Automated classification of hidradenitis suppurativa disease severity by convolutional neural network analyses using calibrated clinical images.

BACKGROUND: The assessment of hidradenitis suppurativa (HS) severity requires detailed, and error-prone lesion counts. This proof-of-concept study aimed to automatically classify HS disease severity using machine learning of clinical smartphone images. METHODS: 777 ambient-light and size-controlled images were used to build a class-balanced synthetic dataset (n = 7675). Convolutional neural networks (CNN) were used for automated severity classification (scale 0-3), and to assess disease-dynamics. International Hidradenitis Suppurativa Severity Score System (IHS4) served as reference. A U-NET algorithm was implemented for automated localization of diseased skin. RESULTS: CNNs were able to distinguish no/mild from moderate/severe disease with an overall prediction accuracy of 78% [receiver operating curve (AUC) 0.85]. Correct IHS4 classification was achieved with an overall accuracy of 72% (AUC 0.84-0.89). In addition, disease dynamics using IHS4 numerical values aligned with CNN outputs (NRMSE 0.262). The UNET algorithm localized lesions with a pixel accuracy of 88.1% and test loss of 0.42. LIMITATIONS: Limitations in assessing tattooed and hairy skin. Limited number of patients with dark skin colour and Hurley I. CONCLUSION: CNNs were able to distinguish no/mild from moderate/severe disease, classify disease severity over time, and automatically identify diseased skin areas and the skin phototype. This study breaks new grounds for fast, reliable, reproducible and easy-to-use HS severity assessments using clinical images.

Anti-BP180 autoantibody levels at diagnosis correlate with 1-year mortality rates in patients with bullous pemphigoid.

BACKGROUND: Bullous pemphigoid (BP) is the most frequent autoimmune blistering disease mainly affecting elderly patients. Among several published risk factors, a recent post hoc analysis linked anti-BP180 autoantibodies (AABs) to fatal outcomes in BP. To date, this finding has not been confirmed independently. OBJECTIVE: To investigate the potential of anti-BP180-AAB levels as a marker of prognosis and to identify a cut-off level indicative of an increased risk for early death. Secondly, to characterize parameters associated with mortality. METHODS: Retrospective, single-centre study of BP patients diagnosed between 2001 and 2012. Analyses included epidemiological and patient- and disease-specific characteristics as well as immunological parameters at diagnosis and during follow-up. Standardized mortality ratios as well as uni- and multivariate regression analyses were calculated. RESULTS: One hundred patients (56 women, 44 men) with a median age of 81 years (interquartile range 74-86) were followed up for a median of 775 days (interquartile range 162-1617). One-year mortality rates were 25.0% implying a 2.4-fold increased risk of death compared with the general population. High anti-BP180 autoantibody levels at diagnosis (CI95 1.30-2.89; P = 0.001), dementia (CI95 1.13-6.72; P =0.03), length of hospitalization (CI95 1.16-2.41; P = 0.01) and age (CI95 1.23-4.19; P = 0.009) correlated significantly with 1-year mortality. BP180-AAB concentrations of ≥61 U/mL characterized a subgroup of patients with a particular higher risk for early death compared with the general population (CI95 1.81-3.81; P < 0.0001). CONCLUSION: In bullous pemphigoid, serum concentrations of BP180 autoantibodies at diagnosis could help to identify patients at risk for death within the first year after diagnosis (cut-off value 61 U/mL).