Skin Disease Detection By Using Machine Learning

Skin is the largest organ in our body.It has three layers: the epidermis, dermis, and subcutaneous tissues.Skin protects us from harmful bacteria, pollution, and sunlight.It can be affected by many things, like damage from chemicals or viruses, problems with the immune system, or genetic conditions.Skin diseases can make life difficult.Sometimes, people try to treat skin problems at home, but this can be risky if the treatment isn't right for the condition.Skin diseases can spread from person to person, so it's important to treat them early.Doctors rely on their experience and judgment to diagnose patients' symptoms.If they make a mistake or take too long, it can harm people's health.That's why it's important to find better ways to detect and diagnose skin diseases early.With technology improving ,we can create systems to monitor skin and detect infections sooner.There are many tools available, like image recognition and pattern matching, to identify different skin diseases.Machine learning is one area that can help accurately identify different types of skin problems.Using machine learning, we can classify diseases based on images.Image classification is when a model is trained to recognize different classes or categories.There are many machine learning and deep learning algorithms that can identify and predict various types of skin diseases.This paper compares machine learning algorithm: CNN.We tested these algorithms on seven types of skin diseases (acne, lichen planus, and SJS-TEN, etc.) using almost 10000 skin samples.Here we are using HAM10000 dataset taken from kaggle which will be easily available.We compared the training accuracy of these algorithms and analyzed the results.