A quantitative description of echographic images of sclerotic skin in patients with systemic sclerosis, as assessed by computerized image analysis on 20 MHz B-scan recordings.
Seidenari S, Belletti B, Conti A
DOI: 10.2340/0001555576361364
Abstract
The aim of our study was to find image descriptors enabling the characterization of sclerotic skin and its differentiation from normal skin, in order to find an objective method for the assessment of skin involvement in systemic sclerosis (SSc). Echographic evaluations were carried out using a 20 MHz B-scanner, on 18 female patients with SSc and on 20 healthy women serving as controls, at 3 different skin sites (forehead, cheek and back of the hand). Images were processed by a program, based on segmentation procedures and object description, employing 5 different amplitude bands and the following parameters: 1) the extension of image areas marked by amplitude bands of interest, 2) the percentage of the image surface reflecting within a homogeneous amplitude band, 3) the number of objects composing the image, 4) the average object size, and 5) the "density" of the objects. At all 3 skin sites, marked differences in the echostructure of the tissue between patients with SSc and the controls were observable. In SSc patients forehead skin appeared thinner and more echogenic, with smaller hypo-reflecting objects and greater hyper-reflecting areas; cheek skin showed an increase in intermediate-high amplitude components, with greater and more numerous hyper-reflecting objects, and smaller and less numerous hypo-reflecting ones; the skin on the back of the hand was thicker, less echogenic, with large hypo-reflecting areas and small hyper-reflecting objects. By image processing these parameters were numerically described. Values referring to sclerotic skin significantly differed from those of normal skin. This echographic procedure is proposed as a method representing a first step towards the quantification of the spontaneous course of SSc and of response to therapy.
Significance
Supplementary content
Comments