A labeled dataset of tongue images

This is a tongue image dataset collected from the inpatient elderly population at a tertiary hospital in Shanghai, China. The data acquisition device, we called the intelligent mirror, is a light-field camera using CIE L*a*b* (CIELAB) with a patent number as CN201520303463.5. The main parameters produced by the intelligent mirror initially are the color of the tongue, tongue coating, lip, and face. We manually labeled the dataset with several other parameters based on traditional Chinese medicine, including tongue shape, size, the degree of smooth or shiny, and the judgment of tooth marks or fissured tongue. This original and continuously growing dataset is suitable as secondary materials to understand the clinical interoperability between Chinese and Western medicine. Another possible future application lies in the commercial field, which is to be the training set as adjuvant therapy deep learning via mHealth tools.