YOLOv8-Based System for Nail Capillary Detection on a Single-Board Computer.

Area of Science:

• Medical Imaging
• Artificial Intelligence
• Dermatology

Background:

• Nail capillaroscopy identifies capillary changes crucial for diagnosing systemic sclerosis and Raynaud's disease.
• Manual capillaroscopic analysis is subjective, time-consuming, and yields inconsistent results.
• Objective and efficient methods are needed to assess capillary morphology (diameter, density, etc.).

Purpose of the Study:

• To develop and evaluate an AI-based system for automated nail bed capillary detection.
• To quantify capillary number, thickness, and density for improved diagnostic accuracy.
• To overcome limitations of manual capillaroscopic evaluation.

Main Methods:

• A YOLOv8s-based artificial intelligence system was developed.
• The system analyzes recorded microscope images of nail bed capillaries.
• It utilizes database systems and software on a Single-Board Computer (SBC).

Main Results:

• The AI system achieved a mean Average Precision (mAP) of 0.882.
• The F1 score for capillary detection was 0.83.
• The system demonstrated proficiency in detecting capillary number, thickness, and density.

Conclusions:

• The proposed AI system offers objective and efficient nail capillary analysis.
• It shows promise for improving early diagnosis and monitoring of systemic sclerosis and Raynaud's disease.
• Automated capillaroscopy can enhance diagnostic consistency and reduce labor.