Scar Tune: An Effective Approach for Scar Care

In the field of dermatology, cosmetic surgery, and wound healing skin scar detection plays very important role in providing accurate treatments by precise scar detection. Accurate and efficient scar detection methods are essential for assessing treatment outcomes. The aim of this paper is to develop a Scar Detection and classification model with the help of Machine Learning Algorithms on a custom dataset. To create intelligent models, we need good machine learning models which need to be trained on scars dataset. In this paper, custom dataset is described briefly which have data and images of different types of scars like Hypertrophic, Keloid, Atrophic, and Contracture. After this, there is a comparative study between three different deep learning models: MobileNetV2, InceptionV3, EfficientNetB0 which are used to classify different types of scars, followed by the results obtained from each model.