Your browser doesn't support javascript.
BND-VGG-19: A deep learning algorithm for COVID-19 identification utilizing X-ray images.
Cao, Zili; Huang, Junjian; He, Xing; Zong, Zhaowen.
  • Cao Z; Chongqing Key Laboratory of Nonlinear Circuits and Intelligent Information Processing, College of Electronic and Information Engineering, Southwest University, Chongqing 400715, PR China.
  • Huang J; Chongqing Key Laboratory of Nonlinear Circuits and Intelligent Information Processing, College of Electronic and Information Engineering, Southwest University, Chongqing 400715, PR China.
  • He X; Chongqing Key Laboratory of Nonlinear Circuits and Intelligent Information Processing, College of Electronic and Information Engineering, Southwest University, Chongqing 400715, PR China.
  • Zong Z; Department of Training Base for Health Care, Army Medical University, Chongqing 400038, PR China.
Knowl Based Syst ; 258: 110040, 2022 Dec 22.
Article in English | MEDLINE | ID: covidwho-2082723
ABSTRACT
During the past two years, a highly infectious virus known as COVID-19 has been damaging and harming the health of people all over the world. Simultaneously, the number of patients is rising in various countries, with many new cases appearing daily, posing a significant challenge to hospital medical staff. It is necessary to improve the efficiency of virus detection. To this end, we combine modern technology and visual assistance to detect COVID-19. Based on the above facts, for accurate and rapid identification of infected persons, the BND-VGG-19 method was proposed. This method is based on VGG-19 and further incorporates batch normalization and dropout layers between the layers to improve network accuracy. Then, the COVID-19 dataset including viral pneumonia, COVID-19, and normal X-ray images, are used to diagnose lung abnormalities and test the performance of the proposed algorithm. The experimental results show the superiority of BND-VGG-19 with a 95.48% accuracy rate compared with existing COVID-19 diagnostic methods.
Keywords

Full text: Available Collection: International databases Database: MEDLINE Language: English Journal: Knowl Based Syst Year: 2022 Document Type: Article

Similar

MEDLINE

...
LILACS

LIS


Full text: Available Collection: International databases Database: MEDLINE Language: English Journal: Knowl Based Syst Year: 2022 Document Type: Article