ABSTRACT
In the smart era, big data analysis based on sensor units is important in intelligent motion. In this study, a dance sports and injury monitoring system (DIMS) based on a recyclable flexible triboelectric nanogenerator (RF-TENG) sensor module, a data processing hardware module, and an upper computer intelligent analysis module are developed to promote intelligent motion. The resultant RF-TENG exhibits an ultra-fast response time of 17 ms, coupled with robust stability demonstrated over 4200 operational cycles, with 6% variation in output voltage. The DIMS enables immersive training by providing visual feedback on sports status and interacting with virtual games. Combined with machine learning (K-nearest neighbor), good classification results are achieved for ground-jumping techniques. In addition, it shows some potential in sports injury prediction (i.e., ankle sprains, knee hyperextension). Overall, the sensing system designed in this study has broad prospects for future applications in intelligent motion and healthcare.
ABSTRACT
The rapid growth of the Internet of Things and wearable sensors has led to advancements in monitoring technology in the field of health. One such advancement is the development of wearable respiratory sensors, which offer a new approach to real-time respiratory monitoring compared to traditional methods. However, the energy consumption of these sensors raises concerns about environmental pollution. To address the issue, this study proposes the use of a triboelectric nanogenerator (TENG) as a sustainable energy source. The electrical conductivity of the TENG is improved by incorporating chitosan and carbon nanotubes, with the added benefit of chitosan's biodegradability reducing negative environmental impact. A wireless intelligent respiratory monitoring system (WIRMS) is then introduced, which utilizes a degradable triboelectric nanogenerator for real-time respiratory monitoring, diagnosis, and prevention of obstructive respiratory diseases. WIRMS offers stable and highly accurate respiratory information monitoring, while enabling real-time and nondestructive transmission of information. In addition, machine learning technology is used for sleep respiration state analysis. The potential applications of WIRMS extend to wearables, medical monitoring and sports monitoring, thereby presenting innovative ideas for modern medical and sports monitoring.
Subject(s)
Nanotubes, Carbon , Wearable Electronic Devices , Wireless Technology , Wireless Technology/instrumentation , Humans , Nanotubes, Carbon/chemistry , Monitoring, Physiologic/instrumentation , Monitoring, Physiologic/methods , Sleep/physiology , Sports , Chitosan/chemistry , Nanotechnology/instrumentation , Nanotechnology/methods , Respiration , Electric Conductivity , Electric Power SuppliesABSTRACT
Surface electromyogram (sEMG) signals have been used in human motion intention recognition, which has significant application prospects in the fields of rehabilitation medicine and cognitive science. However, some valuable dynamic information on upper-limb motions is lost in the process of feature extraction for sEMG signals, and there exists the fact that only a small variety of rehabilitation movements can be distinguished, and the classification accuracy is easily affected. To solve these dilemmas, first, a multiscale time-frequency information fusion representation method (MTFIFR) is proposed to obtain the time-frequency features of multichannel sEMG signals. Then, this paper designs the multiple feature fusion network (MFFN), which aims at strengthening the ability of feature extraction. Finally, a deep belief network (DBN) was introduced as the classification model of the MFFN to boost the generalization performance for more types of upper-limb movements. In the experiments, 12 kinds of upper-limb rehabilitation actions were recognized utilizing four sEMG sensors. The maximum identification accuracy was 86.10% and the average classification accuracy of the proposed MFFN was 73.49%, indicating that the time-frequency representation approach combined with the MFFN is superior to the traditional machine learning and convolutional neural network.