Sensors (Basel, Switzerland)
-
Wearable inertial measurement unit (IMU) sensors are powerful enablers for acquisition of motion data. Specifically, in human activity recognition (HAR), IMU sensor data collected from human motion are categorically combined to formulate datasets that can be used for learning human activities. However, successful learning of human activities from motion data involves the design and use of proper feature representations of IMU sensor data and suitable classifiers. ⋯ The proposed feature extraction approach combined with the data augmentation ensemble produces state-of-the-art accuracy results in HAR. A performance evaluation of each augmentation approach is performed to show the influence on classification accuracy. Finally, in addition to using our own dataset, the proposed data augmentation technique is evaluated against the University of California, Irvine (UCI) public online HAR dataset and yields state-of-the-art accuracy results at various learning rates.
-
Comparative Study Controlled Clinical Trial
Continuous Monitoring of Respiratory Rate in Emergency Admissions: Evaluation of the RespiraSense™ Sensor in Acute Care Compared to the Industry Standard and Gold Standard.
Respiratory Rate (RR) is the best marker to indicate deterioration but measurement are often inaccurate. The RespiraSense™ is a non-invasive, wireless, body worn, motion-tolerant and continuous respiratory rate monitor. We aimed to determine whether the performance of RespiraSense™ was equivalent to that of a gold standard measurement technique of capnography and the industry standard of manual counts. ⋯ At rest, RespiraSense™ has a bias of 0.38 and limits of agreement of 1.0 to 1.8 bpm, when compared to the capnography derived RR. Measurements were within pre-defined limits of error at rest. Continuous measurement of RR with RespiraSense™ in patients admitted as acute emergencies is both feasible and reliable.
-
Machine learning has emerged with big data technologies and high-performance computing to create new opportunities for data intensive science in the multi-disciplinary agri-technologies domain. In this paper, we present a comprehensive review of research dedicated to applications of machine learning in agricultural production systems. ⋯ The filtering and classification of the presented articles demonstrate how agriculture will benefit from machine learning technologies. By applying machine learning to sensor data, farm management systems are evolving into real time artificial intelligence enabled programs that provide rich recommendations and insights for farmer decision support and action.
-
In the last few years, estimating ground reaction forces by means of wearable sensors has come to be a challenging research topic paving the way to kinetic analysis and sport performance testing outside of labs. One possible approach involves estimating the ground reaction forces from kinematic data obtained by inertial measurement units (IMUs) worn by the subject. As estimating kinetic quantities from kinematic data is not an easy task, several models and protocols have been developed over the years. ⋯ The identified papers were classified according to the methodology proposed: (i) methods based on direct modelling; (ii) methods based on machine learning. The methods based on direct modelling were further classified according to the task studied (walking, running, jumping, etc.). Finally, we comparatively examined the methods in order to identify the most reliable approaches for the implementation of a ground reaction force estimator based on IMU data.
-
Finding a target quickly is one of the most important tasks in drone operations. In particular, rapid target detection is a critical issue for tasks such as finding rescue victims during the golden period, environmental monitoring, locating military facilities, and monitoring natural disasters. Therefore, in this study, an improved hierarchical probabilistic target search algorithm based on the collaboration of drones at different altitudes is proposed. ⋯ Several drone collaboration scenarios that can be performed by two drones at different altitudes are described and compared to the proposed algorithm. Through simulations, the performances of the proposed algorithm and the cooperation scenarios are analyzed. It is demonstrated that methods utilizing hierarchical searches with drones are comparatively excellent and that the proposed algorithm is approximately 13% more effective than a previous method and much better compared to other scenarios.