Improved YOLOv8n model for retinal macular degeneration detection based on multi-scale feature fusion

Wenhao Wei; Yahong Li; Yu Tang; Xufen Xie; Kexian Li; Yafang Song

doi:10.1117/12.3015652

29 December 2023 Improved YOLOv8n model for retinal macular degeneration detection based on multi-scale feature fusion

Wenhao Wei, Yahong Li, Yu Tang, Xufen Xie, Kexian Li, Yafang Song

Proceedings Volume 12975, 6th Optics Young Scientist Summit (OYSS 2023); 129750P (2023) https://doi.org/10.1117/12.3015652
Event: 6th Optics Young Scientist Summit, 2023, Changsha, China

Abstract

To solve the problem that the type and area of macular disease are not easy to identify due to the irregular scale and unobvious characteristics of macular disease region on retinal OCT images, an improved YOLOv8n macular disease detection model is proposed, and a data set of retinal macular disease detection is established. Firstly, the feature pyramid module of bidirectional weighted feature fusion was added. Secondly, the attention mechanism was introduced. Finally, the novel loss function was replaced. The improved model can complete the multi-scale and irregular multi-objective training task of retinal maculopathy. The experimental results show that the improved model has a good effect on the self-built data set. The accuracy of central serous macular degeneration, macular hole, and choroidal neovasculation can reach 97.7%, 97.8%, and 97.4%, respectively, and can accurately identify the location of the lesion.

(2023) Published by SPIE. Downloading of the abstract is permitted for personal use only.

Citation Download Citation

Wenhao Wei, Yahong Li, Yu Tang, Xufen Xie, Kexian Li, and Yafang Song "Improved YOLOv8n model for retinal macular degeneration detection based on multi-scale feature fusion", Proc. SPIE 12975, 6th Optics Young Scientist Summit (OYSS 2023), 129750P (29 December 2023); https://doi.org/10.1117/12.3015652

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available