Your browser doesn't support javascript.
Mostrar: 20 | 50 | 100
Resultados 1 - 1 de 1
Filtrar
Adicionar filtros

Base de dados
Ano de publicação
Tipo de documento
Intervalo de ano
1.
2nd International Conference on Intelligent and Cloud Computing, ICICC 2021 ; 286:455-462, 2022.
Artigo em Inglês | Scopus | ID: covidwho-1826298

RESUMO

COVID-19 turned into a critical health problem around the world. Since the start of its spreading, numerous Artificial Intelligence-based models have been created for foreseeing the conduct of the infection and recognizing its contamination. One of the efficient methods of deciding the COVID-19, pneumonia disease is through the chest X-ray images analysis. As there are lots of patients in emergency clinical conditions, it would be tedious and difficult to analyze loads of X-ray images physically. So, an automated, AI-based system can be helpful to predict the infection due to COVID-19 in less time. In this study, a Modified Convolution Neural Network (CNN) model is suggested to predict the COVID-19 infections from the chest X-ray images. Proposed model is designed based on the state-of-art models like GoogleNet, U-Net, VGGNet. The model is fine-tuned using less number of layers than the existing model to get acceptable accuracy. The model is implemented on 724 chest X-ray images from COVID-19 image data collection and is able to produce 93.5% accuracy, 93.0% precision, 93.5% recall, and 92.5% F1-Score, respectively. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

SELEÇÃO DE REFERÊNCIAS
DETALHE DA PESQUISA