Automated segmentation of skin strata in Reflectance confocal microscopy depth stacks

Samuel C. Hames, Marco Ardigò, H. Peter Soyer, Andrew P. Bradley, Tarl W. Prow

Research output: Contribution to journalArticle

7 Citations (Scopus)

Abstract

Reflectance confocal microscopy (RCM) is a powerful tool for in-vivo examination of a variety of skin diseases. However, current use of RCM depends on qualitative examination by a human expert to look for specific features in the different strata of the skin. Developing approaches to quantify features in RCM imagery requires an automated understanding of what anatomical strata is present in a given en-face section. This work presents an automated approach using a bag of features approach to represent en-face sections and a logistic regression classifier to classify sections into one of four classes (stratum corneum, viable epidermis, dermal-epidermal junction and papillary dermis). This approach was developed and tested using a dataset of 308 depth stacks from 54 volunteers in two age groups (20-30 and 50-70 years of age). The classification accuracy on the test set was 85.6%. The mean absolute error in determining the interface depth for each of the stratum corneum/viable epidermis, viable epidermis/dermal-epidermal junction and dermal-epidermal junction/papillary dermis interfaces were 3.1 μm, 6.0 μm and 5.5 μm respectively. The probabilities predicted by the classifier in the test set showed that the classifier learned an effective model of the anatomy of human skin.

Original languageEnglish
Article numbere0153208
JournalPLoS One
Volume11
Issue number4
DOIs
Publication statusPublished - Apr 1 2016

Fingerprint

Confocal microscopy
epidermis (animal)
skin (animal)
Confocal Microscopy
Skin
Classifiers
dermis
cornea
Epidermis
animal anatomy
Dermis
Cornea
skin diseases
volunteers
bags
Logistics
testing
Imagery (Psychotherapy)
Skin Diseases
Volunteers

ASJC Scopus subject areas

  • Agricultural and Biological Sciences(all)
  • Biochemistry, Genetics and Molecular Biology(all)
  • Medicine(all)

Cite this

Automated segmentation of skin strata in Reflectance confocal microscopy depth stacks. / Hames, Samuel C.; Ardigò, Marco; Soyer, H. Peter; Bradley, Andrew P.; Prow, Tarl W.

In: PLoS One, Vol. 11, No. 4, e0153208, 01.04.2016.

Research output: Contribution to journalArticle

Hames, Samuel C. ; Ardigò, Marco ; Soyer, H. Peter ; Bradley, Andrew P. ; Prow, Tarl W. / Automated segmentation of skin strata in Reflectance confocal microscopy depth stacks. In: PLoS One. 2016 ; Vol. 11, No. 4.
@article{b424a045f62d4537ae2409d62afa6c5e,
title = "Automated segmentation of skin strata in Reflectance confocal microscopy depth stacks",
abstract = "Reflectance confocal microscopy (RCM) is a powerful tool for in-vivo examination of a variety of skin diseases. However, current use of RCM depends on qualitative examination by a human expert to look for specific features in the different strata of the skin. Developing approaches to quantify features in RCM imagery requires an automated understanding of what anatomical strata is present in a given en-face section. This work presents an automated approach using a bag of features approach to represent en-face sections and a logistic regression classifier to classify sections into one of four classes (stratum corneum, viable epidermis, dermal-epidermal junction and papillary dermis). This approach was developed and tested using a dataset of 308 depth stacks from 54 volunteers in two age groups (20-30 and 50-70 years of age). The classification accuracy on the test set was 85.6{\%}. The mean absolute error in determining the interface depth for each of the stratum corneum/viable epidermis, viable epidermis/dermal-epidermal junction and dermal-epidermal junction/papillary dermis interfaces were 3.1 μm, 6.0 μm and 5.5 μm respectively. The probabilities predicted by the classifier in the test set showed that the classifier learned an effective model of the anatomy of human skin.",
author = "Hames, {Samuel C.} and Marco Ardig{\`o} and Soyer, {H. Peter} and Bradley, {Andrew P.} and Prow, {Tarl W.}",
year = "2016",
month = "4",
day = "1",
doi = "10.1371/journal.pone.0153208",
language = "English",
volume = "11",
journal = "PLoS One",
issn = "1932-6203",
publisher = "Public Library of Science",
number = "4",

}

TY - JOUR

T1 - Automated segmentation of skin strata in Reflectance confocal microscopy depth stacks

AU - Hames, Samuel C.

AU - Ardigò, Marco

AU - Soyer, H. Peter

AU - Bradley, Andrew P.

AU - Prow, Tarl W.

PY - 2016/4/1

Y1 - 2016/4/1

N2 - Reflectance confocal microscopy (RCM) is a powerful tool for in-vivo examination of a variety of skin diseases. However, current use of RCM depends on qualitative examination by a human expert to look for specific features in the different strata of the skin. Developing approaches to quantify features in RCM imagery requires an automated understanding of what anatomical strata is present in a given en-face section. This work presents an automated approach using a bag of features approach to represent en-face sections and a logistic regression classifier to classify sections into one of four classes (stratum corneum, viable epidermis, dermal-epidermal junction and papillary dermis). This approach was developed and tested using a dataset of 308 depth stacks from 54 volunteers in two age groups (20-30 and 50-70 years of age). The classification accuracy on the test set was 85.6%. The mean absolute error in determining the interface depth for each of the stratum corneum/viable epidermis, viable epidermis/dermal-epidermal junction and dermal-epidermal junction/papillary dermis interfaces were 3.1 μm, 6.0 μm and 5.5 μm respectively. The probabilities predicted by the classifier in the test set showed that the classifier learned an effective model of the anatomy of human skin.

AB - Reflectance confocal microscopy (RCM) is a powerful tool for in-vivo examination of a variety of skin diseases. However, current use of RCM depends on qualitative examination by a human expert to look for specific features in the different strata of the skin. Developing approaches to quantify features in RCM imagery requires an automated understanding of what anatomical strata is present in a given en-face section. This work presents an automated approach using a bag of features approach to represent en-face sections and a logistic regression classifier to classify sections into one of four classes (stratum corneum, viable epidermis, dermal-epidermal junction and papillary dermis). This approach was developed and tested using a dataset of 308 depth stacks from 54 volunteers in two age groups (20-30 and 50-70 years of age). The classification accuracy on the test set was 85.6%. The mean absolute error in determining the interface depth for each of the stratum corneum/viable epidermis, viable epidermis/dermal-epidermal junction and dermal-epidermal junction/papillary dermis interfaces were 3.1 μm, 6.0 μm and 5.5 μm respectively. The probabilities predicted by the classifier in the test set showed that the classifier learned an effective model of the anatomy of human skin.

UR - http://www.scopus.com/inward/record.url?scp=84964664256&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84964664256&partnerID=8YFLogxK

U2 - 10.1371/journal.pone.0153208

DO - 10.1371/journal.pone.0153208

M3 - Article

AN - SCOPUS:84964664256

VL - 11

JO - PLoS One

JF - PLoS One

SN - 1932-6203

IS - 4

M1 - e0153208

ER -