A Review on Automated Sleep Study.

In recent years, research on automated sleep analysis has witnessed significant growth, reflecting advancements in understanding sleep patterns and their impact on overall health. This review synthesizes findings from an exhaustive analysis of 87 papers, systematically retrieved from prominent databases such as Google Scholar, PubMed, IEEE Xplore, and ScienceDirect. The selection criteria prioritized studies focusing on methods employed, signal modalities utilized, and machine learning algorithms applied in automated sleep analysis. The overarching goal was to critically evaluate the strengths and weaknesses of the proposed methods, shedding light on the current landscape and future directions in sleep research. An in-depth exploration of the reviewed literature revealed a diverse range of methodologies and machine learning approaches employed in automated sleep studies. Notably, K-Nearest Neighbors (KNN), Ensemble Learning Methods, and Support Vector Machine (SVM) emerged as versatile and potent classifiers, exhibiting high accuracies in various applications. However, challenges such as performance variability and computational demands were observed, necessitating judicious classifier selection based on dataset intricacies. In addition, the integration of traditional feature extraction methods with deep structures and the combination of different deep neural networks were identified as promising strategies to enhance diagnostic accuracy in sleep-related studies. The reviewed literature emphasized the need for adaptive classifiers, cross-modality integration, and collaborative efforts to drive the field toward more accurate, robust, and accessible sleep-related diagnostic solutions. This comprehensive review serves as a solid foundation for researchers and practitioners, providing an organized synthesis of the current state of knowledge in automated sleep analysis. By highlighting the strengths and challenges of various methodologies, this review aims to guide future research toward more effective and nuanced approaches to sleep diagnostics.

Adversarial bandit approach for RIS-aided OFDM communication.

To assist sixth-generation wireless systems in the management of a wide variety of services, ranging from mission-critical services to safety-critical tasks, key physical layer technologies such as reconfigurable intelligent surfaces (RISs) are proposed. Even though RISs are already used in various scenarios to enable the implementation of smart radio environments, they still face challenges with regard to real-time operation. Specifically, high dimensional fully passive RISs typically need costly system overhead for channel estimation. This paper, however, investigates a semi-passive RIS that requires a very low number of active elements, wherein only two pilots are required per channel coherence time. While in its infant stage, the application of deep learning (DL) tools shows promise in enabling feasible solutions. We propose two low-training overhead and energy-efficient adversarial bandit-based schemes with outstanding performance gains when compared to DL-based reflection beamforming reference methods. The resulting deep learning models are discussed using state-of-the-art model quality prediction trends.

S-Convnet: A Shallow Convolutional Neural Network Architecture for Neuromuscular Activity Recognition Using Instantaneous High-Density Surface EMG Images.

The recent progress in recognizing low-resolution instantaneous high-density surface electromyography (HD-sEMG) images opens up new avenues for the development of more fluid and natural muscle-computer interfaces. However, the existing approaches employed a very large deep convolutional neural network (ConvNet) architecture and complex training schemes for HD-sEMG image recognition, which requires learning of >5.63 million(M) training parameters only during fine-tuning and pre-trained on a very large-scale labeled HD-sEMG training dataset, as a result, it makes high-end resource-bounded and computationally expensive. To overcome this problem, we propose S-ConvNet models, a simple yet efficient framework for learning instantaneous HD-sEMG images from scratch using random-initialization. Without using any pre-trained models, our proposed S-ConvNet demonstrate very competitive recognition accuracy to the more complex state of the art, while reducing learning parameters to only ≈ 2M and using ≈ 12 × smaller dataset. The experimental results proved that the proposed S-ConvNet is highly effective for learning discriminative features for instantaneous HD-sEMG image recognition, especially in the data and high-end resource-constrained scenarios.

Surrogate analysis of fractal dimensions from SEMG sensor array as a predictor of chronic low back pain.

In this paper, a method based on nonlinear analysis of sEMG sensor array signals (2 arrays of 5×13 sensors) to detect chronic low back pain is presented. The use of an FFT based surrogate analysis method isolates the nonlinear structure of the signals from the effect of the power spectrum. The fractal dimension is used for the nonlinear characteristic. From the sensor arrays, a certain number of channels which exhibits the most nonlinearity for a subject are kept as input of a small neural network. A leave-one-out type cross-validation method shows a success rate of 80%.

Trunk motor variability in patients with non-specific chronic low back pain.

PURPOSE: To identify and characterize trunk neuromuscular adaptations during muscle fatigue in patients with chronic low back pain (LBP) and healthy participants. METHODS: Forty-six patients with non-specific chronic LBP and 23 healthy controls were asked to perform a trunk muscles fatigue protocol. Surface electromyography was recorded using two adhesive matrix of 64 electrodes applied bilaterally over the erector spinae. Pain score, kinesiophobia and physical disability were analyzed through different questionnaires. To characterize motor variability, dispersion of muscular activity center of gravity was computed. Motor variability between groups was compared using repeated-measures analyses of variance. RESULTS: Score of disability and kinesiophobia were significantly higher in patients with LBP. Results indicated a significant group effect characterized by an increased motor variability in the healthy group through the entire fatigue task on the left (p = 0.003) and right side (p = 0.048). Interestingly, increasing muscle fatigue led to increased motor variability in both groups (on both sides (p < 0.001) but with a greater increase in the healthy group. CONCLUSION: Muscle recruitment is altered in patients with chronic LBP in the presence of muscle fatigue. Consequently, these patients exhibit changes in muscle recruitment pattern and intensity (lower levels of motor variability) during sustained isometric contraction that may be attributed to variation in the control of motor units within and between muscles. However, patients with LBP are able to increase their motor variability over time but with a lower increase compared to healthy participants.

High-Speed OQPSK and Efficient Power Transfer Through Inductive Link for Biomedical Implants.

Biomedical implants require wireless power and bidirectional data transfer. We pursue our previous work on a novel topology for a multiple carrier inductive link by presenting the fabricated coils. We show that the coplanar geometry approach is better suited for displacement tolerance. We provide a theoretical analysis of the efficiency of power transfer and phase-shift-keying communications through an inductive link. An efficiency of up to 61% has been achieved experimentally for power transfer and a data rate of 4.16 Mb/s with a bit-error rate of less than 2 × 10(-6) has been obtained with our fabricated offset quadrature phase-shift keying modules due to the inductive link optimization presented in this paper.