In Vivo Reflectance Confocal Microscopy: Emerging Role in Noninvasive Diagnosis and Monitoring of Eczematous Dermatoses

Automating reflectance confocal microscopy image analysis for dermatological research: a review

SIGNIFICANCE

Reflectance confocal microscopy (RCM) is a noninvasive, in vivo technology that offers near histopathological resolution at the cellular level. It is useful in the study of phenomena for which obtaining a biopsy is impractical or would cause unnecessary tissue damage and trauma to the patient.

AIM

This review covers the use of RCM in the study of skin and the use of machine learning to automate information extraction. It has two goals: (1) an overview of information provided by RCM on skin structure and how it changes over time in response to stimuli and in disease and (2) an overview of machine learning approaches developed to automate the extraction of key morphological features from RCM images.

APPROACH

A PubMed search was conducted with additional literature obtained from references lists.

RESULTS

The application of RCM as an in vivo tool in dermatological research and the biologically relevant information derived from it are presented. Algorithms for image classification to epidermal layers, delineation of the dermal-epidermal junction, classification of skin lesions, and demarcation of individual cells within an image, all important factors in the makeup of the skin barrier, were reviewed. Application of image analysis methods in RCM is hindered by low image quality due to noise and/or poor contrast. Use of supervised machine learning is limited by time-consuming manual labeling of RCM images.

CONCLUSIONS

RCM has great potential in the study of skin structures. The use of artificial intelligence could enable an easier, more reproducible, precise, and rigorous study of RCM images for the understanding of skin structures, skin barrier, and skin inflammation and lesions. Although several attempts have been made, further work is still needed to provide a definite gold standard and overcome issues related to image quality, limited labeled datasets, and lack of phenotype variability in available databases.

Irritant Contact Dermatitis — a Review

Purpose of Review

Recent Findings

Summary

Dermoscopy, confocal microscopy and optical coherence tomography features of main inflammatory and autoimmune skin diseases: A systematic review

BACKGROUND/OBJECTIVES

Non-invasive skin imaging features of main skin inflammatory and autoimmune diseases have been reported, although a comprehensive review of their correlation with histopathologic features is currently lacking. Therefore, the aim of this paper was to review the correlation of dermoscopic, reflectance confocal microscopy (RCM) and optical coherence tomography (OCT) criteria of main inflammatory and autoimmune skin diseases with their corresponding histopathologic criteria correlation.

METHODS

Studies on human subjects affected by main inflammatory and autoimmune diseases, defining the correlation of dermoscopic, RCM or OCT with histopathologic criteria, were included in the review. Five groups of diseases were identified and described: psoriasiform, spongiotic and interface dermatitis, bullous diseases and scleroderma.

RESULTS

Psoriasiform dermatitis was typified by white scales, corresponding to hyperkeratosis, and vessels, observed with RCM and OCT. Spongiosis, corresponding to dark areas within the epidermis with RCM and OCT, was the main feature of spongiotic dermatitis. Interface dermatitis was characterised by dermoepidermal junction obscuration. Blisters, typical of bullous diseases, were visualised as dark areas with RCM and OCT while scleroderma lesions were characterised by dermoscopic fibrotic beams, related to dermal thickness variations, with specific OCT and histopathologic correlations.

CONCLUSIONS

Although the role of RCM and OCT has yet to be defined in clinical practice, non-invasive skin imaging shows promising results on inflammatory and autoimmune skin diseases, due to the correlation with histopathologic features.

Non-invasive intradermal imaging of cystine crystals in cystinosis

IMPORTANCE

Development of noninvasive methodology to reproducibly measure tissue cystine crystal load to assess disease status and guide clinical care in cystinosis, an inherited lysosomal storage disorder characterized by widespread cystine crystal accumulation.

OBJECTIVE

To develop an unbiased and semi-automated imaging methodology to quantify dermal cystine crystal accumulation in patients to correlate with disease status.

DESIGN, SETTING AND PARTICIPANTS

101 participants, 70 patients and 31 healthy controls, were enrolled at the University of California, San Diego, Cystinosis Clinics, Rady Children's Hospital, San Diego and at the annual Cystinosis Research Foundation family conference for an ongoing prospective longitudinal cohort study of cystinosis patients with potential yearly follow-up.

EXPOSURES

Intradermal reflectance confocal microscopy (RCM) imaging, blood collection via standard venipuncture, medical record collection, and occasional skin punch biopsies.

MAIN OUTCOMES AND MEASURES

The primary outcome was to establish an automated measure of normalized confocal crystal volume (nCCV) for each subject. Secondary analysis examined the association of nCCV with various clinical indicators to assess nCCV's possible predictive potential.

RESULTS

Over 2 years, 57 patients diagnosed with cystinosis (median [range] age: 15.1 yrs [0.8, 54]; 41.4% female) were intradermally assessed by RCM to produce 84 image stacks. 27 healthy individuals (38.7 yrs [10, 85]; 53.1% female) were also imaged providing 37 control image stacks. Automated 2D crystal area quantification revealed that patients had significantly elevated crystal accumulation within the superficial dermis. 3D volumetric analysis of this region was significantly higher in patients compared to healthy controls (mean [SD]: 1934.0 μm3 [1169.1] for patients vs. 363.1 μm3 [194.3] for controls, P<0.001). Medical outcome data was collected from 43 patients with infantile cystinosis (media [range] age: 11 yrs [0.8, 54]; 51% female). nCCV was positively associated with hypothyroidism (OR = 19.68, 95% CI: [1.60, 242.46], P = 0.02) and stage of chronic kidney disease (slope estimate = 0.53, 95%CI: [0.05, 1.00], P = 0.03).

CONCLUSIONS AND RELEVANCE

This study used non-invasive RCM imaging to develop an intradermal cystine crystal quantification method. Results showed that cystinosis patients had increased nCCV compared to healthy controls. Level of patient nCCV correlated with several clinical outcomes suggesting nCCV may be used as a potential new biomarker for cystinosis to monitor long-term disease control and medication compliance.

Melanoma diagnosis using deep learning techniques on dermatoscopic images

BACKGROUND

Melanoma has become more widespread over the past 30 years and early detection is a major factor in reducing mortality rates associated with this type of skin cancer. Therefore, having access to an automatic, reliable system that is able to detect the presence of melanoma via a dermatoscopic image of lesions and/or skin pigmentation can be a very useful tool in the area of medical diagnosis.

METHODS

Among state-of-the-art methods used for automated or computer assisted medical diagnosis, attention should be drawn to Deep Learning based on Convolutional Neural Networks, wherewith segmentation, classification and detection systems for several diseases have been implemented. The method proposed in this paper involves an initial stage that automatically crops the region of interest within a dermatoscopic image using the Mask and Region-based Convolutional Neural Network technique, and a second stage based on a ResNet152 structure, which classifies lesions as either "benign" or "malignant".

RESULTS

Training, validation and testing of the proposed model was carried out using the database associated to the challenge set out at the 2017 International Symposium on Biomedical Imaging. On the test data set, the proposed model achieves an increase in accuracy and balanced accuracy of 3.66% and 9.96%, respectively, with respect to the best accuracy and the best sensitivity/specificity ratio reported to date for melanoma detection in this challenge. Additionally, unlike previous models, the specificity and sensitivity achieve a high score (greater than 0.8) simultaneously, which indicates that the model is good for accurate discrimination between benign and malignant lesion, not biased towards any of those classes.

CONCLUSIONS

The results achieved with the proposed model suggest a significant improvement over the results obtained in the state of the art as far as performance of skin lesion classifiers (malignant/benign) is concerned.