Artificial Intelligence in Dermatology: A Systematic Review

Introduction: Due to the large volume of diverse data regularly received, automation of routine processes in dermatology is a highly relevant task. Artificial intelligence (AI) may provide effective solutions for automating various processes in dermatology. Objective We aimed to assess the current state of development and implementation of AI in dermatology and identify key challenges hindering AI integration into clinical practice. Materials and Methods: A literature search was conducted in PubMed and the Russian Science Citation Index (RSCI) databases, as well as in the Register of Medical Devices to identify registered medical devices incorporating AI. The time frame covered 2019 to 2025. Bibliometric data, research avenue, types of studied pathologies, methodological characteristics, AI and human diagnostic performance, number and experience of involved medical personnel, and implementation outcomes were extracted. Study quality was assessed using the QUADAS-CAD. Results: A total of 41 out of 270 identified references were included in the systematic review. Most studies focused on diagnosing malignant skin neoplasms (65.85%), primarily melanoma (51.22%). In the analyzed studies, AI demonstrated high diagnostic performance comparable to those of experienced specialists. Median value (n=27) for accuracy of AI in diagnosing malignant skin neoplasms were 80% (95%CI: 76.55–83.45%). Among the included algorithms, eight have received medical device certification, and four are mobile applications for diagnosing skin conditions. Discussion: AI implementation in dermatology is at an advanced stage, with 19.5% of analyzed studies reaching commercial deployment and product dissemination levels. However, further research is needed to improve methodological quality in evaluating AI’s accuracy.