Detection and classification methodology for movements in the bed that supports continuous pressure injury risk assessment and repositioning compliance. Issue 1 (February 2019)

• Record Type: Journal Article
• Authors: Duvall, Jonathan; Karg, Patricia; Brienza, David; Pearlman, Jon
• Is Part Of: Journal of tissue viability. Volume 28:Issue 1(2019)
• Page Start: 7
• Page End: 13
• Publication Date: 2019-02
• Subjects: Wounds and injuries -- Periodicals; Ulcers -- Periodicals; Bedsores -- Periodicals; Bedsores; Ulcers; Wounds and injuries; Electronic journals; Periodicals; 617.1406
• DOI: 10.1016/j.jtv.2018.12.001
• Languages: English
• ISSNs: 0965-206X

Abstract

Abstract: Pressure injuries are costly to the healthcare system and mostly preventable, yet incidence rates remain high. Recommendations for improved care and prevention of pressure injuries from the Joint Commission revolve around continuous monitoring of prevention protocols and prompts for the care team. The E-scale is a bed weight monitoring system with load cells placed under the legs of a bed. This study investigated the feasibility of the E-scale system for detecting and classifying movements in bed which are relevant for pressure injury risk assessment using a threshold-based detection algorithm and a K-nearest neighbor classification approach. The E-scale was able to detect and classify four types of movements (rolls, turns in place, extremity movements and assisted turns) with >94% accuracy. This analysis showed that the E-scale could be used to monitor movements in bed, which could be used to prompt the care team when interventions are needed and support research investigating the effectiveness of care plans. Highlights: The E-scale (weight scale under the legs of a bed) can monitor and detect postural movements in bed with 97% accuracy. The movements can be classified into 4 categories (Rolls, Turns, Extremity Movements and Assisted Turns) with 96% accuracy. The largest misclassification is between Assisted Turns and Turns.