Registration of OCT fundus images with color fundus photographs based on blood vessel ridges

Ying Li, Giovanni Gregori, Robert W. Knighton, Brandon Lujan, Philip J. Rosenfeld

Research output: Contribution to journalArticle

41 Citations (Scopus)

Abstract

This paper proposes an algorithm to register OCT fundus images (OFIs) with color fundus photographs (CFPs). This makes it possible to correlate retinal features across the different imaging modalities. Blood vessel ridges are taken as features for registration. A specially defined distance, incorporating information of normal direction of blood vessel ridge pixels, is designed to calculate the distance between each pair of pixels to be matched in the pair image. Based on this distance a similarity function between the pair image is defined. Brute force search is used for a coarse registration and then an Iterative Closest Point (ICP) algorithm for a more accurate registration. The registration algorithm was tested on a sample set containing images of both normal eyes and eyes with pathologies. Three transformation models (similarity, affine and quadratic models) were tested on all image pairs respectively. The experimental results showed that the registration algorithm worked well. The average root mean square errors for the affine model are 31 μm (normal) and 59 μm (eyes with disease). The proposed algorithm can be easily adapted to registration for other modality retinal images.

Original languageEnglish (US)
Pages (from-to)7-16
Number of pages10
JournalOptics Express
Volume19
Issue number1
DOIs
StatePublished - Jan 3 2011
Externally publishedYes

Fingerprint

blood vessels
photographs
ridges
color
pixels
retinal images
root-mean-square errors
registers
pathology

ASJC Scopus subject areas

  • Atomic and Molecular Physics, and Optics

Cite this

Registration of OCT fundus images with color fundus photographs based on blood vessel ridges. / Li, Ying; Gregori, Giovanni; Knighton, Robert W.; Lujan, Brandon; Rosenfeld, Philip J.

In: Optics Express, Vol. 19, No. 1, 03.01.2011, p. 7-16.

Research output: Contribution to journalArticle

Li, Ying ; Gregori, Giovanni ; Knighton, Robert W. ; Lujan, Brandon ; Rosenfeld, Philip J. / Registration of OCT fundus images with color fundus photographs based on blood vessel ridges. In: Optics Express. 2011 ; Vol. 19, No. 1. pp. 7-16.
@article{c3f46817ac3b492396bd3522d7dfdcf1,
title = "Registration of OCT fundus images with color fundus photographs based on blood vessel ridges",
abstract = "This paper proposes an algorithm to register OCT fundus images (OFIs) with color fundus photographs (CFPs). This makes it possible to correlate retinal features across the different imaging modalities. Blood vessel ridges are taken as features for registration. A specially defined distance, incorporating information of normal direction of blood vessel ridge pixels, is designed to calculate the distance between each pair of pixels to be matched in the pair image. Based on this distance a similarity function between the pair image is defined. Brute force search is used for a coarse registration and then an Iterative Closest Point (ICP) algorithm for a more accurate registration. The registration algorithm was tested on a sample set containing images of both normal eyes and eyes with pathologies. Three transformation models (similarity, affine and quadratic models) were tested on all image pairs respectively. The experimental results showed that the registration algorithm worked well. The average root mean square errors for the affine model are 31 μm (normal) and 59 μm (eyes with disease). The proposed algorithm can be easily adapted to registration for other modality retinal images.",
author = "Ying Li and Giovanni Gregori and Knighton, {Robert W.} and Brandon Lujan and Rosenfeld, {Philip J.}",
year = "2011",
month = "1",
day = "3",
doi = "10.1364/OE.19.000007",
language = "English (US)",
volume = "19",
pages = "7--16",
journal = "Optics Express",
issn = "1094-4087",
publisher = "The Optical Society",
number = "1",

}

TY - JOUR

T1 - Registration of OCT fundus images with color fundus photographs based on blood vessel ridges

AU - Li, Ying

AU - Gregori, Giovanni

AU - Knighton, Robert W.

AU - Lujan, Brandon

AU - Rosenfeld, Philip J.

PY - 2011/1/3

Y1 - 2011/1/3

N2 - This paper proposes an algorithm to register OCT fundus images (OFIs) with color fundus photographs (CFPs). This makes it possible to correlate retinal features across the different imaging modalities. Blood vessel ridges are taken as features for registration. A specially defined distance, incorporating information of normal direction of blood vessel ridge pixels, is designed to calculate the distance between each pair of pixels to be matched in the pair image. Based on this distance a similarity function between the pair image is defined. Brute force search is used for a coarse registration and then an Iterative Closest Point (ICP) algorithm for a more accurate registration. The registration algorithm was tested on a sample set containing images of both normal eyes and eyes with pathologies. Three transformation models (similarity, affine and quadratic models) were tested on all image pairs respectively. The experimental results showed that the registration algorithm worked well. The average root mean square errors for the affine model are 31 μm (normal) and 59 μm (eyes with disease). The proposed algorithm can be easily adapted to registration for other modality retinal images.

AB - This paper proposes an algorithm to register OCT fundus images (OFIs) with color fundus photographs (CFPs). This makes it possible to correlate retinal features across the different imaging modalities. Blood vessel ridges are taken as features for registration. A specially defined distance, incorporating information of normal direction of blood vessel ridge pixels, is designed to calculate the distance between each pair of pixels to be matched in the pair image. Based on this distance a similarity function between the pair image is defined. Brute force search is used for a coarse registration and then an Iterative Closest Point (ICP) algorithm for a more accurate registration. The registration algorithm was tested on a sample set containing images of both normal eyes and eyes with pathologies. Three transformation models (similarity, affine and quadratic models) were tested on all image pairs respectively. The experimental results showed that the registration algorithm worked well. The average root mean square errors for the affine model are 31 μm (normal) and 59 μm (eyes with disease). The proposed algorithm can be easily adapted to registration for other modality retinal images.

UR - http://www.scopus.com/inward/record.url?scp=78651348679&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=78651348679&partnerID=8YFLogxK

U2 - 10.1364/OE.19.000007

DO - 10.1364/OE.19.000007

M3 - Article

C2 - 21263537

AN - SCOPUS:78651348679

VL - 19

SP - 7

EP - 16

JO - Optics Express

JF - Optics Express

SN - 1094-4087

IS - 1

ER -