Wound Segmentation for healing prediction

Abstract

Accurate wound assessment is essential for effective medical treatment and monitoring. Traditional wound evaluation methods are subjective and time-consuming. This paper proposes a deep learning-based approach for wound segmentation and healing prediction using medical images. The U-Net architecture is employed for precise segmentation of wound regions. Extracted features are used to analyse healing progression. Experimental results demonstrate high accuracy in segmentation and reliable healing prediction, making the system useful for clinical decision support