Automatic analysis of diabetic peripheral neuropathy using multi-scale quantitative morphology of nerve fibres in corneal confocal microscopy imaging

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

Research output: Contribution to journalArticle

101 Citations (Scopus)

Abstract

Diabetic peripheral neuropathy (DPN) is one of the most common long term complications of diabetes. Corneal confocal microscopy (CCM) image analysis is a novel non-invasive technique which quantifies corneal nerve fibre damage and enables diagnosis of DPN. This paper presents an automatic analysis and classification system for detecting nerve fibres in CCM images based on a multi-scale adaptive dual-model detection algorithm. The algorithm exploits the curvilinear structure of the nerve fibres and adapts itself to the local image information. Detected nerve fibres are then quantified and used as feature vectors for classification using random forest (RF) and neural networks (NNT) classifiers. We show, in a comparative study with other well known curvilinear detectors, that the best performance is achieved by the multi-scale dual model in conjunction with the NNT classifier. An evaluation of clinical effectiveness shows that the performance of the automated system matches that of ground-truth defined by expert manual annotation.

Original languageEnglish
Pages (from-to)738-747
Number of pages10
JournalMedical Image Analysis
Volume15
Issue number5
DOIs
Publication statusPublished - Oct 2011
Externally publishedYes

Fingerprint

Diabetic Neuropathies
Confocal microscopy
Peripheral Nervous System Diseases
Nerve Fibers
Confocal Microscopy
Imaging techniques
Fibers
Classifiers
Neural networks
Diabetes Complications
Medical problems
Image analysis
Detectors

Keywords

  • Corneal confocal microscopy
  • Curvilinear structures
  • Diabetic neuropathy
  • Image quantification

ASJC Scopus subject areas

  • Radiological and Ultrasound Technology
  • Radiology Nuclear Medicine and imaging
  • Computer Vision and Pattern Recognition
  • Health Informatics
  • Computer Graphics and Computer-Aided Design

Cite this

Automatic analysis of diabetic peripheral neuropathy using multi-scale quantitative morphology of nerve fibres in corneal confocal microscopy imaging. / Dabbah, M. A.; Graham, J.; Petropoulos, I. N.; Tavakoli, M.; Malik, Rayaz.

In: Medical Image Analysis, Vol. 15, No. 5, 10.2011, p. 738-747.

Research output: Contribution to journalArticle

@article{244835620c644466b48849ae535977b1,
title = "Automatic analysis of diabetic peripheral neuropathy using multi-scale quantitative morphology of nerve fibres in corneal confocal microscopy imaging",
abstract = "Diabetic peripheral neuropathy (DPN) is one of the most common long term complications of diabetes. Corneal confocal microscopy (CCM) image analysis is a novel non-invasive technique which quantifies corneal nerve fibre damage and enables diagnosis of DPN. This paper presents an automatic analysis and classification system for detecting nerve fibres in CCM images based on a multi-scale adaptive dual-model detection algorithm. The algorithm exploits the curvilinear structure of the nerve fibres and adapts itself to the local image information. Detected nerve fibres are then quantified and used as feature vectors for classification using random forest (RF) and neural networks (NNT) classifiers. We show, in a comparative study with other well known curvilinear detectors, that the best performance is achieved by the multi-scale dual model in conjunction with the NNT classifier. An evaluation of clinical effectiveness shows that the performance of the automated system matches that of ground-truth defined by expert manual annotation.",
keywords = "Corneal confocal microscopy, Curvilinear structures, Diabetic neuropathy, Image quantification",
author = "Dabbah, {M. A.} and J. Graham and Petropoulos, {I. N.} and M. Tavakoli and Rayaz Malik",
year = "2011",
month = "10",
doi = "10.1016/j.media.2011.05.016",
language = "English",
volume = "15",
pages = "738--747",
journal = "Medical Image Analysis",
issn = "1361-8415",
publisher = "Elsevier",
number = "5",

}

TY - JOUR

T1 - Automatic analysis of diabetic peripheral neuropathy using multi-scale quantitative morphology of nerve fibres in corneal confocal microscopy imaging

AU - Dabbah, M. A.

AU - Graham, J.

AU - Petropoulos, I. N.

AU - Tavakoli, M.

AU - Malik, Rayaz

PY - 2011/10

Y1 - 2011/10

N2 - Diabetic peripheral neuropathy (DPN) is one of the most common long term complications of diabetes. Corneal confocal microscopy (CCM) image analysis is a novel non-invasive technique which quantifies corneal nerve fibre damage and enables diagnosis of DPN. This paper presents an automatic analysis and classification system for detecting nerve fibres in CCM images based on a multi-scale adaptive dual-model detection algorithm. The algorithm exploits the curvilinear structure of the nerve fibres and adapts itself to the local image information. Detected nerve fibres are then quantified and used as feature vectors for classification using random forest (RF) and neural networks (NNT) classifiers. We show, in a comparative study with other well known curvilinear detectors, that the best performance is achieved by the multi-scale dual model in conjunction with the NNT classifier. An evaluation of clinical effectiveness shows that the performance of the automated system matches that of ground-truth defined by expert manual annotation.

AB - Diabetic peripheral neuropathy (DPN) is one of the most common long term complications of diabetes. Corneal confocal microscopy (CCM) image analysis is a novel non-invasive technique which quantifies corneal nerve fibre damage and enables diagnosis of DPN. This paper presents an automatic analysis and classification system for detecting nerve fibres in CCM images based on a multi-scale adaptive dual-model detection algorithm. The algorithm exploits the curvilinear structure of the nerve fibres and adapts itself to the local image information. Detected nerve fibres are then quantified and used as feature vectors for classification using random forest (RF) and neural networks (NNT) classifiers. We show, in a comparative study with other well known curvilinear detectors, that the best performance is achieved by the multi-scale dual model in conjunction with the NNT classifier. An evaluation of clinical effectiveness shows that the performance of the automated system matches that of ground-truth defined by expert manual annotation.

KW - Corneal confocal microscopy

KW - Curvilinear structures

KW - Diabetic neuropathy

KW - Image quantification

UR - http://www.scopus.com/inward/record.url?scp=80052181311&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=80052181311&partnerID=8YFLogxK

U2 - 10.1016/j.media.2011.05.016

DO - 10.1016/j.media.2011.05.016

M3 - Article

C2 - 21719344

AN - SCOPUS:80052181311

VL - 15

SP - 738

EP - 747

JO - Medical Image Analysis

JF - Medical Image Analysis

SN - 1361-8415

IS - 5

ER -