Healing with Pixels: Superior Wound Type Identification Using Enhanced Deep Learning Models

In this study, a deep convolutional neural network with fifty layers called ResNet-50, which is effective in picture classification, is utilized to categorize six different types of wounds. These wounds include a stab wound, an abrasion wound, a bruise wound, a burn wound, a cut wound, and a laceration wound. The total accuracy that was attained for this model is 87%. Track metrics for class stabbed: precision 1.00, recall 1.00, and F1-score 1.00; abrasion wound with precision 0.80, recall 0.80, and F1-score 0.80; bruises wound with precision 1.00, recall 0.87, and F1-score 0.93; burn wound with precision 0.97 and recall 0.92, and F1-score 0.94; cut the wound with precision 0.55, recall 0.75, and F1-score 0.63; and laceration wound with precision 0.73, recall 0.80, and F1-score 0.76. Based on these findings, this model has the potential to be of use to practitioners by giving the appropriate classification of wounds.