Intelligent progress monitoring of healing wound tissues based on classification models

The evolution of wound monitoring techniques has seen a significant shift from traditional methods like ruler-based measurements to the use of AI-assisted assessment of wound tissues. This progression has been driven by the need for more accurate, efficient, and non-invasive methods for wound assessment and treatment planning. The proposed approach aims to automate wound analysis and reduce efforts to manage chronic wounds. The snake's approach is used to extract wound areas and geometrical measures are used to monitor the rate of wound healing. A segmentation based on the color thresholding and K-means technique was carried out and demonstrated the effectiveness of the thresholding technique in mapping the wound tissues. The three proportions of wound tissues necrosis, slough, granulation and wound size are combined with three features from the patient's medical record and transmitted to the Support Vector Machine (SVM), Naive Bayes (NB) and Decision Tree (DT) classifiers. Finally, this work is ended with a comparative study that shows the efficiency and the interest of the proposed approach.