Deep Facial Diagnosis: Deep Transfer Learning From Face Recognition to Facial Diagnosis

The relationship between face and disease has been discussed from thousands years ago, which leads to the occurrence of facial diagnosis. In existing system, Old methods like screening by a physician in laboratories with equipment’s which take time to generate results and or also cost in effective. Such system also required for us to know what we are looking for from the start. Also many alternatives to this old system uses machine learning to detect patient diseases. But they suffer from the low accuracy. In proposed system, we are using Deep learning and neural network to capture the faces of people and detect any possible disease associated to them. Deep learning offers increased accuracy for detection of disease and it is highly scalable. We used data augmentation technique to handle imbalance of data in the system. It is also allows us to reduce over fitting and hence generate better accuracies than ever before.

Keywords: Facial Diagnosis, Deep Transfer Learning (DTL), Face Recognition, Beta-Thalassemia, Hyperthyroidism, Down Syndrome, Leprosy.