Modelling of edge profiles in pigmented skin lesions

The sharpness of the lesion boundary and the contrast between the lesion and the surrounding skin provide important diagnostic information in the assessment of pigmented skin lesions. This paper presents a new method for computing these parameters by employing an edge model based on a sigmoid function. For each radial profile of the lesion, optimal parameters are found by using an iterative least-squares method. The level and nature of variability of these parameters along the boundary will be correlated with lesion diagnosis in a future study. The method also returns the location equivalent to “zero-crossing” for each profile, thus producing a lesion outline. Additionally, the spread of the pigment beyond the zero-crossing point is computed, thus showing the lesion extent.