Axial super-resolution optical coherence tomography imaging by complex-valued enhanced deep residual network

Lingyun Wang; Shaohua Pi; Jianhua Mo

doi:10.1117/12.3005634

12 March 2024 Axial super-resolution optical coherence tomography imaging by complex-valued enhanced deep residual network

Lingyun Wang, Shaohua Pi, Jianhua Mo

Author Affiliations +

Proceedings Volume 12830, Optical Coherence Tomography and Coherence Domain Optical Methods in Biomedicine XXVIII; 128300G (2024) https://doi.org/10.1117/12.3005634
Event: SPIE BiOS, 2024, San Francisco, California, United States

Conference Poster

Abstract

Optical coherence tomography (OCT) has been widely used in ophthalmology with its micron-resolution, depth-resolving capability in imaging bio-tissues in vivo. Recently, deep learning methods are emerging to achieve axial super-resolution (SR) in OCT, aimed to reduce the cost of broad-band light source. However, all of those deep learning methods were developed based on real-valued networks, ignoring the phase information of complex-valued OCT image which contains structural information. In this study, we proposed a complex-valued enhanced deep super-resolution network (Cv-EDSR) to obtain OCT axial super-resolution. We validated the superior performance of Cv-EDSR over the traditional EDSR on two datasets (swine esophagus and human retina), and demonstrated three benefits of Cv-EDSR: a) Cv-EDSR generated more realistic SR images, b) Cv-EDSR achieved an improved quality of SR images, c) Cv-EDSR possessed a better generalization performance.

Citation Download Citation

Lingyun Wang, Shaohua Pi, and Jianhua Mo "Axial super-resolution optical coherence tomography imaging by complex-valued enhanced deep residual network", Proc. SPIE 12830, Optical Coherence Tomography and Coherence Domain Optical Methods in Biomedicine XXVIII, 128300G (12 March 2024); https://doi.org/10.1117/12.3005634

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