Computer-Aided Analysis of Nailfold Capillaroscopy Images

Abstract

Nailfold capillaroscopy (NC) is a non-invasive imaging technique employed to assess the condition of blood capillaries in the nailfold, and is routinely used for the detection of scleroderma spectral disorders, Raynaud's phenomenon and other connective tissue diseases. In this chapter, we present computer-aided approaches for capillary inspection, in particular focussing on the tasks of image enhancement, binarisation and skeletonisation. We evaluate the performance of a number image enhancement/noise removal techniques for NC images, as a pre-cursor to edge detection aimed at identifying capillaries. Results show that bilateral filters and enhancers provide the best overall image quality. Following noise removal, NC images typically get converted into binary form. For this purpose, we employed a difference-of-Gaussian approach before thresholding. The final stage is that of skeletonisation, which can be effectively performed using a rule-based thinning algorithm. Thus the complete imaging pipeling of pre-processing, binarisation and skeletonisation is represented in this chapter.