YOLO-Based Deep Learning Model for Pressure Ulcer Detection and Classification

Area of Science:

• Medical Imaging
• Artificial Intelligence
• Computer Vision

Background:

• Pressure ulcers pose a significant global health challenge, especially for patients with limited mobility.
• Early detection and accurate staging are critical for effective management and prevention of complications.

Purpose of the Study:

• To develop and evaluate a novel approach for automated detection and classification of pressure ulcers using YOLOv5.
• To categorize pressure ulcers into four distinct stages and differentiate them from non-pressure ulcers.

Main Methods:

• Implementation of the YOLOv5 object detection model for pressure ulcer analysis.
• Utilization of data augmentation techniques to enhance dataset size and model robustness.

Main Results:

• Achieved an overall mean average precision (mAP) of 76.9%.
• Demonstrated high class-specific mAP50 values, ranging from 66% to 99.5% across different stages.
• Outperformed previous CNN-based methods in efficiency and accuracy.

Conclusions:

• The YOLOv5-based approach offers a promising, efficient, and accurate solution for pressure ulcer detection and classification.
• This technology has the potential to significantly improve early diagnosis and treatment, leading to better patient outcomes and reduced healthcare expenditures.