Machine Learning Opportunities for Automatic Tongue Diagnosis Systems

Machine Learning approaches and the availability of digital data are favoring the rising of applications in biomedical field. Smartphones and wearable devices are facilitating the acquisition of images and audio files of the population and this is fueling patient-centric approaches as regards to the diagnosis of specific diseases. This trend is desired by the World Health Organization both for a view to monitoring and empowering the population and for encouraging the containment of medical costs. Tongue inspection is a clinical practice of Traditional Chinese Medicine (TCM) that has been existing for thousands of years, and allows to intercept diseases and severity of internal organs. While in the past, the traditional diagnosis of the tongue depended on the observation of the doctor, now the use of automatic computerized diagnosis systems is conceivable. In this paper, we focus on the topics of interest from which to start to prototype mobile solutions for tongue diagnosis based on artificial intelligence.