Automated denoising and segmentation of optical coherence tomography images

Sohini Roychowdhury, Dara Koozekanani, Keshab K Parhi

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Citations (Scopus)

Abstract

This paper presents a novel automated system that denoises and segments seven sub-retinal layers in optical coherence tomography (OCT) images. First, the OCT images are subjected to Wiener deconvolution by varying the noise variance from 10-1 to 10-15. A new Fourier-domain structural error is introduced in this paper, and the deconvolved OCT image with the least structural error is selected as the denoised image. The properties of the structural error metric are studied, and it is shown that the error metric satisfies convexity property. For each image, the proposed denoising method increases the image SNR by 6.9 dB on average compared to 5 dB increase reported so far, and attains a mean peak SNR (PSNR) of 23.036 dB. Next, highpass filters are applied to the denoised images in an iterative manner to extract the seven sub-retinal layers. The proposed system requires on average 10.65 seconds for denoising an image and 22.07 seconds for segmenting seven sub-retinal layers. This is a significant improvement over manual segmentation that requires up to 12 minutes per image.

Original languageEnglish (US)
Title of host publicationConference Record of the 47th Asilomar Conference on Signals, Systems and Computers
PublisherIEEE Computer Society
Pages258-262
Number of pages5
ISBN (Print)9781479923908
DOIs
StatePublished - Jan 1 2013
Event2013 47th Asilomar Conference on Signals, Systems and Computers - Pacific Grove, CA, United States
Duration: Nov 3 2013Nov 6 2013

Other

Other2013 47th Asilomar Conference on Signals, Systems and Computers
CountryUnited States
CityPacific Grove, CA
Period11/3/1311/6/13

Fingerprint

Optical tomography
Image denoising
Deconvolution

Keywords

  • denoising
  • Fourier-domain representation
  • iterative segmentation
  • Optical coherence tomography

Cite this

Roychowdhury, S., Koozekanani, D., & Parhi, K. K. (2013). Automated denoising and segmentation of optical coherence tomography images. In Conference Record of the 47th Asilomar Conference on Signals, Systems and Computers (pp. 258-262). [6810272] IEEE Computer Society. https://doi.org/10.1109/ACSSC.2013.6810272

Automated denoising and segmentation of optical coherence tomography images. / Roychowdhury, Sohini; Koozekanani, Dara; Parhi, Keshab K.

Conference Record of the 47th Asilomar Conference on Signals, Systems and Computers. IEEE Computer Society, 2013. p. 258-262 6810272.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Roychowdhury, S, Koozekanani, D & Parhi, KK 2013, Automated denoising and segmentation of optical coherence tomography images. in Conference Record of the 47th Asilomar Conference on Signals, Systems and Computers., 6810272, IEEE Computer Society, pp. 258-262, 2013 47th Asilomar Conference on Signals, Systems and Computers, Pacific Grove, CA, United States, 11/3/13. https://doi.org/10.1109/ACSSC.2013.6810272
Roychowdhury S, Koozekanani D, Parhi KK. Automated denoising and segmentation of optical coherence tomography images. In Conference Record of the 47th Asilomar Conference on Signals, Systems and Computers. IEEE Computer Society. 2013. p. 258-262. 6810272 https://doi.org/10.1109/ACSSC.2013.6810272
Roychowdhury, Sohini ; Koozekanani, Dara ; Parhi, Keshab K. / Automated denoising and segmentation of optical coherence tomography images. Conference Record of the 47th Asilomar Conference on Signals, Systems and Computers. IEEE Computer Society, 2013. pp. 258-262
@inproceedings{4de7001c3e5a44cf96aaf68f5003880f,
title = "Automated denoising and segmentation of optical coherence tomography images",
abstract = "This paper presents a novel automated system that denoises and segments seven sub-retinal layers in optical coherence tomography (OCT) images. First, the OCT images are subjected to Wiener deconvolution by varying the noise variance from 10-1 to 10-15. A new Fourier-domain structural error is introduced in this paper, and the deconvolved OCT image with the least structural error is selected as the denoised image. The properties of the structural error metric are studied, and it is shown that the error metric satisfies convexity property. For each image, the proposed denoising method increases the image SNR by 6.9 dB on average compared to 5 dB increase reported so far, and attains a mean peak SNR (PSNR) of 23.036 dB. Next, highpass filters are applied to the denoised images in an iterative manner to extract the seven sub-retinal layers. The proposed system requires on average 10.65 seconds for denoising an image and 22.07 seconds for segmenting seven sub-retinal layers. This is a significant improvement over manual segmentation that requires up to 12 minutes per image.",
keywords = "denoising, Fourier-domain representation, iterative segmentation, Optical coherence tomography",
author = "Sohini Roychowdhury and Dara Koozekanani and Parhi, {Keshab K}",
year = "2013",
month = "1",
day = "1",
doi = "10.1109/ACSSC.2013.6810272",
language = "English (US)",
isbn = "9781479923908",
pages = "258--262",
booktitle = "Conference Record of the 47th Asilomar Conference on Signals, Systems and Computers",
publisher = "IEEE Computer Society",

}

TY - GEN

T1 - Automated denoising and segmentation of optical coherence tomography images

AU - Roychowdhury, Sohini

AU - Koozekanani, Dara

AU - Parhi, Keshab K

PY - 2013/1/1

Y1 - 2013/1/1

N2 - This paper presents a novel automated system that denoises and segments seven sub-retinal layers in optical coherence tomography (OCT) images. First, the OCT images are subjected to Wiener deconvolution by varying the noise variance from 10-1 to 10-15. A new Fourier-domain structural error is introduced in this paper, and the deconvolved OCT image with the least structural error is selected as the denoised image. The properties of the structural error metric are studied, and it is shown that the error metric satisfies convexity property. For each image, the proposed denoising method increases the image SNR by 6.9 dB on average compared to 5 dB increase reported so far, and attains a mean peak SNR (PSNR) of 23.036 dB. Next, highpass filters are applied to the denoised images in an iterative manner to extract the seven sub-retinal layers. The proposed system requires on average 10.65 seconds for denoising an image and 22.07 seconds for segmenting seven sub-retinal layers. This is a significant improvement over manual segmentation that requires up to 12 minutes per image.

AB - This paper presents a novel automated system that denoises and segments seven sub-retinal layers in optical coherence tomography (OCT) images. First, the OCT images are subjected to Wiener deconvolution by varying the noise variance from 10-1 to 10-15. A new Fourier-domain structural error is introduced in this paper, and the deconvolved OCT image with the least structural error is selected as the denoised image. The properties of the structural error metric are studied, and it is shown that the error metric satisfies convexity property. For each image, the proposed denoising method increases the image SNR by 6.9 dB on average compared to 5 dB increase reported so far, and attains a mean peak SNR (PSNR) of 23.036 dB. Next, highpass filters are applied to the denoised images in an iterative manner to extract the seven sub-retinal layers. The proposed system requires on average 10.65 seconds for denoising an image and 22.07 seconds for segmenting seven sub-retinal layers. This is a significant improvement over manual segmentation that requires up to 12 minutes per image.

KW - denoising

KW - Fourier-domain representation

KW - iterative segmentation

KW - Optical coherence tomography

UR - http://www.scopus.com/inward/record.url?scp=84901259298&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84901259298&partnerID=8YFLogxK

U2 - 10.1109/ACSSC.2013.6810272

DO - 10.1109/ACSSC.2013.6810272

M3 - Conference contribution

SN - 9781479923908

SP - 258

EP - 262

BT - Conference Record of the 47th Asilomar Conference on Signals, Systems and Computers

PB - IEEE Computer Society

ER -