Optimize transfer learning for lung diseases in bronchoscopy using a new concept: Sequential fine-tuning

Tan, Tao; Li, Zhang; Liu, Haixia; Zanjani, Farhad G.; Ouyang, Quchang; Tang, Yuling; Hu, Zheyu; Li, Qiang

Abstract

Bronchoscopy inspection, as a follow-up procedure next to the radiological imaging, plays a key role in the diagnosis and treatment design for lung disease patients. When performing bronchoscopy, doctors have to make a decision immediately whether to perform a biopsy. Because biopsies may cause uncontrollable and life-threatening bleeding of the lung tissue, thus doctors need to be selective with biopsies. In this paper, to help doctors to be more selective on biopsies and provide a second opinion on diagnosis, we propose a computer-aided diagnosis (CAD) system for lung diseases, including cancers and tuberculosis (TB). Based on transfer learning (TL), we propose a novel TL method on the top of DenseNet: sequential fine-tuning (SFT). Compared with traditional fine-tuning (FT) methods, our method achieves the best performance. In a data set of recruited 81 normal cases, 76 TB cases and 277 lung cancer cases, SFT provided an overall accuracy of 82% while other traditional TL methods achieved an accuracy from 70% to 74%. The detection accuracy of SFT for cancers, TB, and normal cases are 87%, 54%, and 91%, respectively. This indicates that the CAD system has the potential to improve lung disease diagnosis accuracy in bronchoscopy and it may be used to be more selective with biopsies.

• Journal Article Type: Article
• Online Publication Date: Aug 16, 2018
• Publication Date: Aug 16, 2018
• Deposit Date: Jan 16, 2023
• Journal: IEEE Journal of Translational Engineering in Health and Medicine
• Electronic ISSN: 2168-2372
• Publisher: Institute of Electrical and Electronics Engineers (IEEE)
• Peer Reviewed: Peer Reviewed
• Volume: 6
• Pages: 1-8
• DOI: https://doi.org/10.1109/JTEHM.2018.2865787
• Keywords: Biomedical Engineering; General Medicine
• Public URL: https://uwe-repository.worktribe.com/output/10339150
• Publisher URL: https://ieeexplore.ieee.org/document/8438558