4D Graph-Based Segmentation for Reproducible and Sensitive Choroid Quantification From Longitudinal OCT Scans

Ipek Oguz; Michael D Abramoff; Li Zhang; Kyungmoo Lee; Ellen Ziyi Zhang; Milan Sonka

doi:10.1167/iovs.15-18924

Back

4D Graph-Based Segmentation for Reproducible and Sensitive Choroid Quantification From Longitudinal OCT Scans

Journal article

Open access

Peer reviewed

4D Graph-Based Segmentation for Reproducible and Sensitive Choroid Quantification From Longitudinal OCT Scans

Ipek Oguz, Michael D Abramoff, Li Zhang, Kyungmoo Lee, Ellen Ziyi Zhang and Milan Sonka

Investigative ophthalmology & visual science, Vol.57(9), pp.OCT621-OCT630

07/01/2016

DOI: 10.1167/iovs.15-18924

PMCID: PMC5215413

PMID: 27936264

Files and links (1)

url

https://doi.org/10.1167/iovs.15-18924View

Published (Version of record) Open Access

Abstract

Longitudinal imaging is becoming more commonplace for studies of disease progression, response to treatment, and healthy maturation. Accurate and reproducible quantification methods are desirable to fully mine the wealth of data in such datasets. However, most current retinal OCT segmentation methods are cross-sectional and fail to leverage the inherent context present in longitudinal sequences of images. We propose a novel graph-based method for segmentation of multiple three-dimensional (3D) scans over time (termed 3D + time or 4D). The usefulness of this approach in retinal imaging is illustrated in the segmentation of the choroidal surfaces from longitudinal optical coherence tomography (OCT) scans. A total of 3219 synthetic (3070) and patient (149) OCT images were segmented for validation of our approach. The results show that the proposed 4D segmentation method is significantly more reproducible (P < 0.001) than the 3D approach and is significantly more sensitive to temporal changes (P < 0.0001) achieved by the substantial increase of measurement robustness. This is the first automated 4D method for jointly quantifying choroidal thickness in longitudinal OCT studies. Our method is robust to image noise and produces more reproducible choroidal thickness measurements than a sequence of independent 3D segmentations, without sacrificing sensitivity to temporal changes.

Choroidal Neovascularization - diagnosis

Reproducibility of Results

Humans

Female

Male

ROC Curve

Aged

Image Processing, Computer-Assisted - methods

Tomography, Optical Coherence - methods

Choroid - pathology

Details

Title: Subtitle: 4D Graph-Based Segmentation for Reproducible and Sensitive Choroid Quantification From Longitudinal OCT Scans
Creators: Ipek Oguz - University of Iowa, The University of Iowa Institute for Vision Research
Michael D Abramoff - University of Iowa, The University of Iowa Institute for Vision Research
Li Zhang - Iowa Institute for Biomedical Imaging, The University of Iowa, Iowa City, Iowa, United States
Kyungmoo Lee - Iowa Institute for Biomedical Imaging, The University of Iowa, Iowa City, Iowa, United States
Ellen Ziyi Zhang - Wellman Center for Photomedicine, Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts, United States
Milan Sonka - University of Iowa, The University of Iowa Institute for Vision Research
Resource Type: Journal article
Publication Details: Investigative ophthalmology & visual science, Vol.57(9), pp.OCT621-OCT630
DOI: 10.1167/iovs.15-18924
PMID: 27936264
PMCID: PMC5215413
NLM abbreviation: Invest Ophthalmol Vis Sci
ISSN: 1552-5783
eISSN: 1552-5783
Publisher: United States
Grant note: P41 EB015903 / NIBIB NIH HHS R01 EB004640 / NIBIB NIH HHS R01 EY018853 / NEI NIH HHS R01 EY019112 / NEI NIH HHS R01 NS094456 / NINDS NIH HHS R01 CA163528 / NCI NIH HHS
Language: English
Date published: 07/01/2016
Academic Unit: Roy J. Carver Department of Biomedical Engineering; Electrical and Computer Engineering; Psychiatry; Radiation Oncology; Injury Prevention Research Center; Ophthalmology and Visual Sciences
Record Identifier: 9983806393102771

Metrics

38 Record Views