Classification of Covid-19 Infection Based on Chest X-ray Pictures Using OpenCv and Convolution Neural Networks
0
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
COVID-19 is a pathogen called SARS-CoV-2, an RNA virus that can infect various animals, including humans. As COVID-19 spread globally, the World Health Organization upgraded it to a pandemic in March 2020. In addition to solving the problem of shortage of medical personnel, rapid and accurate classification of infected patients emerged as an important issue. Therefore, we propose a deep learning-based chest X-ray image reading model that can notify the doctor whether the patient is infected. The goal is to achieve multiclass classification, which not only classifies COVID-19 infections, but also other lung diseases to help the medical community. The proposed method is a combination model. It involves pre-processing the chest X-ray image using the image augmentation method and various convolutional neural network (CNN) models. The purpose of the proposed method is to classify COVID-19, normal people, and viral pneumonia appropriately. Overall, 15,153 X-ray images were used in the study. By using the proposed method, we obtained a model with high accuracy through improved image data. Characteristically, some models tend to detect COVID-19 and pneumonia properly. Finally, an ensemble model was created using models made by the proposed method. Eventually, we obtained a high accuracy (0.981) model for detecting infections appropriately.
Abstract
COVID-19 is a pathogen called SARS-CoV-2, an RNA virus that can infect various animals, including humans. As COVID-19 spread globally, the World Health Organization upgraded it to a pandemic in March 2020. In addition to solving the problem of shortage of medical personnel, rapid and accurate classification of infected patients emerged as an important issue. Therefore, we propose a deep learning-based chest X-ray image reading model that can notify the doctor whether the patient is infected. The goal is to achieve multiclass classification, which not only classifies COVID-19 infections, but also other lung diseases to help the medical community. The proposed method is a combination model. It involves pre-processing the chest X-ray image using the image augmentation method and various convolutional neural network (CNN) models. The purpose of the proposed method is to classify COVID-19, normal people, and viral pneumonia appropriately. Overall, 15,153 X-ray images were used in the study. By using the proposed method, we obtained a model with high accuracy through improved image data. Characteristically, some models tend to detect COVID-19 and pneumonia properly. Finally, an ensemble model was created using models made by the proposed method. Eventually, we obtained a high accuracy (0.981) model for detecting infections appropriately.
참고문헌 (20)
- "Wuhan Municipal Health Commission’s briefing on the current situation of pneumonia in our city" (2022)[URL]
- "Coronavirus disease (COVID-19)" (2022)[URL]
- "Timeline: WHO’s COVID-19 response" (2022)[URL]
- "Chest X-ray" (2022)[URL]
- Nkwentsha X, "Automatic classification of medical X-ray images with convolutional neural networks", International SAUPEC, PRASA Conference IEEE, pp.2020[DOI]
- 안경희, "딥 러닝 기반 코로나19 흉부 X선 판독 기법", 문화기술의 융합, 6(4), pp.789~795 (2020), 국제문화기술진흥원
- Zhang H, "Stackgan++ : realistic image synthesis with stacked generative adversarial networks", IEEE TPAMI, 41, pp.1947~1962 (2018)
- Dcunha M, "Pneumonia detection in X-rays using OpenCV and deep learning", IEEE Int’l Conf. on Communication information and Computing Technology, pp.1~6 (2021)
- Chowdhury MEH, "Can AI help in screening viral and COVID-19 pneumonia", IEEE Access, 8, pp.132665~132676 (2020)
- Rahman T, "Exploring the effect of image enhancement techniques on COVID-19 detection using Chest X-ray images", Comput Biol Med, 132, pp.1~34 (2020)
- Wu H, "Predict pneumonia with Chest X-ray images based on convolutional deep neural learning networks", J Intell Fuzzy Syst, 39, pp.2893~2907 (2020)
- 이용학, "Integrated Analytic Methodology Using Visual Image and Meta-Data for Product Recommendation", Quantitative Bio-Science, 41(1), pp.27~35 (2022), 자연과학연구소[DOI]
- Huang G, "Densely connected convolutional networks", IEEE Int’l Conf. on Computer Vision and Pattern Recognition, pp.4700~4708 (2017)[DOI]
- Simonyan K, "Very deep convolutional networks for large-scale image recognition", arXiv preprint arXiv:1409.1556, ~ (2015)
- Tan M, "Efficientnet : Rethinking model scaling for convolutional neural networks", arXiv preprint arXiv:1905.11946;6105-6114, ~ (2019)
- He K, "Deep residual learning for image recognition", IEEE Int’l Conf. on Computer Vision and Pattern Recognition, pp.770~778 (2016)
- Krizhevsky A, "ImageNet classification with deep convolutional neural networks", Commun ACM, 60, pp.84~90 (2017)
- "Basic concepts of ensemble" (2022)[URL]
- Zummel N, "Practical data science with R" (2014), Manning Publishing Company
- "Cohen’s kappa" (2022)[URL]