Ya-Wen Tu

Muscle fatigue can result in decreased athletic performance and even muscle damage. Therefore, a deeper understanding of muscle fatigue is essential for enhancing athletic performance and injury prevention. This study collected sEMG signals from 40 participants, and their fatigue levels were quantified using the Borg Rating of Perceived Exertion (RPE) scale. Subsequently, fatigue identification was performed using the model proposed in this study. The model leverages features extracted from both the time and frequency domains of sEMG signals. The time domain branch primarily consists of Temporal Convolutional Networks (TCN) and self-attention mechanisms, while the frequency domain branch comprises a series of 1D Convolutional Neural Networks (CNNs). Finally, the proposed model was employed to classify the collected sEMG signals, with results achieving an accuracy of 90.07%.

Speech sound disorder (SSD) is a common communication disorder among children that can lead to difficulties in accurately articulating speech sounds. Automated detection of SSD can provide timely assessment and screening for a large population of children. This paper presents a novel approach for automatically detecting and diagnosing SSD in Chinese-speaking children. The proposed approach utilizes a text-dependent end-to-end model incorporating linguistic and acoustic features. CTC (Connectionist Temporal Classification) is used as the loss function. We collected a corpus of speech samples from 100 children aged three to nine, focusing on the speech patterns and characteristics of early childhood. Reliable annotations were provided by multiple professional speech and language pathologists (SLPs) for model training and evaluation. Experimental results demonstrate that the proposed approach achieves superior performance compared with the baseline model CNN-RNN-CTC in phoneme recognition, mispronunciation detection and diagnosis, as well as the identification of common phonological processes. The overall average F-measure for recognizing common phonological processes reached 66.16%.

The goal of stroke rehabilitation is to establish a robust protocol for patients to live independently in community. Firstly, we examined the impact of 3 hybridized transcranial direct current stimulation (tDCS)-mirror therapy interventions on activities of daily life (ADL) in stroke patients. Secondly, we explored the underlying therapeutic mechanisms with theory-driven electroencephalography (EEG) indexes in the alpha band. This was achieved by identifying the unique contributions of alpha power in motor production to ADL in relation to the premotor cortex (PMC), primary cortex (M1), and Sham tDCS with mirror therapy. The results showed that, although post-intervention ADL improvement was comparable among the three tDCS groups, one of the EEG indexes differentiated the interventions. Neural-behavioral correlation analyses revealed that different types of ADL improvements consistently corresponded with alpha power in the temporal lobe exclusively in the PMC tDCS group (all rs > 0.39). By contrast, alterations in alpha power in the central-frontal region were found to vary, with ADL primarily in the M1 tDCS group (r = −0.6 or 0.7), with the benefit depending on the complexity of the ADL. In conclusion, this research suggested two potential therapeutic mechanisms and demonstrated the additive benefits of introducing theory-driven neural indexes in explaining ADL.

This article describes a system for analyzing acoustic data to assist in the diagnosis and classification of children’s speech sound disorders (SSDs) using a computer. The analysis concentrated on identifying and categorizing four distinct types of Chinese SSDs. The study collected and generated a speech corpus containing 2540 stopping, backing, final consonant deletion process (FCDP), and affrication samples from 90 children aged 3–6 years with normal or pathological articulatory features. Each recording was accompanied by a detailed diagnostic annotation by two speech–language pathologists (SLPs). Classification of the speech samples was accomplished using three well-established neural network models for image classification. The feature maps were created using three sets of MFCC (Mel-frequency cepstral coefficients) parameters extracted from speech sounds and aggregated into a three-dimensional data structure as model input. We employed six techniques for data augmentation to augment the available dataset while avoiding overfitting. The experiments examine the usability of four different categories of Chinese phrases and characters. Experiments with different data subsets demonstrate the system’s ability to accurately detect the analyzed pronunciation disorders. The best multi-class classification using a single Chinese phrase achieves an accuracy of 74.4 percent.

Surface ElectroMyoGraphy (sEMG) is widely applied to a variety of applications. Managing the power consumption of battery-constrained sEMG Wireless Body Sensor Networks (WBSN) is an important topic. In this paper, we use fatigue assessments as a case study. We apply the concept of distributed computing to explore the impact of computation allocations on the client power consumption and the requirement of architecture specifications. Regarding the CPU clock rate, we propose a power-saving method based on the ping-pong buffer mechanism and evaluate all the crucial factors which affect the power consumption such as sEMG sample rates, algorithmic computational costs, wireless throughputs, and selection of wireless technologies. To sum up, we conduct a comprehensive analysis of all possible distributed computing architectures of WBSN to determine the lowest-power WBSN architecture. The results show that the implementation based on the lowest-power WBSN architecture has lower power consumption compared with other hardwares. The average current of the proposed architecture can be reduced by 81.7% compared with the previous work. Besides, the battery life is 4.48 times that of the previous work under the continuous wireless connection equipped with the same 300mAh lithium battery. Compared with the commercial device, the battery life is 1.6 times that of the commercial one.

We study the foot plantar sensor placement by a deep reinforcement learning algorithm without using any prior knowledge of the foot anatomical area. To apply a reinforcement learning algorithm, we propose a sensor placement environment and reward system that aims to optimize fitting the center of pressure (COP) trajectory during the self-selected speed running task. In this environment, the agent considers placing eight sensors within a 7 × 20 grid coordinate system, and then the final pattern becomes the result of sensor placement. Our results show that this method (1) can generate a sensor placement, which has a low mean square error in fitting ground truth COP trajectory, and (2) robustly discovers the optimal sensor placement in a large number of combinations, which is more than 116 quadrillion. This method is also feasible for solving different tasks, regardless of the self-selected speed running task.

Recently, surface electromyography (sEMG) has been used to detect running-related works. sEMG provides a non-invasive and real-time method that allows quantification of muscle energy. However, noises in sEMG signals are a serious issue to be considered as these will interrupt the analysis of muscular activity. Hence, this work aims at distinguishing between sEMG valid signals and noises during running exercise by taking advantage of the combination of 3D-CNN and LSTM, which we called 3D-LCNN. Furthermore, according to the possible cases that happen in the sEMG data-collection procedure, we proposed two data-augmentation approaches to expend our sEMG dataset, which are the simulation of the surface electrodes displacement on the skin and the muscle fatigue. Experiment results show that the classification accuracy of the proposed 3D-LCNN can achieve 90.52%. Additionally, this work provides excellent service-oriented architecture (SOA). The recognition process can be done after the subject placed the sEMG sensors and performed a trial. Therefore, the process can help clinicians or therapists to distinguish between sEMG valid signals and noises more efficiently.