A fusing Transformer and CNN on Interpretable COVID-19 Detection

DOI

10.2991/978-94-6463-264-4_46How to use a DOI?

Keywords

COVID-19; CNN; Transformer; Lung segmentation; Transfer Learning

Abstract

Although computer-aided diagnosis has become an important tool for rapid detection of lung diseases, the reliability of algorithm visualization on chest X-ray (CXR) images remains a challenge. This study explores the detection performance of a fusion model combining Transformer and CNN models. A decision constraint module was designed to achieve interpretable pneumonia detection. The performance of the decision constraint module was observed using Grad-CAM technique, and experimental results demonstrate that it outperforms lung mask segmentation. By activating transfer learning, our parallel combination model effectively identifies COVID-19 categories with a test set accuracy of 98.65%.

Copyright

© 2024 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Zhuohui Pan
AU  - Yujuan Chen
PY  - 2023
DA  - 2023/09/28
TI  - A fusing Transformer and CNN on Interpretable COVID-19 Detection
BT  - Proceedings of the 2023 3rd International Conference on Education, Information Management and Service Science (EIMSS 2023)
PB  - Atlantis Press
SP  - 410
EP  - 419
SN  - 2589-4900
UR  - https://doi.org/10.2991/978-94-6463-264-4_46
DO  - 10.2991/978-94-6463-264-4_46
ID  - Pan2023
ER  -