Digital Image Analysis of Vitiligo for Monitoring of Vitiligo Treatment

• Main Author: Nugroho, Hermawan
• Format: Thesis
• Language: English
• Published: 2007
• Subjects: TK Electrical engineering. Electronics Nuclear engineering
• Online Access: http://utpedia.utp.edu.my/8285/1/2008%20Master%20-%20Digital%20Analysis%20of%20Vitiligo%20for%20Monitoring%20of%20Vitiligo%20Treatment.pdf; http://utpedia.utp.edu.my/8285/

Vitiligo is an acquired pigmentary skin disorder characterized by depigmented macules that result from damage to and destruction of epidermal melanocytes. Visually, the vitiligous areas are paler in contrast to normal skin or completely white due to the lack of pigment melanin. The course of vitiligo is unpredictable where the vitiligous skin lesions may remain stable for years before worsening. Vitiligo treatments have two objectives, to arrest disease progression and to re-pigment the vitiligous skin lesions. To monitor the efficacy of the treatment, dermatologists observe the disease directly, or indirectly using digital photos. Currently there is no objective method to determine the efficacy of the vitiligo treatment. Physician's Global Assessment (PGA) scale is the current scoring system used by dermatologists to evaluate the treatment. The scale is based on the degree of repigmentation within lesions over time. This quantitative tool however may not be help to detect slight changes due to treatment as it would still be largely dependent on the human eye and judgment to produce the scorings. In addition, PGA score is also subjective, as it varies with dermatologists. The progression of vitiligo treatment can be very slow and can take more than 6 months. It is observed that dermatologists find it visually hard to determine the areas of skin repigmentation due to this slow progress and as a result the observations are made after a longer time frame. The objective of this research is to develop a tool that enables dermatologists to determine and quantify areas of repigmentation objectively over a shorter time frame during treatment. The approaches towards achieving this objective are based on digital image processing techniques. Skin color is due to the combination of skin histological parameters, namely pigment melanin and haemoglobin. However in digital imaging, color is produced by combining three different spectral bands, namely red, green, and blue (RGB). It is believed that the spatial distribution of melanin and haemoglobin in skin image could be separated. It is found that skin color distribution lies on a two-dimensional melanin-haemoglobin color subspace. In order to determine repigmentation (due to pigment melanin) it is necessary to perform a conversion from RGB skin image to this two-dimensional color subspace. Using principal component analysis (PCA) as a dimensional reduction tool, the two-dimensional subspace can be represented by its first and second principal components. Independent component analysis is employed to convert the twodimensional subspace into a skin image that represents skin areas due to melanin and haemoglobin only. In the skin image that represents skin areas due to melanin, vitiligous skin lesions are identified as skin areas that lack melanin. Segmentation is performed to separate the healthy skin and the vitiligous lesions. The difference in the vitiligous surface areas between skin images before and after treatment will be expressed as a percentage of repigmentation in each vitiligo lesion. This percentage will represent the repigmentation progression of a particular body region. Results of preliminary and pre-clinical trial study show that our vitiligo monitoring system has been able to determine repigmentation progression objectively and thus treatment efficacy on a shorter time cycle. An intensive clinical trial is currently undertaken in Hospital Kuala Lumpur using our developed system. VII