We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
Deep Convolutional Neural Network With Image Processing Techniques And Resnet252v2 For Detection Of Covid19 From X-Ray Images.
- Authors
D., Kavitha Rajalakshmi.; Bharathisindhu
- Abstract
The 2019 coronavirus disease, or SARS-CoV-2, is a rapidly spreading viral infection that has affected millions of people around the world. Due to its rapid spread and increasing numbers, it becomes overwhelming for health professionals to quickly diagnose the disease and prevent its spread. Therefore, automation of the diagnostic procedure has become essential. This improves work efficiency and keeps healthcare workers protected from exposure to viruses. Medical image analysis is one of the emerging fields of research where this problem can be addressed even more precisely. This paper presents the prediction of SARS-CoV-2 using chest roentgen rays images and the implementation of an image processing system using deep learning and neural networks. This study presents a methodology that utilises deep learning techniques, including machine learning and convolutional neural networks, to predict the presence of SARS-CoV-2 infection in patients based on the analysis of chest radiographs. The utilisation of deep architectures is essential in the categorization of roentgen rays images due to the intricate nature of their structures. In light of this, we present a neural network including a sequential arrangement of ResNet50 and Res Net152V2 networks. The network demonstrated superior accuracy by employing various features derived from two robust networks. In order to assess the performance of our network, a comprehensive evaluation was conducted on a dataset consisting of 15,602 images. This evaluation aimed to determine the accuracy achieved by the network under real-world conditions. The network under consideration has an average accuracy of 93% in detecting both SARS-CoV-2 and normal cases, rendering it a potentially valuable tool inside the radiology department.
- Subjects
X-ray imaging; CONVOLUTIONAL neural networks; IMAGE processing; COVID-19 testing; DEEP learning; MACHINE learning
- Publication
Mapana Journal of Sciences, 2023, Vol 22, p129
- ISSN
0975-3303
- Publication type
Article
- DOI
10.12723/mjs.sp2.8