Abstract
Fatigue is one of the physiological phenomena felt by living organisms. It is a symptom that includes both physical and mental components. People feel fatigue in daily life, and the accumulation of fatigue can lead to the onset of diseases and accidents. For this reason, understanding fatigue is an important issue. Questionnaire-based monitoring is now being conducted in companies. However, due to the burden on the participants and the management, the monitoring is conducted only once or twice a year, which leads to a practicality problem. To increase the practicality, a simple and highly accurate monitoring technology is desired. Since the relationship between fatigue and skin condition has been reported in many past studies, we thought there was the possibility of constructing a fatigue prediction technique using skin information. In this study, we apply deep learning technology to predict subjective fatigue assessments and physiological fatigue assessments by blood and urine from the morphological information of epidermal stratum corneum cells obtained by tape stripping. As a result, we can predict several fatigue states from stratum corneum images. This study shows the new possibility of using the skin analysis technology developed in cosmetics research for the realization of a healthy society.
