Dual-model automatic detection of nerve-fibres in corneal confocal microscopy images.

M. A. Dabbah, J. Graham, I. Petropoulos, M. Tavakoli, Rayaz Malik

Research output: Contribution to journalArticle

40 Citations (Scopus)

Abstract

Corneal Confocal Microscopy (CCM) imaging is a non-invasive surrogate of detecting, quantifying and monitoring diabetic peripheral neuropathy. This paper presents an automated method for detecting nerve-fibres from CCM images using a dual-model detection algorithm and compares the performance to well-established texture and feature detection methods. The algorithm comprises two separate models, one for the background and another for the foreground (nerve-fibres), which work interactively. Our evaluation shows significant improvement (p approximately 0) in both error rate and signal-to-noise ratio of this model over the competitor methods. The automatic method is also evaluated in comparison with manual ground truth analysis in assessing diabetic neuropathy on the basis of nerve-fibre length, and shows a strong correlation (r = 0.92). Both analyses significantly separate diabetic patients from control subjects (p approximately 0).

Original languageEnglish
Pages (from-to)300-307
Number of pages8
JournalMedical image computing and computer-assisted intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention
Volume13
Issue numberPt 1
Publication statusPublished - 2010
Externally publishedYes

Fingerprint

Nerve Fibers
Confocal Microscopy
Diabetic Neuropathies
Signal-To-Noise Ratio
Peripheral Nervous System Diseases

ASJC Scopus subject areas

  • Medicine(all)

Cite this

@article{0380f51485d64055b6487541a3e4eac9,
title = "Dual-model automatic detection of nerve-fibres in corneal confocal microscopy images.",
abstract = "Corneal Confocal Microscopy (CCM) imaging is a non-invasive surrogate of detecting, quantifying and monitoring diabetic peripheral neuropathy. This paper presents an automated method for detecting nerve-fibres from CCM images using a dual-model detection algorithm and compares the performance to well-established texture and feature detection methods. The algorithm comprises two separate models, one for the background and another for the foreground (nerve-fibres), which work interactively. Our evaluation shows significant improvement (p approximately 0) in both error rate and signal-to-noise ratio of this model over the competitor methods. The automatic method is also evaluated in comparison with manual ground truth analysis in assessing diabetic neuropathy on the basis of nerve-fibre length, and shows a strong correlation (r = 0.92). Both analyses significantly separate diabetic patients from control subjects (p approximately 0).",
author = "Dabbah, {M. A.} and J. Graham and I. Petropoulos and M. Tavakoli and Rayaz Malik",
year = "2010",
language = "English",
volume = "13",
pages = "300--307",
journal = "Medical image computing and computer-assisted intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention",
number = "Pt 1",

}

TY - JOUR

T1 - Dual-model automatic detection of nerve-fibres in corneal confocal microscopy images.

AU - Dabbah, M. A.

AU - Graham, J.

AU - Petropoulos, I.

AU - Tavakoli, M.

AU - Malik, Rayaz

PY - 2010

Y1 - 2010

N2 - Corneal Confocal Microscopy (CCM) imaging is a non-invasive surrogate of detecting, quantifying and monitoring diabetic peripheral neuropathy. This paper presents an automated method for detecting nerve-fibres from CCM images using a dual-model detection algorithm and compares the performance to well-established texture and feature detection methods. The algorithm comprises two separate models, one for the background and another for the foreground (nerve-fibres), which work interactively. Our evaluation shows significant improvement (p approximately 0) in both error rate and signal-to-noise ratio of this model over the competitor methods. The automatic method is also evaluated in comparison with manual ground truth analysis in assessing diabetic neuropathy on the basis of nerve-fibre length, and shows a strong correlation (r = 0.92). Both analyses significantly separate diabetic patients from control subjects (p approximately 0).

AB - Corneal Confocal Microscopy (CCM) imaging is a non-invasive surrogate of detecting, quantifying and monitoring diabetic peripheral neuropathy. This paper presents an automated method for detecting nerve-fibres from CCM images using a dual-model detection algorithm and compares the performance to well-established texture and feature detection methods. The algorithm comprises two separate models, one for the background and another for the foreground (nerve-fibres), which work interactively. Our evaluation shows significant improvement (p approximately 0) in both error rate and signal-to-noise ratio of this model over the competitor methods. The automatic method is also evaluated in comparison with manual ground truth analysis in assessing diabetic neuropathy on the basis of nerve-fibre length, and shows a strong correlation (r = 0.92). Both analyses significantly separate diabetic patients from control subjects (p approximately 0).

UR - http://www.scopus.com/inward/record.url?scp=80052142762&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=80052142762&partnerID=8YFLogxK

M3 - Article

VL - 13

SP - 300

EP - 307

JO - Medical image computing and computer-assisted intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention

JF - Medical image computing and computer-assisted intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention

IS - Pt 1

ER -