Structured layer surface segmentation for retina OCT using fully convolutional regression networks

•A novel formulation of deep network to output continuous, smooth and topology correct surfaces without post-processing.•End-to-end optimization of multiple structured surface segmentation.•An effective multitask deep network for retinal layer surface and lesion segmentation. [Display omitted] Optic...

Full description

Saved in:
Bibliographic Details
Published in:Medical image analysis Vol. 68; p. 101856
Main Authors: He, Yufan, Carass, Aaron, Liu, Yihao, Jedynak, Bruno M., Solomon, Sharon D., Saidha, Shiv, Calabresi, Peter A., Prince, Jerry L.
Format: Journal Article
Language:English
Published: Netherlands Elsevier B.V 01.02.2021
Elsevier BV
Subjects:
Algorithms
Automation
Biomarkers
Boundaries
Continuity (mathematics)
Deep learning
Deep learning segmentation
Diabetes mellitus
Diabetic Retinopathy - diagnostic imaging
Edema
Humans
Image segmentation
Machine learning
Macular Edema
Multiple sclerosis
Optical Coherence Tomography
Optical communication
Pixels
Retina
Retina - diagnostic imaging
Retina OCT
Surface segmentation
Tomography, Optical Coherence
Topology
68T45
Surface segmentation
Retina OCT
41A10
Deep learning segmentation
65D05
65D17
41A05
ISSN:1361-8415, 1361-8423, 1361-8423
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:•A novel formulation of deep network to output continuous, smooth and topology correct surfaces without post-processing.•End-to-end optimization of multiple structured surface segmentation.•An effective multitask deep network for retinal layer surface and lesion segmentation. [Display omitted] Optical coherence tomography (OCT) is a noninvasive imaging modality with micrometer resolution which has been widely used for scanning the retina. Retinal layers are important biomarkers for many diseases. Accurate automated algorithms for segmenting smooth continuous layer surfaces with correct hierarchy (topology) are important for automated retinal thickness and surface shape analysis. State-of-the-art methods typically use a two step process. Firstly, a trained classifier is used to label each pixel into either background and layers or boundaries and non-boundaries. Secondly, the desired smooth surfaces with the correct topology are extracted by graph methods (e.g., graph cut). Data driven methods like deep networks have shown great ability for the pixel classification step, but to date have not been able to extract structured smooth continuous surfaces with topological constraints in the second step. In this paper, we combine these two steps into a unified deep learning framework by directly modeling the distribution of the surface positions. Smooth, continuous, and topologically correct surfaces are obtained in a single feed forward operation. The proposed method was evaluated on two publicly available data sets of healthy controls and subjects with either multiple sclerosis or diabetic macular edema, and is shown to achieve state-of-the art performance with sub-pixel accuracy.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
content type line 23
Author Statment
Yufan He: Conceptualization, Methodology, Software, Validation, Writing - Original Draft Aaron Carass: Conceptualization, Writing - Review & Editing Yihao Liu: Conceptualization, Writing - review & editing Bruno M. Jedynak: Formal analysis, Writing - review & editing Sharon D. Solomon: Resources, Writing - review & editing Shiv Saidha: Resources, Writing - review & editing Peter A. Calabresi: Resources, Writing - review & editing Jerry L. Prince: Conceptualization, Supervision, Project administration, Writing - review & editing
ISSN:1361-8415
1361-8423
1361-8423
DOI:10.1016/j.media.2020.101856