Tracking the Optic Nervehead in OCT Video Using Dual Eigenspaces and an Adaptive Vascular Distribution Model

Dara Koozekanani, Kim L. Boyer, Cynthia Roberts

Research output: Contribution to journalArticle

14 Citations (Scopus)

Abstract

Optical coherence tomography (OCT) is a new ophthalmic imaging modality generating cross sectional views of the retina. OCT systems are essentially Michelson interferometers that form images in 1.5 s by directing a super-luminescent diode (SLD) beam over the retinal surface. Involuntary eye motions frequently cause incorrect locations to be imaged. This motion may leave no obvious artifacts in the scan data and can easily go undetected. For glaucoma monitoring especially, knowing the measurement path, typically a circle concentric with the nerve head, is crucial. The commercially available OCT system displays a near-infrared video of the retina showing the SLD beam. This paper presents a prototype system to detect the nerve head and SLD beam in the video, and report the true scan path relative to the nerve head. Low image contrast and limited resolution make the reliable detection of retinal features difficult. In an adaptive model construction phase, the system directly detects retinal vasculature and the nerve head and incrementally builds a model of the current subject's vascular pattern relative to the optic disk. The nerve head identification is multitiered, using a novel dual eigenspace technique and a geometric comparison of detected vessel positions and nerve head hypotheses. In its operational phase, a correspondence is achieved between the currently detected vasculature and the model. Using subjects not included in training, the system located the optic nerve head to within 5 pixels (0.07 optic disk diameters, an error well below clinical significance) in 99.75% of 2800 video fields. In current clinical practice, motions as large as 1-2 disc diameters may go undetected, so this is a vast improvement.

Original languageEnglish (US)
Pages (from-to)1519-1536
Number of pages18
JournalIEEE Transactions on Medical Imaging
Volume22
Issue number12
DOIs
StatePublished - Dec 1 2003

Fingerprint

Optical tomography
Optic Disk
Optical Coherence Tomography
Blood Vessels
Optics
Diodes
Retina
Michelson interferometers
Glaucoma
Artifacts
Pixels
Infrared radiation
Imaging techniques
Monitoring

Keywords

  • Eigenspaces
  • Fundus imaging
  • Ophthalmology
  • Optical coherence tomography
  • Retina
  • Video tracking

Cite this

Tracking the Optic Nervehead in OCT Video Using Dual Eigenspaces and an Adaptive Vascular Distribution Model. / Koozekanani, Dara; Boyer, Kim L.; Roberts, Cynthia.

In: IEEE Transactions on Medical Imaging, Vol. 22, No. 12, 01.12.2003, p. 1519-1536.

Research output: Contribution to journalArticle

@article{48daec702a8541a78dbdead3b13e3537,
title = "Tracking the Optic Nervehead in OCT Video Using Dual Eigenspaces and an Adaptive Vascular Distribution Model",
abstract = "Optical coherence tomography (OCT) is a new ophthalmic imaging modality generating cross sectional views of the retina. OCT systems are essentially Michelson interferometers that form images in 1.5 s by directing a super-luminescent diode (SLD) beam over the retinal surface. Involuntary eye motions frequently cause incorrect locations to be imaged. This motion may leave no obvious artifacts in the scan data and can easily go undetected. For glaucoma monitoring especially, knowing the measurement path, typically a circle concentric with the nerve head, is crucial. The commercially available OCT system displays a near-infrared video of the retina showing the SLD beam. This paper presents a prototype system to detect the nerve head and SLD beam in the video, and report the true scan path relative to the nerve head. Low image contrast and limited resolution make the reliable detection of retinal features difficult. In an adaptive model construction phase, the system directly detects retinal vasculature and the nerve head and incrementally builds a model of the current subject's vascular pattern relative to the optic disk. The nerve head identification is multitiered, using a novel dual eigenspace technique and a geometric comparison of detected vessel positions and nerve head hypotheses. In its operational phase, a correspondence is achieved between the currently detected vasculature and the model. Using subjects not included in training, the system located the optic nerve head to within 5 pixels (0.07 optic disk diameters, an error well below clinical significance) in 99.75{\%} of 2800 video fields. In current clinical practice, motions as large as 1-2 disc diameters may go undetected, so this is a vast improvement.",
keywords = "Eigenspaces, Fundus imaging, Ophthalmology, Optical coherence tomography, Retina, Video tracking",
author = "Dara Koozekanani and Boyer, {Kim L.} and Cynthia Roberts",
year = "2003",
month = "12",
day = "1",
doi = "10.1109/TMI.2003.817753",
language = "English (US)",
volume = "22",
pages = "1519--1536",
journal = "IEEE Transactions on Medical Imaging",
issn = "0278-0062",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "12",

}

TY - JOUR

T1 - Tracking the Optic Nervehead in OCT Video Using Dual Eigenspaces and an Adaptive Vascular Distribution Model

AU - Koozekanani, Dara

AU - Boyer, Kim L.

AU - Roberts, Cynthia

PY - 2003/12/1

Y1 - 2003/12/1

N2 - Optical coherence tomography (OCT) is a new ophthalmic imaging modality generating cross sectional views of the retina. OCT systems are essentially Michelson interferometers that form images in 1.5 s by directing a super-luminescent diode (SLD) beam over the retinal surface. Involuntary eye motions frequently cause incorrect locations to be imaged. This motion may leave no obvious artifacts in the scan data and can easily go undetected. For glaucoma monitoring especially, knowing the measurement path, typically a circle concentric with the nerve head, is crucial. The commercially available OCT system displays a near-infrared video of the retina showing the SLD beam. This paper presents a prototype system to detect the nerve head and SLD beam in the video, and report the true scan path relative to the nerve head. Low image contrast and limited resolution make the reliable detection of retinal features difficult. In an adaptive model construction phase, the system directly detects retinal vasculature and the nerve head and incrementally builds a model of the current subject's vascular pattern relative to the optic disk. The nerve head identification is multitiered, using a novel dual eigenspace technique and a geometric comparison of detected vessel positions and nerve head hypotheses. In its operational phase, a correspondence is achieved between the currently detected vasculature and the model. Using subjects not included in training, the system located the optic nerve head to within 5 pixels (0.07 optic disk diameters, an error well below clinical significance) in 99.75% of 2800 video fields. In current clinical practice, motions as large as 1-2 disc diameters may go undetected, so this is a vast improvement.

AB - Optical coherence tomography (OCT) is a new ophthalmic imaging modality generating cross sectional views of the retina. OCT systems are essentially Michelson interferometers that form images in 1.5 s by directing a super-luminescent diode (SLD) beam over the retinal surface. Involuntary eye motions frequently cause incorrect locations to be imaged. This motion may leave no obvious artifacts in the scan data and can easily go undetected. For glaucoma monitoring especially, knowing the measurement path, typically a circle concentric with the nerve head, is crucial. The commercially available OCT system displays a near-infrared video of the retina showing the SLD beam. This paper presents a prototype system to detect the nerve head and SLD beam in the video, and report the true scan path relative to the nerve head. Low image contrast and limited resolution make the reliable detection of retinal features difficult. In an adaptive model construction phase, the system directly detects retinal vasculature and the nerve head and incrementally builds a model of the current subject's vascular pattern relative to the optic disk. The nerve head identification is multitiered, using a novel dual eigenspace technique and a geometric comparison of detected vessel positions and nerve head hypotheses. In its operational phase, a correspondence is achieved between the currently detected vasculature and the model. Using subjects not included in training, the system located the optic nerve head to within 5 pixels (0.07 optic disk diameters, an error well below clinical significance) in 99.75% of 2800 video fields. In current clinical practice, motions as large as 1-2 disc diameters may go undetected, so this is a vast improvement.

KW - Eigenspaces

KW - Fundus imaging

KW - Ophthalmology

KW - Optical coherence tomography

KW - Retina

KW - Video tracking

UR - http://www.scopus.com/inward/record.url?scp=0344875703&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0344875703&partnerID=8YFLogxK

U2 - 10.1109/TMI.2003.817753

DO - 10.1109/TMI.2003.817753

M3 - Article

C2 - 14649743

AN - SCOPUS:0344875703

VL - 22

SP - 1519

EP - 1536

JO - IEEE Transactions on Medical Imaging

JF - IEEE Transactions on Medical Imaging

SN - 0278-0062

IS - 12

ER -