Login / Signup

Proposal for a new noncontact method for measuring tongue moisture to assist in tongue diagnosis and development of the tongue image analyzing system, which can separately record the gloss components of the tongue.

Toshiya NakaguchiKanako TakedaYuya IshikawaTakeshi OjiSatoshi YamamotoNorimichi TsumuraKeigo UedaKoichi NagamineTakao NamikiYoichi Miyake
Published in: BioMed research international (2015)
Tongue diagnosis is a noninvasive diagnosis and is traditionally one of the most important tools for physicians who practice Kampo (traditional Japanese) medicine. However, it is a subjective process, and its results can depend on the experience of the physician performing it. Previous studies have reported how to measure and evaluate the shape and color of the tongue objectively. Therefore, this study focused on the glossy component in order to quantify tongue moisture in tongue diagnosis. We hypothesized that moisture appears as a gloss in captured images and measured the amount of water on the tongue surface in 13 subjects. The results showed a high correlation between the degree of gloss and the amount of water on the tongue surface and suggested that the moisture on the tongue can be estimated by the degree of gloss in a captured image. Because the moisture level on the tongue changes during the course of taking photos, it became clear that we had to wait at least 3 minutes between photos. Based on these results, we established the tongue image analyzing system (TIAS), which can consistently record the gloss and color of the tongue surface simultaneously.
Keyphrases
  • primary care
  • deep learning
  • healthcare
  • machine learning
  • optical coherence tomography
  • sleep quality
  • case control