Atopic dermatitis (AD) is a severe and common chronic skin disease that impacts the quality of life of patients, and has a worldwide prevalence of 7-30% in children and 1-10% in adults. Therefore, reliable and accurate evaluation methods for early diagnosis and treatment of AD are very important. In recent years, the use of machine learning (ML) in medical fields has grown significantly, thanks to its ability to identify and classify skin lesions and histopathologic images, as well as to improve disease predictions and personalize treatment.
Recently, a team from Third Xiangya Hospital at Central South University in China, led by Ph.D. candidate Songjiang Wu, has developed several relatively stable and reliable prediction models for AD diagnosis and efficacy evaluation employing three ML algorithms: lasso, linear regression, and random forest. The models showed outstanding performance in distinguishing between AD lesions and non-lesions (with an AUC over 0.8).
The models also revealed a positive correlation of the scores with SCORAD (Scoring Atopic Dermatitis) in patients treated with biological therapy, as well as a negative correlation with treatment duration, indicating a positive trend in effectiveness. Unfortunately, due to the small sample size and poor sample quality, the correlation coefficients between the models and the SCORAD were not high enough.
These findings, recently published in the journal Fundamental Research, suggest the potential of ML-based models in clinical AD diagnosis and treatment efficacy. The research team plans to collect more samples in the future for better validation and model stability.
The Third Xiangya Hospital at Central South University is affiliated to the Chinese Academy of Sciences, specializing in comprehensive medical and research services.
Songjiang Wu is a Ph.D. candidate in dermatology at Third Xiangya Hospital at Central South University in China. He focuses his research on the development of machine learning algorithms to assess diagnostic and therapeutic evaluation of atopic dermatitis.
