In The American journal of pathology ; h5-index 54.0
The u-serrated immunodeposition pattern in direct immunofluorescence (DIF) microscopy is a recognizable feature and confirmative for diagnosis of epidermolysis bullosa acquisita (EBA). Due to unfamiliarity with serrated patterns, serration pattern recognition is still of limited use in routine DIF microscopy. The objective of this study is to investigate the feasibility of using convolutional neural networks (CNNs) for the recognition of u-serrated patterns that can assist in diagnosis of EBA. Nine most commonly used CNNs are trained and validated by using 220,800 manually delineated DIF image patches from 106 images of 46 different patients. The dataset was split into 10 subsets; 9 training subsets from 42 patients to train CNNs and the remaining subset from the remaining 4 patients for validation dataset of diagnostic accuracy. This process was repeated 10 times with a different subset used for validation. The best performing CNN achieved specificity of 89.3% and corresponding sensitivity of 89.3% in the classification of u-serrated DIF image patches, a diagnostic accuracy of expert level. Experiments and results demonstrate the effectiveness of convolutional neural networks approaches for u-serrated patterns recognition with a high accuracy. The proposed approach can assist clinicians and pathologists in recognition of u-serrated patterns in DIF images, and facilitate diagnosis of EBA.
Shi Chenyu, Meijer Joost M, Azzopardi George, Diercks Gilles F H, Guo Jiapan, Petkov Nicolai
Convolutional neural network, direct immunofluorescence, epidermolysis bullosa acquisita, machine learning, pemphigoid, serration pattern analysis