COVID-19 Chest X-Ray Classification Using Compact Convolutional Transformer

Xin Hui Tan, Jit Yan Lim, Kian Ming Lim, Chin Poo Lee

Research output: Chapter in Book/Conference proceedingConference contributionpeer-review

Abstract

The outbreak of Covid-19 in 2019 had a significant impact worldwide, causing long-term breathing problems in many affected individuals. Some people may experience white spots on their lungs after recovering from Covid-19, which can be difficult to identify. One promising approach for identifying abnormal lungs is through image classification. In this work, we utilize three datasets for image classification: the COVID-19 Radiography Dataset, the Chest X-ray Dataset, and the COVID-19 Dataset. To achieve accurate classification, a pre-trained Compact Convolution Transformer (CCT) has been utilized with transfer learning. Our results show that the COVID-19 Radiography Dataset achieved an accuracy of 89.28%, the Chest X-ray Dataset achieved 95.11% accuracy, and the COVID-19 X-ray Dataset achieved an impressive 97.50% accuracy. These findings demonstrate the potential of using image classification to identify abnormal lungs and pave the way for further research in this area.

Original languageEnglish
Title of host publication2023 11th International Conference on Information and Communication Technology, ICoICT 2023
PublisherIEEE Computer Society
Pages266-270
Number of pages5
ISBN (Electronic)9798350321982
DOIs
Publication statusPublished - 2023
Externally publishedYes
Event11th International Conference on Information and Communication Technology, ICoICT 2023 - Melaka, Malaysia
Duration: 23 Aug 202324 Aug 2023

Publication series

NameInternational Conference on ICT Convergence
Volume2023-August
ISSN (Print)2162-1233
ISSN (Electronic)2162-1241

Conference

Conference11th International Conference on Information and Communication Technology, ICoICT 2023
Country/TerritoryMalaysia
CityMelaka
Period23/08/2324/08/23

Keywords

  • CCT
  • Chest X-Ray
  • Compact Convolution Transformer
  • Covid-19
  • CXR

ASJC Scopus subject areas

  • Information Systems
  • Computer Networks and Communications

Fingerprint

Dive into the research topics of 'COVID-19 Chest X-Ray Classification Using Compact Convolutional Transformer'. Together they form a unique fingerprint.

Cite this