Retinal layer segmentation in rodent OCT images: Local intensity profiles & fully convolutional neural networks

•Two approaches to detect retinal layers in rat OCT images are presented.•One approach is based on classical image processing and the other on deep learning.•Both methods significantly outperform a commercial image segmentation software.•Deep-learning-based method obtains promising results on unseen...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Computer methods and programs in biomedicine Ročník 198; s. 105788
Hlavní autoři: Morales, Sandra, Colomer, Adrián, Mossi, José M., del Amor, Rocío, Woldbye, David, Klemp, Kristian, Larsen, Michael, Naranjo, Valery
Médium: Journal Article
Jazyk:angličtina
Vydáno: Ireland Elsevier B.V 01.01.2021
Témata:
Animals
Convolutional neural networks
Intensity profile
Layer segmentation
Neural Networks, Computer
Optical coherence tomography
Rat OCT
Rats
Retina - diagnostic imaging
Rodent OCT
Rodentia
Software
Tomography, Optical Coherence
Rodent OCT
Intensity profile
Rat OCT
Optical coherence tomography
Convolutional neural networks
Layer segmentation
ISSN:0169-2607, 1872-7565, 1872-7565
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:•Two approaches to detect retinal layers in rat OCT images are presented.•One approach is based on classical image processing and the other on deep learning.•Both methods significantly outperform a commercial image segmentation software.•Deep-learning-based method obtains promising results on unseen databases.•A new rat OCT database with expert-reviewed segmentations is publicly available. Background and Objective: Optical coherence tomography (OCT) is a useful technique to monitor retinal layer state both in humans and animal models. Automated OCT analysis in rats is of great relevance to study possible toxic effect of drugs and other treatments before human trials. In this paper, two different approaches to detect the most significant retinal layers in a rat OCT image are presented. Methods: One approach is based on a combination of local horizontal intensity profiles along with a new proposed variant of watershed transformation and the other is built upon an encoder-decoder convolutional network architecture. Results: After a wide validation, an averaged absolute distance error of 3.77 ± 2.59 and 1.90 ± 0.91 µm is achieved by both approaches, respectively, on a batch of the rat OCT database. After a second test of the deep-learning-based method using an unseen batch of the database, an averaged absolute distance error of 2.67 ± 1.25 µm is obtained. The rat OCT database used in this paper is made publicly available to facilitate further comparisons. Conclusions: Based on the obtained results, it was demonstrated the competitiveness of the first approach since outperforms the commercial Insight image segmentation software (Phoenix Research Labs) as well as its utility to generate labelled images for validation purposes speeding significantly up the ground truth generation process. Regarding the second approach, the deep-learning-based method improves the results achieved by the more conventional method and also by other state-of-the-art techniques. In addition, it was verified that the results of the proposed network can be generalized to new rat OCT images.
Bibliografie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ISSN:0169-2607
1872-7565
1872-7565
DOI:10.1016/j.cmpb.2020.105788