<i>In vivo</i> reflectance confocal microscopy: automated diagnostic image analysis of melanocytic skin tumours

Koller, Sílvia Helena; Wiltgen, Marco; Ahlgrimm‐Siess, Verena; Weger, Wolfgang; Hofmann‐Wellenhof, Rainer; Richtig, Erika; Smolle, Josef; Gerger, Armin

doi:10.1111/j.1468-3083.2010.03834.x

Cited by 55 publications

(43 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To address the need for training, machine learning-based image analysis is being investigated to provide quantitative and objective approaches for reading images. (Gareau et al, 2010; Koller et al, 2011; Kurugol et al, 2011; Wiltgen et al, 2008) One of the first studies reported a method based on texture analysis for automated identification of diagnostically significant regions in RCM images of melanocytic lesions(Koller et al, 2011; Wiltgen et al, 2008). Another group of researchers developed a method to automatically quantify the spread of pagetoid melanocytes in epidermis and disarray at the DEJ level in order to detect superficial spreading melanomas (Gareau et al, 2010).…”

Section: Introductionmentioning

confidence: 99%

Automated Delineation of Dermal–Epidermal Junction in Reflectance Confocal Microscopy Image Stacks of Human Skin

Kurugol

Kose

Park

et al. 2015

Journal of Investigative Dermatology

View full text Add to dashboard Cite

Reflectance confocal microscopy (RCM) images skin non-invasively, with optical sectioning and nuclear-level resolution comparable to that of pathology. Based on assessment of the dermal-epidermal junction (DEJ) and morphologic features in its vicinity, skin cancer can be diagnosed in vivo with high sensitivity and specificity. However, the current visual, qualitative approach for reading images leads to subjective variability in diagnosis. We hypothesize that machine learning-based algorithms may enable a more quantitative, objective approach. Testing and validation was performed with two algorithms that can automatically delineate the DEJ in RCM stacks of normal human skin. The test set was composed of 15 fair and 15 dark skin stacks (30 subjects) with expert labellings. In dark skin, in which the contrast is high due to melanin, the algorithm produced an average error of 7.9±6.4μm. In fair skin, the algorithm delineated the DEJ as a transition zone, with average error of 8.3±5.8μm for the epidermis-to-transition zone boundary and 7.6±5.6μm for the transition zone-to-dermis. Our results suggest that automated algorithms may quantitatively guide the delineation of the DEJ, to assist in objective reading of RCM images. Further development of such algorithms may guide assessment of abnormal morphological features at the DEJ.

show abstract

Section: Introductionmentioning

confidence: 99%

Automated Delineation of Dermal–Epidermal Junction in Reflectance Confocal Microscopy Image Stacks of Human Skin

Kurugol

Kose

Park

et al. 2015

Journal of Investigative Dermatology

View full text Add to dashboard Cite

show abstract

“…These assessment approaches have focused on one of three main categories: 1) Quantifying and detecting specific features such as counting keratinocytes [5], detecting pagetoid cells [6] and evaluating photoageing [7], 2) Computer aided diagnosis of malignant melanocytic lesions [8] and 3) Identifying the anatomical structures of human skin [9,10]. Although both Somoza et al and Kurugol et al considered the problem of understanding human skin their work is limited: Kurugol et al consider only the location of the dermal-epidermal junction and showed good performance only in darker skin types: 89% of the epidermis and 87% of the dermis were correctly classified in dark skin, whereas in light skin only 64% and 75% of the epidermis and dermis were correctly classified.…”

Section: Introductionmentioning

confidence: 99%

Automated Segmentation of Skin Strata in Reflectance Confocal Microscopy Depth Stacks

et al. 2016

View full text Add to dashboard Cite

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.

show abstract

“…A direct comparison of dermoscopy to RCM found no significant difference between the sensitivity and specificity for melanoma between these two methods [56]. Attempts have also been made to integrate automated computer analysis with RCM images and have resulted in 93.6% sensitivity and 90.4% specificity for melanoma detection in learning sets [66]. An additional strength of RCM is the ability to identify amelanotic lesions [67] and early melanoma in situ with a sensitivity of 85% and specificity of 76% for these tumors [68,69].…”

Section: Reflectance Confocal Microscopy and The Vivascopementioning

confidence: 99%

Screening and Non-Invasive Evaluative Devices for Melanoma Detection A Comparison of Commercially Available Devices and Dermoscopic Evaluation

Grossman¹

2015

CDT

View full text Add to dashboard Cite

A number of commercially available devices have been developed to help dermatologists distinguish benign pigmented lesions from malignant melanoma, but it is unclear if and how these technologies should be adopted into clinical practice. This review summarizes the reported diagnostic accuracies of these devices and compares them to clinical exam and dermoscopy. Screening devices appear to have the highest utility in patient populations with a high prevalence of melanoma. However, even in high-risk patients the diagnostic accuracy of evaluative devices is not clearly superior to dermatologists skilled with dermoscopy, potentially limiting their cost effectiveness. General practitioners not skilled with dermoscopy are likely to have larger improvements in diagnostic accuracy when using evaluative devices, however further investigation into the best way to employ them is needed.

show abstract

In vivo reflectance confocal microscopy: automated diagnostic image analysis of melanocytic skin tumours

Abstract: The automated RCM image analysis procedure holds promise for further investigations. However, to date our system cannot be applied to routine skin tumour screening.

Cited by 55 publications

References 24 publications

Automated Delineation of Dermal–Epidermal Junction in Reflectance Confocal Microscopy Image Stacks of Human Skin

Automated Delineation of Dermal–Epidermal Junction in Reflectance Confocal Microscopy Image Stacks of Human Skin

Automated Segmentation of Skin Strata in Reflectance Confocal Microscopy Depth Stacks

Screening and Non-Invasive Evaluative Devices for Melanoma Detection A Comparison of Commercially Available Devices and Dermoscopic Evaluation

Contact Info

Product

Resources

About