Title :
Unrestrained multi-sensor systems for real-time prediction of bed-leaving behavior patterns
Author :
Madokoro, Hirokazu ; Shimoi, Nobuhiro ; Sato, Kiminori
Author_Institution :
Dept. of Machine Intell. & Syst. Eng., Akita Prefectural Univ., Yurihonjo, Japan
Abstract :
This paper presents an unrestrained sensor system that analyzes predictive behavior patterns that occur when a patient leaves a bed. Our system comprises three sensors: pad sensors installed under a bed mat, a pillow sensor to detect head movements, and a bolt sensor mounted to a bed handrail. We used a triaxial accelerometer for a pillow sensor and piezoelectric elements for the pad sensors and bolt sensor. The salient features of these sensors are their easy installation, low cost, high reliability, and toughness. Moreover, we developed a method of recognizing bed-leaving behavior patterns using machine-learning algorithms from signals obtained using the sensors. We evaluated our system by examining ten subjects in an environment representing a clinical site. The mean recognition accuracy for seven behavior patterns was 75.5%. Furthermore, the recognition accuracies for longitudinal sitting, terminal sitting, and left the bed were 83.3%, 98.3%, and 95.0%, respectively. In contrast, falsely recognized patterns remained inside of respective categories of sleeping and sitting. We believe that our system is applicable to an actual environment as a novel sensor system with no restraint of patients.
Keywords :
accelerometers; behavioural sciences computing; learning (artificial intelligence); sensor fusion; signal denoising; bed handrail; bed mat; bed-leaving behavior pattern recognition; bolt sensor; clinical site; head movement detection; left-the-bed pattern; longitudinal sitting pattern; machine-learning algorithms; mean recognition accuracy; pad sensor installation; piezoelectric elements; pillow sensor; predictive behavior pattern analysis; real-time bed-leaving behavior pattern prediction; sleeping pattern; terminal sitting pattern; triaxial accelerometer; unrestrained multisensor systems; Biomedical imaging; Fasteners; Monitoring; Noise; Pattern recognition; Prototypes; Sensor systems; Accelerometer; Bed-Leaving; Machine Leaning; Piezoelectric elements; Quality of Life;
Conference_Titel :
SICE Annual Conference (SICE), 2014 Proceedings of the
Conference_Location :
Sapporo
DOI :
10.1109/SICE.2014.6935309