healthy patients based on chest radiography images. In this study, we present a novel approach for effectively handling the intricate structural complexity of images. Our proposed method involves the utilization of a neural network architecture that combines features extracted from two state-of-the-art convolutional neural networks, namely ZFNet and VGG -16 Net. By leveraging the strengths of these networks, we aim to enhance the overall performance of our model in managing the complex structural characteristics of images. In order to evaluate the performance of our network in real-world scenarios, a comprehensive testing was conducted on a dataset consisting of 7940 images. The purpose of this evaluation was to assess the network's effectiveness and efficiency in handling various real-world situations. By subjecting the network to this extensive testing, we aimed to gain insights into its capabilities and limitations, thereby enabling us to make informed decisions regarding its deployment and potential improvements. In this study, we present a novel network architecture designed for the detection of normal, bacterial pneumonia, viral pneumonia, pleurisy and COVID-19 cases. The network demonstrates a commendable average accuracy of 95%, making it a promising tool for assisting radiologists in their diagnostic processes. The detection of COVID-19 cases has become a critical task in the field of radiology, given the rapid spread and severity of the disease. Traditional diagnostic methods often rely on visual inspection of radiographic images, which can be time-consuming and prone to human error. Therefore, the development of automated systems that can accurately identify different cases is of utmost importance. Our proposed network leverages advanced machine learning techniques to effectively distinguish between normal, bacterial pneumonia, viral pneumonia, pleurisy and COVID-19 cases. By training the network on a large dataset of radiographic images, we were able to optimize its performance.
Downloads
Metrics
References
S. Militante, N. Dionisio and B. Sibbaluca, "Pneumonia Detection through Adaptive Deep Learning Models of Convolutional Neural Networks".
S. V. Militante and B. G. Sibbaluca, "Pneumonia Detection Using Convolutional Neural Networks", International Journal of Scientific & Technology Research, vol. 9, no. 04, pp. 1332-1337, April 2020.
Jiang F, Deng L, Zhang L, Cai Y, Cheung CW, Xia Z. Review of the clinical characteristics of Coronavirus disease 2019 (COVID-19). J Gen Intern Med [Internet]. 2020;35(5):1545–9.
Word Health Organization [Internet]. Word Health Organization. Available from: https://www.who.in
Sun D, Li H, Lu X-X, Xiao H, Ren J, Zhang F-R, et al. Clinical features of severe pediatric patients with coronavirus disease 2019 in Wuhan: a single center’s observational study. World J Pediatr [Internet]. 2020;16(3):251–9.
An P, Chen H, Ren H, Su J, Ji M, Kang J, et al. Gastrointestinal symptoms onset in COVID-19 patients in Wuhan, China. Dig Dis Sci [Internet]. 2021;66(10):3578–87.
Harsh Panwar PK, Gupta MK, Siddiqui R, Morales-Menendez V. Application of deep learning for fast detection of COVID-19 in X-Rays using nCOVnet, Chaos. Chaos, Solitons & Fractals. 2020;138.
Chakraborty M, Dhavale SV, Ingole J. Corona-Nidaan: lightweight deep convolutional neural network for chest X-Ray based COVID-19 infection detection. Appl Intell [Internet]. 2021;51(5):3026–43.
Wang S, Kang B, Ma J, Zeng X, Xiao M, Guo J, et al. A deep learning algorithm using CT images to screen for Corona virus disease (COVID-19). Eur Radiol [Internet]. 2021;31(8):6096–104.
Gozes O, Frid-Adar M, Greenspan H, Browning PD, Zhang H, Ji W, et al. Rapid ai development cycle for the coronavirus (covid-19) pandemic: initial results for automated detection & patient monitoring using deep learning ct image analysis.2003.05037. 2003.
Wang L, Wong A. Covid-net: a tailored deep convolutional neural network design for detection of covid-19 cases from chest radiography images. 2003.
Narin A, Kaya C, Pamuk Z. Automatic detection of coron- avirus disease (covid-19) using x-ray images and deep convolutional neural networks. 2003.
Li L, Qin L, Xu Z, Yin Y, Wang X, Kong B, et al. Artificial intelligence distinguishes covid-19 from community acquired pneumonia on chest ct. Radiology. 2020.
Downloads
Published
How to Cite
Issue
Section
License

This work is licensed under a Creative Commons Attribution 4.0 International License.
You are free to:
- Share — copy and redistribute the material in any medium or format
- Adapt — remix, transform, and build upon the material for any purpose, even commercially.
Terms:
- Attribution — You must give appropriate credit, provide a link to the license, and indicate if changes were made. You may do so in any reasonable manner, but not in any way that suggests the licensor endorses you or your use.
- No additional restrictions — You may not apply legal terms or technological measures that legally restrict others from doing anything the license permits.