| Shavkat Fazilov, Shukhrat Mamadjanov System Analysis of Acne Diagnostics Using Artificial Intelligence |
|---|
| Abstract. Acne, considered one of the dermatological disorders, is a widespread skin disease that often occurs during crucial developmental stages in adolescents and can lead to long-term psychosocial consequences. Traditional methods, such as assessing overall severity and counting rash elements, are limited by subjectivity and time constraints. This article focuses on a systematic review of the latest advancements in diagnosing acne, detecting and counting rash elements, and assessing severity using artificial intelligence (AI), highlight-ing the potential of AI-based methods to improve objectivity, reproducibility, and clinical effectiveness. A comprehensive literature search was conducted in PubMed, Scopus, arXiv, Embase, and Web of Science databases for stud-ies published from 2017 to February 2026. The search strategy included terms related to "acne" and various AI methodologies (e.g., "neural net-work," "deep learning," "convolutional neural network"). During the research process, 391 articles were reviewed, of which 39 articles met the final crite-ria. Data were extracted on study design, dataset characteristics (including in-ternal and open-access resources such as ACNE04 and AcneSCU), AI archi-tectures (primarily CNN-based models), and performance indicators. Alt-hough AI-based models have demonstrated high accuracy under controlled conditions, the scarcity of large public datasets, the predominance of data from specific ethnic groups, and the lack of comprehensive external valida-tion reveal significant barriers to implementation in clinical practice. The re-search findings indicate that while AI has the potential to standardize acne assessment, reduce observer variability, and enable self-monitoring through mobile platforms, there are considerable challenges in achieving reliable re-al-world application. Future research should prioritize the creation of large, diverse, and openly accessible datasets and conduct prospective clinical trials to ensure fair and effective dermatological care. |
| Keywords: Acne, artificial intelligence, acne position, dermatology, image classification, computer vision |
 Download PDF |
| DOI: https://doi.org/10.54381/itta2026.2.03 |