Explainable ensemble learning method for OCT detection with transfer learning
PLOS ONE · 22.03.2024 · 19 (3) · e0296175
DOI: 10.1371/journal.pone.0296175 ↗Kurz gefasst
Fachartikel in der Zeitschrift PLOS ONE, online erschienen am 22.03.2024. Verfasst von Jiasheng Yang und 4 weiteren Autor:innen. Beteiligte Einrichtung: Academician Workstation, Changsha Medical University, Changsha, Hunan, China. Hauptthemen laut MEDLINE: Macular Edema, Diabetic Retinopathy, Deep Learning. Die Arbeit wurde bisher 60-mal zitiert (Crossref) und nennt 49 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: The Hunan Key Laboratory of the Research and Development of Novel Pharmaceutical Preparations, The Hunan Provincial Key Laboratory of the TCM Agricultural Biogenomics, Jiangsu Key Laboratory of Regional Resource Exploitation and Medicinal Research.
Abstract (Originalsprache)
The accuracy and interpretability of artificial intelligence (AI) are crucial for the advancement of optical coherence tomography (OCT) image detection, as it can greatly reduce the manual labor required by clinicians. By prioritizing these aspects during development and application, we can make significant progress towards streamlining the clinical workflow. In this paper, we propose an explainable ensemble approach that utilizes transfer learning to detect fundus lesion diseases through OCT imaging. Our study utilized a publicly available OCT dataset consisting of normal subjects, patients with dry age-related macular degeneration (AMD), and patients with diabetic macular edema (DME), each with 15 samples. The impact of pre-trained weights on the performance of individual networks was first compared, and then these networks were ensemble using majority soft polling. Finally, the features learned by the networks were visualized using Grad-CAM and CAM. The use of pre-trained ImageNet weights improved the performance from 68.17% to 92.89%. The ensemble model consisting of the three CNN models with pre-trained parameters loaded performed best, correctly distinguishing between AMD patients, DME patients and normal subjects 100% of the time. Visualization results showed that Grad-CAM could display the lesion area more accurately. It is demonstrated that the proposed approach could have good performance of both accuracy and interpretability in retinal OCT image detection.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Autor:innen und Institutionen
- Academician Workstation, Changsha Medical University, Changsha, Hunan, China.
- School of Mathematics and Statistics, Nanjing University of Information Science and Technology, Nanjing, Jiangsu, China.
- Geneis Beijing Co. Ltd., Beijing, China.
- School of International Education, Anhui University of Technology, Maanshan, Anhui, China.
ORCID-Profile: Jiasheng Yang
Förderung
- The Hunan Key Laboratory of the Research and Development of Novel Pharmaceutical Preparations
- The Hunan Provincial Key Laboratory of the TCM Agricultural Biogenomics
- Jiangsu Key Laboratory of Regional Resource Exploitation and Medicinal Research
Zitieren
APA
Yang, J., Wang, G., Xiao, X., Bao, M., & Tian, G. (2024). Explainable ensemble learning method for OCT detection with transfer learning. PLOS ONE, 19(3), e0296175. https://doi.org/10.1371/journal.pone.0296175
Vancouver
Yang J, Wang G, Xiao X, Bao M, Tian G. Explainable ensemble learning method for OCT detection with transfer learning. PLOS ONE. 2024;19(3):e0296175. doi:10.1371/journal.pone.0296175
BibTeX
@article{Yang2024,
author = {Yang, Jiasheng and Wang, Guanfang and Xiao, Xu and Bao, Meihua and Tian, Geng},
title = {Explainable ensemble learning method for OCT detection with transfer learning},
journal = {PLOS ONE},
year = {2024},
volume = {19},
number = {3},
pages = {e0296175},
doi = {10.1371/journal.pone.0296175}
}
EUVIMED