Automated scoring of vitiligo using superpixel-generated computerized digital image analysis of clinical photographs: a novel and consistent way to score vitiligo

Non-invasive skin measurement methods and diagnostics for vitiligo: a systematic review

Vitiligo is a multifaceted autoimmune depigmenting disorder affecting around 0.5 to 2.0% of individuals globally. Standardizing diagnosis and therapy tracking can be arduous, as numerous clinical evaluation methods are subject to interobserver variability and may not be validated. Therefore, there is a need for diagnostic tools that are objective, dependable, and preferably non-invasive.

Aims

This systematic review provides a comprehensive overview of the non-invasive objective skin measurement methods that are currently used to evaluate the diagnosis, severity, and progression of vitiligo, as well as the advantages and limitations of each technique.

Methods

The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) checklist was used for the systematic review. Scopus, Embase, Cochrane Library, and Web of Science databases were comprehensively searched for non-invasive imaging and biophysical skin measuring methods to diagnose, evaluate the severity of, or monitor the effects of vitiligo treatment. The risk of bias in included articles was assessed using the QUADAS-2 quality assessment scale.

Results

An extensive literature search resulted in 64 studies for analysis, describing eight imaging techniques (reflectance confocal microscopy, computer-aided imaging analysis, optical coherence tomography, infrared photography, third-harmonic generation microscopy, multiphoton microscopy, ultraviolet light photography, and visible light/digital photograph), and three biophysical approaches (dermoscopy, colorimetry, spectrometry) used in diagnosing and assessing vitiligo. Pertinent information about functionality, mechanisms of action, sensitivity, and specificity was obtained for all studies, and insights into the strengths and limitations of each diagnostic technique were addressed. Methodological study quality was adequate; however, statistical analysis was not achievable because of the variety of methods evaluated and the non-standardized reporting of diagnostic accuracy results.

Conclusions

The results of this systematic review can enhance clinical practice and research by providing a comprehensive overview of the spectrum of non-invasive imaging and biophysical techniques in vitiligo assessment. Studies with larger sample sizes and sound methodology are required to develop verified methods for use in future practice and research.

Systematic review registration

(PROSPERO) database, (CRD42023395996).

Image Analysis Systems to Calculate the Surface Area of Vitiligo Lesions: a Systematic Review of Measurement Properties

Several digital image analysis systems have been developed for surface calculation of vitiligo lesions. Critical assessment of their measurement properties is crucial to support evidence-based recommendations on the most suitable instruments and will reveal the need for future research. A systematic review was performed to systematically summarize, compare, and critically assess the measurement properties of digital and analogue analysis systems for surface calculation of vitiligo lesions following the Consensus-Based Standards for the Selection of Health Measurement Instruments (COSMIN) recommendations. Nineteen clinical trials were selected including 25 different instruments. Manual tracing on transparent sheets (contact planimetry) combined with digital measurement or point counting can be considered as the best validated method for the evaluation of target lesions taking into account the skin curvatures. Two-dimensional digital imaging analysis on photographs seems also robust although confirmatory data of different research groups using the same digital instrument in a wide range of skin types are missing. Analysis based on 3D photography is still in its early stage but is promising for whole-body analysis. However, the reported data on the quality of the instruments for surface area calculation of vitiligo lesions was in general rather limited. Therefore, future high-quality validation studies are required also including full body evaluations.

Segmenting Vitiligo on Clinical Face Images Using CNN Trained on Synthetic and Internet Images

Accurately diagnosing and describing the severity of vitiligo is crucial for prognostication, treatment selection and comparison. Currently, disease severity scores require dermatologists to estimate percentage area of involvement, which is subjected to inter and intra-assessor variability. Previous studies focus on pure skin but vitiligo on the face, which has a more serious impact on patients' quality of life, was completely neglected. Convolutional neural networks (CNNs) have good performance on many segmentation tasks. However, due to data privacy, it is hard to have a large clinical vitiligo face image dataset to train a CNN. To address this challenge, images from two different sources, the Internet and the proposed vitiligo face synthesis algorithm, are employed in training. 843 vitiligo images taken from different viewpoints were collected from the Internet. These images are hugely different from the target clinical images collected according to a newly established international standard. To have more vitiligo face images similar to the target clinical images to enhance segmentation performance, an image synthesis algorithm is proposed. Both synthetic and Internet images are used to train a CNN which is modified from the fully convolutional network (FCN) to segment face vitiligo lesions. The results show that 1) the synthetic images effectively improve segmentation performance; 2) the proposed algorithm achieves 1.06 % error for the face vitiligo area estimation and 3) it is more accurate than two dermatologists and all the previous automated vitiligo segmentation methods, which were designed for segmentation vitiligo on pure skin.

Automated severity scoring of atopic dermatitis patients by a deep neural network

Scoring atopic dermatitis (AD) severity with the Eczema Area and Severity Index (EASI) in an objective and reproducible manner is challenging. Automated measurement of erythema, papulation, excoriation, and lichenification severity using images has not yet been investigated. Our aim was to determine whether convolutional neural networks (CNNs) could assess erythema, papulation, excoriation, and lichenification severity at a level of competence comparable to dermatologists. We created a standard dataset of 8,000 clinical images showing AD. Each component of the EASI was scored from 0 to 3 by three dermatologists. We trained four CNNs (ResNet V1, ResNet V2, GoogLeNet, and VGG-Net) with the image dataset and determined which CNN was the most suitable for erythema, papulation, excoriation, and lichenification scoring. The brightness of the images in each dataset was adjusted to − 80% to + 80% of the original brightness (i.e., 9 levels by 20%) to investigate if the CNNs accurately measured scores if image brightness levels were changed. Compared to the dermatologists’ scoring, accuracy rates of the CNNs were 99.17% for erythema, 93.17% for papulation, 96.00% for excoriation, and 97.17% for lichenification. CNNs trained with brightness-adjusted images achieved a high accuracy without the need to standardize camera settings. These results suggested that CNNs perform at level of competence comparable to dermatologists for scoring erythema, papulation, excoriation, and lichenification severity.

Mechanisms of melanocyte death in vitiligo

Vitiligo is an autoimmune depigment disease results from extensive melanocytes destruction. The destruction of melanocyte is thought to be of multifactorial causation. Genome-wide associated studies have identified single-nucleotide polymorphisms in a panel of susceptible loci as risk factors in melanocyte death. But vitiligo onset can't be solely attributed to a susceptive genetic background. Oxidative stress triggered by elevated levels of reactive oxygen species accounts for melanocytic molecular and organelle dysfunction, a minority of melanocyte demise, and melanocyte-specific antigens exposure. Of note, the self-responsive immune function directly contributes to the bulk of melanocyte deaths in vitiligo. The aberrantly heightened innate immunity, type-1-skewed T helper, and incompetent regulatory T cells tip the balance toward autoreaction and CD8+ cytotoxic T lymphocytes finally execute the killing of melanocytes, possibly alarmed by resident memory T cells. In addition to the well-established apoptosis and necrosis, we discuss several death modalities like oxeiptosis, ferroptosis, and necroptosis that are probably employed in melanocyte destruction. This review focuses on the various mechanisms of melanocytic death in vitiligo pathogenesis to demonstrate a panorama of that. We hope to provide new insights into vitiligo pathogenesis and treatment strategies by the review.

Proposing a novel unsupervised stack ensemble of deep and conventional image segmentation (SEDCIS) method for localizing vitiligo lesions in skin images

BACKGROUND

Vitiligo is an acquired pigmentary skin disorder characterized by depigmented macules and patches which brings many challenges for the patients suffering from. For vitiligo severity assessment, several scoring methods have been proposed based on morphometry and colorimetry. But, all methods suffer from much inter- and intra-observer variations for estimating the depigmented area. For all mentioned assessment methods of vitiligo disorder, accurate segmentation of the skin images for lesion detection and localization is required. The image segmentation for localizing vitiligo skin lesions has many challenges because of illumination variation, different shapes and sizes of vitiligo lesions, vague lesion boundaries and skin hairs and vignette effects. The manual image segmentation is a tedious and time-consuming task. Therefore, using automatic image segmentation methods for lesion detection is necessarily required.

MATERIALS AND METHODS

In this study, a novel unsupervised stack ensemble of deep and conventional image segmentation (SEDCIS) methods is proposed for localizing vitiligo lesions in skin images. Unsupervised segmentation methods do not require prior manual segmentation of vitiligo lesions which is a tedious and time-consuming task with intra- and inter-observer variations.

RESULTS

Our collected dataset includes 877 images taken from 21 patients with the resolution of 5760*3840 pixels suffering from vitiligo disorder. Experimental results show that SEDCIS outperforms the compared methods with accuracy of 97%, sensitivity of 98%, specificity of 96%, area overlapping of 94%, and Dice index of 97%.

CONCLUSION

The proposed method can segment vitiligo lesions with highly reasonable performance and can be used for assessing the vitiligo lesion surface.

Reference method for digital surface measurement of target lesions in vitiligo: a comparative analysis

BACKGROUND

Objective measurement of target lesions in vitiligo is important for clinical practice and trials, yet no preferred tool has been defined. Reported digital tools have shortcomings related to feasibility aspects and often lack information on validity, reliability and responsiveness. Moreover, studies are not yet based on ultraviolet (UV) photography.

OBJECTIVES

To assess the reliability, validity and feasibility of two functions in ImageJ for measurement of target lesions, based on three different types of images including UV pictures.

METHODS

Planimetric measurements were performed on photographs with and without UV, and lesion contours on transparent sheets of 52 vitiligo lesions from 10 patients with vitiligo. The ImageJ functions 'wand' and 'threshold' were used by three and four assessors, respectively. Inter- and intrarater reliability, hypothesis testing for construct validity, and feasibility were evaluated.

RESULTS

The inter- and intrarater reliability for the 'wand' and 'threshold' functions were excellent [intraclass correlation coefficient (ICC) > 0·9] for measurement on pictures (with or without UV). The highest agreement (ICC > 0·95) and lowest variance were obtained for measurements on transparent sheets. All four hypotheses for construct validity were confirmed for all measurements. Overall, all measurement methods scored satisfactorily for user-friendliness. However, measurements on transparent sheets were preferred and the completion time was significantly faster.

CONCLUSIONS

This study confirmed the reliability, validity and feasibility of two functions in ImageJ to measure target lesions in vitiligo. Based on the feasibility and included three-dimensional aspects, transparent sheets measured with the ImageJ 'wand' function can be proposed for future trials as a reference method to investigate the criterion validity of other digital instruments.