Automated Detection of Microorganisms of TINEA using Deep Learning

Medical Sciences have seen a tremendous advancement in terms of diagnosis, medication, as well as discovery of various new families of disorders and diseases. Diagnosis and treatment of such complicated diseases requires keen observation and highly experienced physicians and doctors. The field of Artificial Intelligence and Machine Learning can very constructively provide a reliable support to doctors for many complicated diseases. There are various scenarios, such as diagnosis of different types of Cancers, where Artificial Intelligence systems are already being used for faster and more accurate detection. Medical Science has various branches, one such being the field of Dermatology. It is a field that primarily deals with the skin, nails, hair, and their accompanying diseases, which being the most exposed parts of the human body, encounter various kinds of fungi, bacteria, and viruses. A skin disease called ‘Tinea’ is one such disease, that when approached with the traditional method of observing the specimen under a microscope, requires long durations of diagnosis by even experienced doctors. Our work aims at designing an efficient system, that would assist doctors by predicting the presence of the diseases in the specimen in short periods of time and reducing human intervention for every case of Tinea encountered. There are existing models using deep learning, that can be used to implement the approach we are looking at, which we aim to build on using Transfer Learning. We aim to bring a solution to facilitate deployable options for such models and muster high accuracies for them.