IMAGE CLASSIFICATION OF WOUNDS USING A HYBRID APPROACH BETWEEN ACTIVE CONTOUR AND DECISION TREE

Sari

The segmentation and classification of wound images play a crucial role in medical analysis, particularly in the diagnosis and treatment of wounds. This study develops a hybrid method between active contour segmentation techniques with decision tree classification to enhance accuracy in recognizing wound types in digital images. The active contour technique is employed to precisely separate the wound object from the background, while the decision tree algorithm is used to classify wound types, such as burns, cuts, and punctures, based on extracted features. Using a dataset of 278 wound images, experimental results show an improvement in accuracy in identifying wound types, offering an efficient solution that contributes to the advancement of more sophisticated wound detection systems in the medical field.