People with atopic dermatitis (AD) would benefit from personalised treatment strategies. We aim to develop a computational tool that makes personalised predictions of AD severity dynamics and generates treatment recommendations. We introduced EczemaPred, a computational framework to predict patient-dependent dynamic evolution of AD severity using Bayesian state-space models. We used EczemaPred to predict the dynamic evolution of Patient-Oriented Scoring Atopic Dermatitis (PO-SCORAD) by combining predictions for the nine severity items of PO-SCORAD (six intensity signs, extent of eczema, and two subjective symptoms). We validated this approach using longitudinal data of 347 AD patients with twice-weekly measurements over 17 weeks. We further extended EczemaPred to integrate treatment data available on another dataset of 16 AD patients with daily recording of PO-SCORAD for 12 weeks. We estimated treatment effects and used Bayesian decision analysis to generate treatment recommendations. EczemaPred achieved good performance for predicting PO-SCORAD and its nine severity items daily to weekly. Estimated treatment responses strongly depended on the patient’s clinical phenotype and allowed us to generate patient-specific treatment recommendations. We demonstrated the use of EczemaPred as a coherent framework for personalised AD severity predictions and treatment recommendations, while dealing with missing values and measurement errors. EczemaPred could be applied to other AD severity scores such as EASI or POEM.

ESDR087 - Personalised predictions of AD severity dynamics and treatment recommendations using a Bayesian machine learning approach

The scientific meeting of the European Society for Dermatological Research (ESDR) is held yearly, usually in September. With an international attendance of more than 1000 skin scientists from around the globe, the four-day event includes short oral presentations from peer-reviewed abstracts, keynote lectures from international experts, sponsored symposia, satellite programs, educational events as well as an important focus on posters, networking and building collaborations amongst scientists. For more information about this year’s meeting visit our website.