Computer Aided Diagnosis (CAD) and Classification of Microbial and other Wide Range of Dermal Diseases using AI and Medical Image Processing

Abstract

The human body's largest and most defensive organ is its skin. It covers internal organs and shields the human body from extraneous substances outside of it. Numerous illnesses can harm a person's skin caused by microbes such as bacteria, fungi, and viruses; for example, MRSA (methicillin-resistant Staphylococcus aureus) infection, Herpes zoster, Acne vulgaris, warts, eczema, psoriasis, and the fifth disease. It can also be damaged by carcinogenic and tumor-inducing agents, leading to skin cancer such as melanoma, which is more fatal and life-threatening to human life. Skin diseases can be diagnosed by blood tests, tissue sample collection (biopsy), and skin examination by dermatologists and experts. If non-expert doctors or laboratory technicians examine the skin, it can lead to medical errors and misdiagnosis. A proper and precise diagnosis and detection are required to treat the specific disease. This research aims to detect dermal diseases through sample images and classify and identify the cause of disease with greater accuracy and precision in a time- and cost-efficient way. This research uses medical processing algorithms such as pre-processing and segmentation of the diseased image and image classification algorithms such as deep learning, part of a neural network, to classify the diseased medical images.