Development of a multifunctional nursing bed system for in-bed position recognition and automatic repositioning
Haoyu Zhou,Lingfeng Sang,Jingjing Luo,Hongbo Wang,Yongfei Feng,Xueze Zhang,Li Chen
DOI: https://doi.org/10.1177/09544062231223878
2024-01-20
Proceedings of the Institution of Mechanical Engineers Part C Journal of Mechanical Engineering Science
Abstract:Proceedings of the Institution of Mechanical Engineers, Part C: Journal of Mechanical Engineering Science, Ahead of Print. Bedridden elderly need to be repositioned at regular intervals to prevent pressure ulcers. However, repositioning, being one of the most physically demanding tasks in nursing, faces many challenges, including work-related injuries and limited manpower. To address these problems, we have developed a multifunctional nursing bed system. This system not only assists in repositioning and transitioning positions but also features in-bed position recognition, enabling automatic repositioning based on recognition results. In this paper, the repositioning process was analysed, and the nursing bed mechanism and its control design were developed. An algorithmic framework for in-bed position recognition, including preprocessing and convolutional neural networks, is proposed. The experimental results demonstrate that the nursing bed system facilitates comfortable repositioning and position transition for subjects. In a short-term experiment, the system demonstrated an accuracy rate of 96.83% for in-bed position recognition. Meanwhile, in a long-term experiment, it achieved a success rate of 93.89% for automatic repositioning. These findings indicate that the proposed system demonstrates substantial potential in nursing care, particularly in pressure ulcer prevention.
engineering, mechanical