Deep Learning-Aided Diagnosis of Autoimmune Blistering Diseases

Autoimmune blistering diseases (AIBDs) are rare, chronic disorders of the skin and mucous membranes, with a broad spectrum of clinical manifestations and morphological lesions. Considering that 1) diagnosis of AIBDs is a challenging task, owing to their rarity and heterogeneous clinical features, and 2) misdiagnoses are common, and the resulting diagnostic delay is a major factor in their high mortality rate, patient prognosis stands to benefit greatly from the development of a computer-aided diagnostic (CAD) tool for AIBDs. Artificial intelligence (AI) research into rare skin diseases like AIBDs is severely underrepresented, due to a variety of factors, foremost a lack of large-scale, uniformly curated imaging data. A study by Julia S. et al. finds that, as of 2020, there exists no machine learning studies on rare skin diseases [1], despite the demonstrated success of AI in the field of dermatology. Whereas previous research has primarily looked to improve performance through extensive data collection and preprocessing, this approach remains tedious and impractical for rarer, under-documented skin diseases. This study proposes a novel approach in the development of a deep learning based diagnostic aid for AIBDs. Leveraging the visual similarities between our imaging data with pre-existing repositories, we demonstrate automated classification of AIBDs using techniques such as transfer learning and data augmentation over a convolutional neural network (CNN). A three-loop process for training is used, combining feature extraction and fine-tuning to improve performance on our classification task. Our final model retains an accuracy nearly on par with dermatologists' diagnostic accuracy on more common skin cancers. Given the efficacy of our predictive model despite low amounts of training data, this approach holds the potential to benefit clinical diagnoses of AIBDs. Furthermore, our approach can be extrapolated to the diagnosis of other clinically similar rare diseases.

Anti-Desmocollin Autoantibodies in Autoimmune Blistering Diseases

The presence of anti-desmocollin (Dsc) antibodies is rarely described in autoimmune blistering diseases patients. Moreover, several clinical phenotypes of pemphigus may be associated with these antibodies. In this review we analyze clinicopathological, immunologic and outcome features of anti-Dsc autoimmune blistering diseases patients, to improve their diagnosis and management. We conducted a systematic search of PubMed and Embase (1990-present) for studies reporting cases of autoimmune blistering diseases with anti-Dsc antibodies. We classified the selected patients as patients with exclusively anti-Dsc autoantibodies, and patients with anti-Dsc and other autoantibodies. Of 93 cases with anti-Dsc autoantibodies included, 38 (41%) had exclusively these antibodies. Only 18% of patients presented with the typical clinicopathological phenotype of pemphigus vulgaris or pemphigus foliaceous. Mucosal involvement was seen in approximately half of the patients. Up to 18% of cases were associated with neoplasms. Acantholysis was described in 54% of cases with histopathological information. Treatments and outcomes vary in the different clinical phenotypes. The presence of anti-Dsc antibodies must be suspected mainly in those patients with either atypical pemphigus, in special with clinical pustules, or in cases showing intraepithelial or dermal neutrophilic/eosinophilic infiltrate on histological examination and dual pattern by direct immunofluorescence examination.

Rare forms of autoimmune blistering diseases

The authors summarize up-to-date knowledge about the rare forms of autoimmune blistering diseases, focusing on their clinical features, diferential diagnosis and the modern therapy modalities. Besides the well known pemphigus and pemphigoid diseases, the knowledge about the rare autoimmune blistering diseases has been expanded. Although at this time, there is no signifcant diference between these diseases with regard to their therapy, the diferential diagnosis is important for knowing their exact prevalences and it may lead to individual targeted therapies in the future.

Exploring RNAs Interactions and Polymorphisms in the Pathophysiology of Pemphigus: A Review

Background: Pemphigus consists of a group of rare autoimmune blistering diseases involving the skin and mucous membranes. Pemphigus pathophysiology is mediated by autoantibodies against two desmosomal cadherins, namely, desmoglein (Dsg) 1 and (Dsg) 3 that are present in the skin and mucosal membranes. The involvement of coding and non-coding RNAs in the pathophysiology of pemphigus has been studied in the literature. MicroRNAs are small RNAs that could also be used as diagnostic biomarkers for some autoimmune diseases. The aim of this research was to explore the potential of this specification of some RNAs to be used as biomarkers for diagnosing pemphigus or its severity. This review discussed RNA expressions in patients with pemphigus. Methods: A comprehensive search was performed on published studies from 1990 to May 2020 using different search engines including PubMed, Scopus, and Web of Science. Results: In general, 335 articles were obtained according to search keywords. Then, 41 relevant studies were selected based on the inclusion and exclusion criteria. MiR-338-3p, miR-424-5p, and miR-584-5p were among the miRNAs that were reported to be increased in pemphigus. The C3 mRNA, mRNA of CD36, mRNA of CD163, mRNA of urokinase plasminogen activator (PA), IL23R mRNA, RORγt mRNA, and human leukocyte antigen G1 (HLA-G1) mRNA were coding RNAs that increased in pemphigus in addition to the activity of the mRNA of tissue-type PA while HLA-G2 mRNA decreased in pemphigus. Conclusion: Overall, this study investigated the role of Mir-338-3p, miR-424-5p, MiR-127, miR-584-5p, and some mRNAs in pemphigus, and it was revealed that some RNAs may be impressive on pemphigus. More studies and clinical assessments need more information about the role of RNAs on pemphigus to obtain a better view of their mechanisms and use them as biomarkers for earlier diagnosis or probable treatment.

Bullous Systemic Lupus Erythematosus Mimicking Bullous Pemphigoid: A Case Report

Introduction: Bullous systemic lupus erythematosus (BSLE) is an infrequent but distinct presentation of systemic lupus erythematosus (SLE) in less than 5% of lupus cases. It is characterized by vesicobullous skin eruption in SLE that can develop either before or after SLE diagnosis has been established. Distinguish between BSLE with other autoimmune blistering diseases such as bullous pemphigoid (BP), dermatitis herpetiformis, linear IgA, etc., is very important to prevent misdiagnosis. The physician must be able to combine clinical, histological and immunofluorescence finding for the diagnosis approach. We report a case of blistering skin eruption in SLE patient. Case Report: A 19-year-old female patient complained of tense blistering on her lip, face and wrists since one month ago. She was diagnosed with SLE two weeks ago. The dermatological state showed bullae and vesicle on erythematous/ normal base, erosions, excoriation and blackish red crust on the lip, face, armpit, neck, abdomen and wrists. Histopathological examination of the lesion showed sub-epidermal bullae containing PMN leukocytes consist of abundant neutrophils, only occasional eosinophils and the presence of keratotic plugs. Direct immunofluorescence (DIF) of the skin showed linear deposition of IgG, IgA, IgM and C1q at the dermo-epidermal junction. This patient exhibited similar features to both BSLE and BP with tensed clear blisters and subepidermal cleft. BSLE differ from BP by abundant neutrophils found on histopathological examination, whereas BP has abundant eosinophils. Conclusion: Immunofluorescence examination shows linear IgG in BP, whereas linear or granular IgG in BSLE. Establishing the correct diagnosis is important to prevent misdiagnosis and mistreatment.