Designing a low-cost mattress sensor for automated body position classification.

Determination of body position on a mattress in a clinical setting is important for both diagnosis of sleep disorders and management and prevention of pressure sores. At present, patient body position is measured by visual inspection or by post analysis of either a video monitoring system or expensive, high-resolution arrays of pressure sensors. Due to the high cost and time consumption, often the alternative, a set of accelerometers tethered to the patient is used. However, these accelerometers only provide directional orientation of the patient torso, and yield little information regarding the total body position. We present in this paper a design for a low-cost sensor mat that combines a minimum number of sensors with a feed-forward neural net for autonomous detection and distinction of body position into a set of discrete patterns.