Automatic Classification of Serrated Patterns in Direct Immunofluorescence Images

Direct immunofluorescence (DIF) images are used by clinical experts for the diagnosis of autoimmune blistering diseases. The analysis of serration patterns in DIF images concerns two types of patterns, namely n- and u-serrated. Manual analysis is time-consuming and challenging due to noise. We propose an algorithm for the automatic classification of serrated patterns in DIF images. We first segment the epidermal basement membrane zone (BMZ) where n- and u-serrated patterns are typically found. Then, we apply a bank of B-COSFIRE filters to detect ridges and determine their orientations with respect to the BMZ. Finally, we classify an image by comparing its normalized histogram of relative orientations with those of the training images using a nearest neighbor approach. We achieve a recognition rate of 84.4% on a UMCG data set of 416 DIF images, which is comparable to 83.4% by clinical experts.

[1]  Nicolai Petkov,et al.  Computational models of visual neurons specialised in the detection of periodic and aperiodic oriented visual stimuli: bar and grating cells , 1997, Biological Cybernetics.

[2]  Lambert Schomaker,et al.  Writer identification using directional ink-trace width measurements , 2012, Pattern Recognit..

[3]  R. Vodegel,et al.  Clinical and Laboratory Investigations U-serrated immunodeposition pattern differentiates type VII collagen targeting bullous diseases from other subepidermal bullous autoimmune diseases , 2004 .

[4]  M. Hertl,et al.  Immunofluorescence serration pattern analysis as a diagnostic criterion in antilaminin‐332 mucous membrane pemphigoid: immunopathological findings and clinical experience in 10 Dutch patients , 2011, The British journal of dermatology.

[5]  Nicolai Petkov,et al.  Suppression of contour perception by band-limited noise and its relation to nonclassical receptive field inhibition , 2003, Biological cybernetics.

[6]  George Azzopardi,et al.  Trainable COSFIRE filters for vessel delineation with application to retinal images , 2015, Medical Image Anal..

[7]  Chenyu Shi,et al.  Automatic Differentiation of u- and n-serrated Patterns in Direct Immunofluorescence Images , 2015, CAIP.

[8]  Nikolay Petkov,et al.  Biologically motivated computationally intensive approaches to image pattern recognition , 1995, Future Gener. Comput. Syst..

[9]  John F. Canny,et al.  A Computational Approach to Edge Detection , 1986, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[10]  E. Beutner,et al.  Direct immunofluorescence studies of sodium chloride-separated skin in the differential diagnosis of bullous pemphigoid and epidermolysis bullosa acquisita. , 1990, Journal of the American Academy of Dermatology.

[11]  George Azzopardi,et al.  Trainable COSFIRE Filters for Keypoint Detection and Pattern Recognition , 2013, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[12]  M. Jonkman,et al.  The many faces of epidermolysis bullosa acquisita after serration pattern analysis by direct immunofluorescence microscopy , 2011, The British journal of dermatology.

[13]  J. Meijer,et al.  The n‐ vs. u‐serration is a learnable criterion to differentiate pemphigoid from epidermolysis bullosa acquisita in direct immunofluorescence serration pattern analysis , 2013, The British journal of dermatology.

[14]  George Azzopardi,et al.  A CORF computational model of a simple cell that relies on LGN input outperforms the Gabor function model , 2012, Biological Cybernetics.

[15]  D. Woodley,et al.  Identification of the skin basement-membrane autoantigen in epidermolysis bullosa acquisita. , 1984, The New England journal of medicine.