The fall of elderly patients is still a critical medical issue since it can cause irreversible bone injuries due to the elderly bones weakness. To mitigate the likelihood of the occurrence of a fall, continuously tracking the patients with balance and health issues has been envisaged, despite being unpractical. To address this problem, we propose an efficient automatic fall detection system which is also fitted for the detection of different Activities of Daily Living (ADL). The system relies on a wearable Shimmer device, to transmit some inertial signals via a wireless connection to a computer. Aiming at reducing the size of the transmitted data and minimizing the energy consumption, a compressive sensing (CS) method is applied. In this perspective, we started by creating our dataset from 17 subjects performing a set of movements, then three distinct systems were investigated: one which detects the presence or the absence of the fall, a second which detects static or dynamic movements including the fall, and a third which recognizes the fall and six other ADL activities. In the acquisition and classification steps, first only the data collected by the accelerometer are exploited, then a mixture of the accelerometer and gyroscope measurements are taken into consideration. The two configurations are compared and the resulting system incorporating CS capabilities is shown to achieve up to 99.8% of accuracy.

No abstract

Fatigue is defined as “a loss of force-generating capacity” in a muscle that can intensify tremor. Tremor quantification can facilitate early detection of fatigue onset so that preventative or corrective controls can be taken to minimize work-related injuries and improve the performance of tasks that require high-levels of accuracy. We focused on developing a system that recognizes and classifies voluntary effort and detects phases of fatigue. The experiment was designed to extract and evaluate hand-tremor data during the performance of both rest and effort tasks. The data were collected from the wrist and finger of the participant’s dominant hand. To investigate tremor, time, frequency domain features were extracted from the accelerometer signal for segments of 45 and 90 samples/window. Analysis using advanced signal processing and machine-learning techniques such as decision tree, k-nearest neighbor, support vector machine, and ensemble classifiers were applied to discover models to classify rest and effort tasks and the phases of fatigue. Evaluation of the classifier’s performance was assessed based on various metrics using 5-fold cross-validation. The recognition of rest and effort tasks using an ensemble classifier based on the random subspace and window length of 45 samples was deemed to be the most accurate (96.1%). The highest accuracy (~98%) that distinguished between early and late fatigue phases was achieved using the same classifier and window length.

No abstract