Design and Development of an Artificial Intelligence–Based Image Measurement System to Support Medical Procedures

Abstract

This project seeks to create and develop a measurement system utilizing artificial intelligence (AI) technology to facilitate applications in medical procedures, specifically for assessing wound size to improve treatment planning and decision-making. The implemented system utilizes the You Only Look Once (YOLO) v8 model for precise wound recognition and measurement, alongside a MySQL database for organized wound data storage, facilitating effective analysis and management of wound information. The development process commenced with the acquisition of wound image data and the training of the YOLOv8 model to enhance its ability to accurately detect and measure wound size. The system’s performance was evaluated in terms of accuracy and processing stability. The findings indicated an average accuracy of 84.5% and a confidence stability rate of 83% for the image processing capabilities of YOLOv8. The system also enables organized data storage and retrieval, thereby supporting decision-making in medical treatment procedures. Performance testing indicated that the system processes images efficiently and effectively, suggesting its applicability in medical practice, including monitoring treatment progress and adjusting treatment plans based on real-time data. This study emphasizes the potential of AI technology and systematic data management to improve the quality of medical care. The proposed method functions as an effective tool to enhance the accuracy and efficiency of wound measurement, presenting significant implications for the future of medical practice.