Electronic Sleep Stage Classifiers: A Survey and VLSI Design Methodology

Field-Programmable Gate Array (FPGA)-Based Lock-In Amplifier System with Signal Enhancement: A Comprehensive Review on the Design for Advanced Measurement Applications

Lock-in amplifiers (LIAs) are critical tools in precision measurement, particularly for applications involving weak signals obscured by noise. Advances in signal processing algorithms and hardware synthesis have enabled accurate signal extraction, even in extremely noisy environments, making LIAs indispensable in sensor applications for healthcare, industry, and other services. For instance, the electrical impedance measurement of the human body, organs, tissues, and cells, known as bioelectrical impedance, is commonly used in biomedical and healthcare applications because it is non-invasive and relatively inexpensive. Also, due to its portability and miniaturization capabilities, it has great potential for the development of new point-of-care and portable testing devices. In this document, we highlight existing techniques for high-frequency resolution and precise phase detection in LIA reference signals from field-programmable gate array (FPGA) designs. A comprehensive review is presented under the key requirements and techniques for single- and dual-phase digital LIA architectures, where relevant insights are provided to address the LIAs' digital precision in measurement system configurations. Furthermore, the document highlights a novel method to enhance the spurious-free dynamic range (SFDR), thereby advancing the precision and effectiveness of LIAs in complex measurement environments. Finally, we summarize the diverse applications of impedance measurement, highlighting the wide range of fields that can benefit from the design of high performance in modern measurement technologies.

Single-trial detection of auditory cues from the rat brain using memristors

Implantable devices hold the potential to address conditions currently lacking effective treatments, such as drug-resistant neural impairments and prosthetic control. Medical devices need to be biologically compatible while providing enhanced performance metrics of low-power consumption, high accuracy, small size, and minimal latency to enable ongoing intervention in brain function. Here, we demonstrate a memristor-based processing system for single-trial detection of behaviorally meaningful brain signals within a timeframe that supports real-time closed-loop intervention. We record neural activity from the reward center of the brain, the ventral tegmental area, in rats trained to associate a musical tone with a reward, and we use the memristors built-in thresholding properties to detect nontrivial biomarkers in local field potentials. This approach yields consistent and accurate detection of biomarkers >98% while maintaining power consumption as low as 4.14 nanowatt per channel. The efficacy of our system's capabilities to process real-time in vivo neural data paves the way for low-power chronic neural activity monitoring and biomedical implants.

Computational ensemble expert system classification for the recognition of bruxism using physiological signals

This study aimed to develop an automatic diagnostic scheme for bruxism, a sleep-related disorder characterized by teeth grinding and clenching. The aim was to improve on existing methods, which have been proven to be inefficient and challenging. We utilized a novel hybrid machine learning classifier, facilitated by the Weka tool, to diagnose bruxism from biological signals. The study processed and examined these biological signals by calculating the power spectral density. Data were categorized into normal or bruxism categories based on the EEG channel (C4-A1), and the sleeping phases were classified into wake (w) and rapid eye movement (REM) stages using the ECG channel (ECG1-ECG2). The classification resulted in a maximum specificity of 93% and an accuracy of 95% for the EEG-based diagnosis. The ECG-based classification yielded a supreme specificity of 87% and an accuracy of 96%. Furthermore, combining these phases using the EMG channel (EMG1-EMG2) achieved the highest specificity of 95% and accuracy of 98%. The ensemble Weka tool combined all three physiological signals EMG, ECG, and EEG, to classify the sleep stages and subjects. This integration increased the specificity and accuracy to 97% and 99%, respectively. This indicates that a more precise bruxism diagnosis can be obtained by including all three biological signals. The proposed method significantly improves bruxism diagnosis accuracy, potentially enhancing automatic home monitoring systems for this disorder. Future studies may expand this work by applying it to patients for practical use.

Extracting electromyographic signals from multi-channel LFPs using independent component analysis without direct muscular recording

Electromyography (EMG) has been commonly used for the precise identification of animal behavior. However, it is often not recorded together with in vivo electrophysiology due to the need for additional surgeries and setups and the high risk of mechanical wire disconnection. While independent component analysis (ICA) has been used to reduce noise from field potential data, there has been no attempt to proactively use the removed "noise," of which EMG signals are thought to be one of the major sources. Here, we demonstrate that EMG signals can be reconstructed without direct EMG recording using the "noise" ICA component from local field potentials. The extracted component is highly correlated with directly measured EMG, termed IC-EMG. IC-EMG is useful for measuring an animal's sleep/wake, freezing response, and non-rapid eye movement (NREM)/REM sleep states consistently with actual EMG. Our method has advantages in precise and long-term behavioral measurement in wide-ranging in vivo electrophysiology experiments.

A System-on-Chip for Closed-loop Optogenetic Sleep Modulation

Stimulation of target neuronal populations using optogenetic techniques during specific sleep stages has begun to elucidate the mechanisms and effects of sleep. To conduct closed-loop optogenetic sleep studies in untethered animals, we designed a fully integrated, low-power system-on-chip (SoC) for real-time sleep stage classification and stage-specific optical stimulation. The SoC consists of a 4-channel analog front-end for recording polysomnography signals, a mixed-signal machine-learning (ML) core, and a 16-channel optical stimulation back-end. A novel ML algorithm and innovative circuit design techniques improved the online classification performance while minimizing power consumption. The SoC was designed and simulated in 180 nm CMOS technology. In an evaluation using an expert labeled sleep database with 20 subjects, the SoC achieves a high sensitivity of 0.806 and a specificity of 0.947 in discriminating 5 sleep stages. Overall power consumption in continuous operation is 97 µW.

A real-time sleep scoring framework for closed-loop sleep manipulation in mice

Subtle changes in sleep architecture can accompany and be symptomatic of many diseases or disorders. In order to probe and understand the complex interactions between sleep and health, the ability to model, track, and modulate sleep in preclinical animal models is vital. While various methods have been described for scoring experimental sleep recordings, few are designed to work in real time - a prerequisite for closed-loop sleep manipulation. In the present study, we have developed algorithms and software to classify sleep in real time and validated it on C57BL/6 mice (n = 8). Hidden Markov models of baseline sleep dynamics were fitted using an unsupervised algorithm to electroencephalogram (EEG) and electromyogram (EMG) data for each mouse, and were able to classify sleep in a manner highly concordant with manual scoring (Cohen's Kappa >75%) up to 3 weeks after model construction. This approach produced reasonably accurate estimates of common sleep metrics (proportion, mean duration, and number of bouts). After construction, the models were used to track sleep in real time and accomplish selective rapid eye movement (REM) sleep restriction by triggering non-invasive somatosensory stimulation. During REM restriction trials, REM bout duration was significantly reduced, and the classifier continued to perform satisfactorily despite the disrupted sleep patterns. The software can easily be tailored for use with other commercial or customised methods of sleep disruption (e.g. stir bar, optogenetic stimulation, etc.) and could serve as a robust platform to facilitate closed-loop experimentation. The source code and documentation are freely available upon request from the authors.

Review on Biomedical Sensors, Technologies and Algorithms for Diagnosis of Sleep Disordered Breathing: Comprehensive Survey

This paper provides a comprehensive review of available technologies for measurements of vital physiology related parameters that cause sleep disordered breathing (SDB). SDB is a chronic disease that may lead to several health problems and increase the risk of high blood pressure and even heart attack. Therefore, the diagnosis of SDB at an early stage is very important. The essential primary step before diagnosis is measurement. Vital health parameters related to SBD might be measured through invasive or non-invasive methods. Nowadays, with respect to increase in aging population, improvement in home health management systems is needed more than even a decade ago. Moreover, traditional health parameter measurement techniques such as polysomnography are not comfortable and introduce additional costs to the consumers. Therefore, in modern advanced self-health management devices, electronics and communication science are combined to provide appliances that can be used for SDB diagnosis, by monitoring a patient's physiological parameters with more comfort and accuracy. Additionally, development in machine learning algorithms provides accurate methods of analysing measured signals. This paper provides a comprehensive review of measurement approaches, data transmission, and communication networks, alongside machine learning algorithms for sleep stage classification, to diagnose SDB.

A Novel Hybrid Machine Learning Classification for the Detection of Bruxism Patients Using Physiological Signals

Bruxism is a sleep disorder in which the patient clinches and gnashes their teeth. Bruxism detection using traditional methods is time-consuming, cumbersome, and expensive. Therefore, an automatic tool to detect this disorder will alleviate the doctor workload and give valuable help to patients. In this paper, we targeted this goal and designed an automatic method to detect bruxism from the physiological signals using a novel hybrid classifier. We began with data collection. Then, we performed the analysis of the physiological signals and the estimation of the power spectral density. After that, we designed the novel hybrid classifier to enable the detection of bruxism based on these data. The classification of the subjects into “healthy” or “bruxism” from the electroencephalogram channel (C4-A1) obtained a maximum specificity of 92% and an accuracy of 94%. Besides, the classification of the sleep stages such as the wake (w) stage and rapid eye movement (REM) stage from the electrocardiogram channel (ECG1-ECG2) obtained a maximum specificity of 86% and an accuracy of 95%. The combined bruxism classification and the sleep stages classification from the electroencephalogram channel (C4-P4) obtained a maximum specificity of 90% and an accuracy of 97%. The results show that more accurate bruxism detection is achieved by exploiting the electroencephalogram signal (C4-P4). The present work can be applied for home monitoring systems for bruxism detection.

Orthogonal convolutional neural networks for automatic sleep stage classification based on single-channel EEG

BACKGROUND AND OBJECTIVE

In recent years, several automatic sleep stage classification methods based on convolutional neural networks (CNN) by learning hierarchical feature representation automatically from raw EEG data have been proposed. However, the state-of-the-art of such methods are quite complex. Using a simple CNN architecture to classify sleep stages is important for portable sleep devices. In addition, employing CNNs to learn rich and diverse representations remains a challenge. Therefore, we propose a novel CNN model for sleep stage classification.

METHODS

Generally, EEG signals are better described in the frequency domain; thus, we convert EEG data to a time-frequency representation via Hilbert-Huang transform. To learn rich and effective feature representations, we propose an orthogonal convolutional neural network (OCNN). First, we construct an orthogonal initialization of weights. Second, to avoid destroying the orthogonality of the weights in the training process, orthogonality regularizations are proposed to maintain the orthogonality of weights. Simultaneously, a squeeze-and-excitation (SE) block is employed to perform feature recalibration across different channels.

RESULTS

The proposed method achieved a total classification accuracy of 88.4% and 87.6% on two public datasets, respectively. The classification performances of different convolutional neural networks models were compared to that of the proposed method. The experiment results demonstrated that the proposed method is effective for sleep stage classification.

CONCLUSIONS

Experiment results indicate that the proposed OCNN can learn rich and diverse feature representations from time-frequency images of EEG data, which is important for deep learning. In addition, the proposed orthogonality regularization is simple and can be easily adapted to other architectures.

A Resource-Optimized VLSI Implementation of a Patient-Specific Seizure Detection Algorithm on a Custom-Made 2.2 cm 2 Wireless Device for Ambulatory Epilepsy Diagnostics

A patient-specific epilepsy diagnostic solution in the form of a wireless wearable ambulatory device is presented. First, the design, VLSI implementation, and experimental validation of a resource-optimized machine learning algorithm for epilepsy seizure detection are described. Next, the development of a mini-PCB that integrates a low-power wireless data transceiver and a programmable processor for hosting the seizure detection algorithm is discussed. The algorithm uses only EEG signals from the frontal lobe electrodes while yielding a seizure detection sensitivity and specificity competitive to the standard full EEG systems. The experimental validation of the algorithm VLSI implementation proves the possibility of conducting accurate seizure detection using quickly-mountable dry-electrode headsets without the need for uncomfortable/painful through-hair electrodes or adhesive gels. Details of design and optimization of the algorithm, the VLSI implementation, and the mini-PCB development are presented and resource optimization techniques are discussed. The optimized implementation is uploaded on a low-power Microsemi Igloo FPGA, requires 1237 logic elements, consumes 110 μW dynamic power, and yields a minimum detection latency of 10.2 μs. The measurement results from the FPGA implementation on data from 23 patients (198 seizures in total) shows a seizure detection sensitivity and specificity of 92.5% and 80.1%, respectively. Comparison to the state of the art is presented from system integration, the VLSI implementation, and the wireless communication perspectives.

A 9.2-g Fully-Flexible Wireless Ambulatory EEG Monitoring and Diagnostics Headband With Analog Motion Artifact Detection and Compensation

An 8-channel wearable wireless device for ambulatory surface EEG monitoring and analysis is presented. The entire multi-channel recording, quantization, and motion artifact removal circuitries are implemented on a 4-layer polyimide flexible substrate. The recording electrodes and active shielding are also integrated on the same substrate, yielding the smallest form factor compared to the state of the art. Thanks to the dry non-contact electrodes, the system is quickly mountable with minimal assistance required, making it an ideal ambulatory front- and temporal-lobe EEG monitoring device. The flexible main board is connected to a rechargeable battery on one end and to a 13 × 17 mm ² rigid board on the other end. The mini rigid board hosts a low-power programmable FPGA and a BLE 5.0 transceiver, which add diagnostic capability and wireless connectivity features to the device, respectively. Design considerations for a wearable EEG monitoring and diagnostic device are discussed in details. The theory of the novel fully-analog method for motion artifact detection and removal is described and the detailed circuit implementation is presented. The device performance in terms of voltage gain (260 V/V), bandwidth (DC-300 Hz), motion artifact removal, and wireless communication throughput (up to 1 Mbps) is experimentally validated. The entire wearable solution with the battery weighs 9.2 grams.

Complex-valued unsupervised convolutional neural networks for sleep stage classification

BACKGROUND AND OBJECTIVE

Despite numerous deep learning methods being developed for automatic sleep stage classification, almost all the models need labeled data. However, obtaining labeled data is a subjective process. Therefore, the labels will be different between two experts. At the same time, obtaining labeled data also is a time-consuming task. Even an experienced expert requires hours to annotate the sleep stage patterns. More important, as the development of wearable sleep devices, it is very difficult to obtain labeled sleep data. Therefore, unsupervised training algorithm is very important for sleep stage classification. Hence, a new sleep stage classification method named complex-valued unsupervised convolutional neural networks (CUCNN) is proposed in this study.

METHODS

The CUCNN operates with complex-valued inputs, outputs, and weights, and its training strategy is greedy layer-wise training. It is composed of three phases: phase encoder, unsupervised training and complex-valued classification. Phase encoder is used to translate real-valued inputs into complex numbers. In the unsupervised training phase, the complex-valued K-means is used to learn filters which will be used in the convolution.

RESULTS

The classification performances of handcrafted features are compared with those of learned features via CUCNN. The total accuracy (TAC) and kappa coefficient of the sleep stage from UCD dataset are 87% and 0.8, respectively. Moreover, the comparison experiments indicate that the TACs of the CUCNN from UCD and MIT-BIH datasets outperform these of unsupervised convolutional neural networks (UCNN) by 12.9% and 13%, respectively. Additionally, the convergence of CUCNN is much faster than that of UCNN in most cases.

CONCLUSIONS

The proposed method is fully automated and can learn features in an unsupervised fashion. Results show that unsupervised training and automatic feature extraction on sleep data are possible, which are very important for home sleep monitoring.

A New Method for Automatic Sleep Stage Classification

Traditionally, automatic sleep stage classification is quite a challenging task because of the difficulty in translating open-textured standards to mathematical models and the limitations of handcrafted features. In this paper, a new system for automatic sleep stage classification is presented. Compared with existing sleep stage methods, our method can capture the sleep information hidden inside electroencephalography (EEG) signals and automatically extract features from raw data. To translate open sleep stage standards into machine rules recognized by computers, a new model named fast discriminative complex-valued convolutional neural network (FDCCNN) is proposed to extract features from raw EEG data and classify sleep stages. The new model combines complex-valued backpropagation and the Fisher criterion. It can learn discriminative features and overcome the negative effect of imbalance dataset. More importantly, the orthogonal decision boundaries for the real and imaginary parts of a complex-valued convolutional neuron are proven. A speed-up algorithm is proposed to reduce computational workload and yield improvements of over an order of magnitude compared to the normal convolution algorithm. The classification performances of handcrafted features and different convolutional neural networks are compared with that of the FDCCNN. The total accuracy and kappa coefficient of the proposed method are 92% and 0.84, respectively. Experiment results demonstrated that the performance of our system is comparable to those of human experts.