Algorithm for Estimation of Scratching Time

This paper presents a novel signal-processing algorithm to estimate scratching time. The scratching motion is monitored using various methods to evaluate itching behavior, which is a symptom associated with many diseases. Given that the severity of skin diseases accompanied by itching is related to the length of scratching time, various sensing systems have been developed to measure the scratching motion. The authors previously proposed a bed-sensing system, which helps monitor the scratching motion while the patient is asleep. However, in each of these systems, determining the scratching time is a burdensome task for the scorer, who determines the scratching time by visually observing the signal obtained from the sensor. In addition, the scorer has to analyze several output wave shapes of the signals recorded while the patient is asleep. In this paper, a processing algorithm capable of automatically extracting the scratching time is developed. In addition, the algorithm is applied to sensing systems.