Artificial Intelligence-Assisted Dermatophyte Diagnosis-Image Analytics Approach

8TH INTERNATIONAL EURASIAN CONFERENCE ON BIOLOGICAL AND CHEMICAL SCIENCES, Ankara, Türkiye, 17 - 19 Aralık 2025, ss.296, (Özet Bildiri)

Artificial intelligence–based diagnostic tools have recently gained increasing attention as complementary

technologies for dermatological imaging. Dermatophyte infections, which represent one of the most common

groups of superficial fungal diseases, are often difficult to diagnose accurately using traditional visual inspection

or microscopy alone. Deep learning models, particularly convolutional neural networks (CNNs), can automatically

extract morphological features associated with dermatophytic lesions from dermoscopic and clinical skin images.

These models enhance diagnostic precision by learning fine-scale texture differences, erythema patterns, edge

characteristics and scaling intensity that may not be easily distinguished by clinicians. Recent studies demonstrate

that AI-supported systems can reach accuracy levels comparable to experienced dermatologists, especially in

differentiating dermatophyte infections from eczema, psoriasis and bacterial dermatoses. This study provides an

overview of state-of-the-art image-based AI applications developed for dermatophyte detection, discussing

training dataset requirements, preprocessing workflows, model performance metrics and common challenges

related to class imbalance, image noise and variability in illumination. Furthermore, the integration of AI tools

with tele-dermatology platforms is evaluated, highlighting their potential use in remote triage, early detection and

decision support. Limitations such as the need for standardized image acquisition protocols, dataset diversity and

clinical validation processes are also addressed. Overall, AI-driven image analysis offers a promising, fast and

reproducible approach to dermatophyte diagnosis. When supported by high-quality datasets and rigorous clinical

evaluation, these systems may enhance routine dermatological practice and contribute to more accurate

management strategies for superficial mycoses.

Keywords: Dermatophyte, Artificial Intelligence, CNN, Image Analysis, Digital Diagnosis