BORDER DETECTION OF SKIN CANCER CELLS WITH FRACTAL DIMENSION
Abstract
In this paper we study a model of skin cancer (MM) in vitro, using geometry of fractals as the method of analysis. The fractal dimensions of moles (skin cancer cells) growth pattern have been measured by using the methods of Box-counting method (DB) and Sausage method (DS). The cell growth of this cancer can be modeled by Hidden Markov model (HMM) and percolation model which are depending upon the time complexity. From these models we can find the shape of the irregularity border by using the probability distribution of the cells. The variation in the irregular border of the skin cancer has been found out using ANOVA test and cell's compactness. The fractal approach led to very promising results which improved the determination and examination of the stage of skin cancer.