Human Activity Recognition Using Smartphone Sensor Based on Selective Classifiers

Khatun, Mst. Alema; Yousuf, Mohammad Abu

doi:10.1109/sti50764.2020.9350486

Cited by 6 publications

(1 citation statement)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Deep learning has been applied to the processing of signals within the fields of medicine and health monitoring, such as liver segmentation, 33,34 vertebrae segmentation, [35][36][37] cancer detection, 38 facial expression recognition 39 and human activity recognition. [40][41][42] Deep learning methods achieve higher accuracy than analytical methods. In this work, we used a Convolutional Neural Network (CNN) for in-bed position recognition (Figure 15).…”

Section: Classificationmentioning

confidence: 99%

Development of a multifunctional nursing bed system for in-bed position recognition and automatic repositioning

Zhou,

Sang,

Luo

et al. 2024

Proceedings of the Institution of Mechanical Engineers, Part C:

View full text Add to dashboard Cite

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.

show abstract