Self-fusion for OCT Noise Reduction
Overview
Affiliations
Reducing speckle noise is an important task for improving visual and automated assessment of retinal OCT images. Traditional image/signal processing methods only offer moderate speckle reduction; deep learning methods can be more effective but require substantial training data, which may not be readily available. We present a novel self-fusion method that offers effective speckle reduction comparable to deep learning methods, but without any external training data. We present qualitative and quantitative results in a variety of datasets from fovea and optic nerve head regions, with varying SNR values for input images.
Probabilistic volumetric speckle suppression in OCT using deep learning.
Chintada B, Ruiz-Lopera S, Restrepo R, Bouma B, Villiger M, Uribe-Patarroyo N Biomed Opt Express. 2024; 15(8):4453-4469.
PMID: 39346991 PMC: 11427188. DOI: 10.1364/BOE.523716.
Recurrent Self Fusion: Iterative Denoising for Consistent Retinal OCT Segmentation.
Wei S, Liu Y, Bian Z, Wang Y, Zuo L, Calabresi P Ophthalmic Med Image Anal (2023). 2024; 14096:42-51.
PMID: 38318463 PMC: 10840975. DOI: 10.1007/978-3-031-44013-7_5.
Probabilistic volumetric speckle suppression in OCT using deep learning.
Chintada B, Ruiz-Lopera S, Restrepo R, Bouma B, Villiger M, Uribe-Patarroyo N ArXiv. 2023; .
PMID: 38106457 PMC: 10723542.
Rico-Jimenez J, Jovanovic J, Nolen S, Malone J, Rao G, Levine E Front Ophthalmol (Lausanne). 2023; 3.
PMID: 37275441 PMC: 10238074. DOI: 10.3389/fopht.2023.1141070.
Retinal OCT Denoising with Pseudo-Multimodal Fusion Network.
Hu D, Malone J, Atay Y, Tao Y, Oguz I Ophthalmic Med Image Anal (2020). 2022; 12069:125-135.
PMID: 35775870 PMC: 9241435. DOI: 10.1007/978-3-030-63419-3_13.