UserratedPatch2021

This is a data set that we created in collaboration with the UMCG and used it to evaluate a new method published in the paper "Use of convolutional neural networks for the detection of u-serrated patterns in direct immunofluorescence images to facilitate the diagnosis of epidermolysis bullosa acquisita", American Journal of Pathology. The data set consists of 220,800 DIF image patches (of size 50 x 50 pixels) extracted from 106 images of 46 patients (4,800 patches per each patient). The half of the dataset is u-serrated patterns from DIF biopsies of 23 patients with confirmed diagnoses of EBA and the other half is non-u-serrated patterns from DIF biopsies of the other 23 patients with other pemphigoid diseases. Different staining intensities and slide thicknesses (4 µm, 6 µm, and 8 µm) were included to generate the dataset, representing routine diagnostics. All DIF slides were cut and stained at the Immunodermatology Laboratory of the Center for Blistering Diseases, University Medical Center Groningen, the Netherlands. DIF images were taken under standardized settings with the same microscope (Leica DMRA, Leica, Wetzlar, Germany) and 40x magnification dry objective (Leica HCX PL Fluotar 40x/0.75, Leica, Wetzlar, Germany) and 10x magnification ocular. (2021