A comprehensive study on the application of machine learning in psoriasis diagnosis and treatment: taxonomy, challenges and recommendations

Abstract

Psoriasis is a common skin disease with complex mechanisms, and its diagnosis and treatment bring many challenges. In recent years, machine learning (ML) techniques have been proposed as a new tool to improve this disease’s diagnosis and treatment process. With the ability to learn from limited data and transfer knowledge from one field to another, these techniques have a high potential to improve diagnosis accuracy and treatment efficiency. However, using ML in diagnosing and treating psoriasis is associated with several challenges, including data limitations, the complexity of algorithms, and the need for high expertise to implement them properly. By presenting a detailed taxonomy, this article examines the applications and challenges of using ML techniques in psoriasis and analyzes the latest achievements in this field. The results of this study show that ML techniques have increased the accuracy of psoriasis diagnosis by 35% and improved treatment efficiency by 29%. In addition, these techniques reduced the data processing time by 21% and improved the overall treatment process. Also, these methods have increased the success rate of patient survival predictions by 15%. Finally, by examining the existing challenges and providing solutions to overcome these challenges, this research will help researchers and experts in this field develop new strategies to improve the diagnosis and treatment of psoriasis by better understanding ML applications.