Melanoma Poem Analysis

Melanoma is a cancer of the melanocytes, the cell found in the skin's epidermis that produces melanin. Melanoma most commonly occurs on the trunk or lower extremities. While malignant melanoma is less common than non-melanoma skin cancer , it is considered the most deadly form of skin cancer. This is because melanoma accounts for approximately 75% of deaths associated with skin cancer. In 2013, it is estimated that 76,690 people will be diagnosed with melanoma and 9,480 people will die of melanoma in the United States. Skin lesions are visually screened for melanoma. Visual algorithms[1-3] that doctors use as a guide to assess skin lesions include the ABCD scale . For example, the ABCD scale is an acronym for asymmetry, border irregularity,

This novel skin lesion segmentation algorithm is designed to be used for images taken by a digital camera. The segmentation algorithm uses a set of learned texture distributions and their texture distinctiveness metric (TD metric). The representative texture distributions used to identify pixels that belong to the lesion and skin classes and to find the border of the skin lesion. The proposed segmentation algorithm is referred to as the Texture Distinctiveness Lesion Segmentation

After, the color space transformation we are going to extracts the texture vector from that image using sparse texture model. The texture vectors are represented as a set of distributions which is used to cluster the texture data using K-means clustering algorithm. Finding the number of clusters which consists set of texture distributions used to calculate TD metric. After, calculating the TD metric, the image is over segmented using SRM algorithm, which results the image being divided into large number of regions. Next, each region is independently classified as representing normal skin or lesion based on the textural contents of that region. Finally, to apply morphological dilation method to refine the lesion borders. Figure 3 shows an overall system