Skinscanpro: A Deep Learning-Based Ai Health Assistant For Skin Disease Classification

Author : Manoj Kathoju

Abstract : The impact of skin diseases is felt worldwide by millions of people. It is necessary to have a tool for diagnosing this problem efficiently. Stocked with AI, this study turns public attention to skin disease classification, employing the Deep Learning algorithm DenseNet201--known for its exceptional feature extraction skills. This interactive system deploys a trained model through user input, skin images. A diagnosis with that model is a matter of just a few seconds. With a private dataset, we adopted data augmentation and then saved the trained model via Python's pickle library so that it would run on demand. The results show that up to 4/5 accuracy is achieved when classifying as expected by the feed forward operator all ellipse points in 1d spatial correlation matrix.

Keywords : Skin Disease Classification, Deep Learning, DenseNet201, Feature Extraction, AI in Healthcare, Image-Based Diagnosis.

Conference Name : International Conference on Health Care Reform, Health Economics and Health Policy (ICHCRHEHP-25)

Conference Place : Dharamsala India

Conference Date : 5th Apr 2025