Limits...
Automated Segmentation of Skin Strata in Reflectance Confocal Microscopy Depth Stacks.

Hames SC, Ardigò M, Soyer HP, Bradley AP, Prow TW - PLoS ONE (2016)

Bottom Line: 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.

View Article: PubMed Central - PubMed

Affiliation: Dermatology Research Centre, The University of Queensland, School of Medicine, Translational Research Institute, Brisbane, Australia.

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.

No MeSH data available.


Related in: MedlinePlus

RCM features used in identification of the start and end of each anatomical strata.A) The start of the viable epidermis is marked by the first visible honeycomb pattern (Blue outline). B) The start of the dermal-epidermal junction is marked by the first clearly defined papillae with bright ring of basal cells surrounding a dark centre (Blue arrows). C) The start of the papillary dermis is marked by the absence of features corresponding to the basal layer. D) Alternative definition of the start of the dermal-epidermal junction for light skin types, a dark featureless patch surrounded by keratinocytes with the absence of a bright basal layer (Blue arrow).
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4835045&req=5

pone.0153208.g001: RCM features used in identification of the start and end of each anatomical strata.A) The start of the viable epidermis is marked by the first visible honeycomb pattern (Blue outline). B) The start of the dermal-epidermal junction is marked by the first clearly defined papillae with bright ring of basal cells surrounding a dark centre (Blue arrows). C) The start of the papillary dermis is marked by the absence of features corresponding to the basal layer. D) Alternative definition of the start of the dermal-epidermal junction for light skin types, a dark featureless patch surrounded by keratinocytes with the absence of a bright basal layer (Blue arrow).

Mentions: For the purposes of segmentation four distinct anatomical strata were identified in the skin: the stratum corneum, viable epidermis, dermal-epidermal junction and the papillary dermis. To simplify the problem of establishing a ground truth in a large number of stacks it was assumed that a single anatomical strata was present in each en-face section and the extent of each strata was determined by identifying where three distinct features first occured within the stack, as illustrated in Fig 1. This approach was inspired by work on weakly supervised video classification [15]. The specific features were based on [16]: visible honeycomb patterns corresponding to viable keratinocytes, visible bright bands corresponding to basal cells/dermal papillae and finally the absence of features of the basal layer -or- clearly visible fibrillar structures in the dermis. The en-face section with the first visible honeycomb pattern was considered the first layer of the viable epidermis—all sections above this layer were labelled as stratum corneum. En-face sections from the first honeycomb pattern to the first dermal papillae were labelled as viable epidermis. Sections from the first dermal papillae and up to, but not including, the en-face section where the features of the basal layer were no longer visible were labelled as dermal-epidermal junction. All sections below these were labelled as papillary dermis.


Automated Segmentation of Skin Strata in Reflectance Confocal Microscopy Depth Stacks.

Hames SC, Ardigò M, Soyer HP, Bradley AP, Prow TW - PLoS ONE (2016)

RCM features used in identification of the start and end of each anatomical strata.A) The start of the viable epidermis is marked by the first visible honeycomb pattern (Blue outline). B) The start of the dermal-epidermal junction is marked by the first clearly defined papillae with bright ring of basal cells surrounding a dark centre (Blue arrows). C) The start of the papillary dermis is marked by the absence of features corresponding to the basal layer. D) Alternative definition of the start of the dermal-epidermal junction for light skin types, a dark featureless patch surrounded by keratinocytes with the absence of a bright basal layer (Blue arrow).
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4835045&req=5

pone.0153208.g001: RCM features used in identification of the start and end of each anatomical strata.A) The start of the viable epidermis is marked by the first visible honeycomb pattern (Blue outline). B) The start of the dermal-epidermal junction is marked by the first clearly defined papillae with bright ring of basal cells surrounding a dark centre (Blue arrows). C) The start of the papillary dermis is marked by the absence of features corresponding to the basal layer. D) Alternative definition of the start of the dermal-epidermal junction for light skin types, a dark featureless patch surrounded by keratinocytes with the absence of a bright basal layer (Blue arrow).
Mentions: For the purposes of segmentation four distinct anatomical strata were identified in the skin: the stratum corneum, viable epidermis, dermal-epidermal junction and the papillary dermis. To simplify the problem of establishing a ground truth in a large number of stacks it was assumed that a single anatomical strata was present in each en-face section and the extent of each strata was determined by identifying where three distinct features first occured within the stack, as illustrated in Fig 1. This approach was inspired by work on weakly supervised video classification [15]. The specific features were based on [16]: visible honeycomb patterns corresponding to viable keratinocytes, visible bright bands corresponding to basal cells/dermal papillae and finally the absence of features of the basal layer -or- clearly visible fibrillar structures in the dermis. The en-face section with the first visible honeycomb pattern was considered the first layer of the viable epidermis—all sections above this layer were labelled as stratum corneum. En-face sections from the first honeycomb pattern to the first dermal papillae were labelled as viable epidermis. Sections from the first dermal papillae and up to, but not including, the en-face section where the features of the basal layer were no longer visible were labelled as dermal-epidermal junction. All sections below these were labelled as papillary dermis.

Bottom Line: 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.

View Article: PubMed Central - PubMed

Affiliation: Dermatology Research Centre, The University of Queensland, School of Medicine, Translational Research Institute, Brisbane, Australia.

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.

No MeSH data available.


Related in: MedlinePlus