Skin lesion segmentation using deep convolution networks guided by local unsupervised learning

摘要

A dermoscopic lesion area is identified by: Obtaining a dermoscopic image and running a convolutional neural network image classifier on the dermoscopic image to obtain pixelwise lesion prediction scores. Segmenting the dermoscopic image into super-pixels, and computing for each super-pixel an average of the pixelwise prediction scores for pixels within that super-pixel. Computing a mean prediction score across the plurality of super-pixels. Assigning a confidence indicator of “1” to each super-pixel with a prediction score equal or greater than the mean prediction score, and a confidence indicator of “0” to each super-pixel with a prediction score less than the mean prediction score. Constructing a super-pixel graph G=(V,E,W) wherein; $$ w_{\mathrm{ij}}=\exp \left(-\frac{{x_i-x_j}^2}{\sigma }\right)\mathrm{and}{d}_i=\sum _{i=1}^N{w}_{\mathrm{ij}};$$computing a confidence score function F according to {circumflex over (F)}=arg min(F^TLF+μ∥F−Y∥²); and integrating the confidence score function F with the pixelwise prediction scores to produce a final segmentation of the dermoscopic image into lesion and background areas.