RESUMEN
Deep learning associated with neurological signals is poised to drive major advancements in diverse fields such as medical diagnostics, neurorehabilitation, and brain-computer interfaces. The challenge in harnessing the full potential of these signals lies in the dependency on extensive, high-quality annotated data, which is often scarce and expensive to acquire, requiring specialized infrastructure and domain expertise. To address the appetite for data in deep learning, we present Neuro-BERT, a self-supervised pre-training framework of neurological signals based on masked autoencoding in the Fourier domain. The intuition behind our approach is simple: frequency and phase distribution of neurological signals can reveal intricate neurological activities. We propose a novel pre-training task dubbed Fourier Inversion Prediction (FIP), which randomly masks out a portion of the input signal and then predicts the missing information using the Fourier inversion theorem. Pre-trained models can be potentially used for various downstream tasks such as sleep stage classification and gesture recognition. Unlike contrastive-based methods, which strongly rely on carefully hand-crafted augmentations and siamese structure, our approach works reasonably well with a simple transformer encoder with no augmentation requirements. By evaluating our method on several benchmark datasets, we show that Neuro-BERT improves downstream neurological-related tasks by a large margin.
RESUMEN
In recent years, deep learning (DL) has emerged as a powerful tool in clinical imaging, offering unprecedented opportunities for the diagnosis and treatment of neurological disorders (NDs). This comprehensive review explores the multifaceted role of DL techniques in leveraging vast datasets to advance our understanding of NDs and improve clinical outcomes. Beginning with a systematic literature review, we delve into the utilization of DL, particularly focusing on multimodal neuroimaging data analysis-a domain that has witnessed rapid progress and garnered significant scientific interest. Our study categorizes and critically analyses numerous DL models, including Convolutional Neural Networks (CNNs), LSTM-CNN, GAN, and VGG, to understand their performance across different types of Neurology Diseases. Through particular analysis, we identify key benchmarks and datasets utilized in training and testing DL models, shedding light on the challenges and opportunities in clinical neuroimaging research. Moreover, we discuss the effectiveness of DL in real-world clinical scenarios, emphasizing its potential to revolutionize ND diagnosis and therapy. By synthesizing existing literature and describing future directions, this review not only provides insights into the current state of DL applications in ND analysis but also covers the way for the development of more efficient and accessible DL techniques. Finally, our findings underscore the transformative impact of DL in reshaping the landscape of clinical neuroimaging, offering hope for enhanced patient care and groundbreaking discoveries in the field of neurology. This review paper is beneficial for neuropathologists and new researchers in this field.
RESUMEN
Human visual neurons rely on event-driven, energy-efficient spikes for communication, while silicon image sensors do not. The energy-budget mismatch between biological systems and machine vision technology has inspired the development of artificial visual neurons for use in spiking neural network (SNN). However, the lack of multiplexed data coding schemes reduces the ability of artificial visual neurons in SNN to emulate the visual perception ability of biological systems. Here, we present an artificial visual spiking neuron that enables rate and temporal fusion (RTF) coding of external visual information. The artificial neuron can code visual information at different spiking frequencies (rate coding) and enables precise and energy-efficient time-to-first-spike (TTFS) coding. This multiplexed sensory coding scheme could improve the computing capability and efficacy of artificial visual neurons. A hardware-based SNN with the RTF coding scheme exhibits good consistency with real-world ground truth data and achieves highly accurate steering and speed predictions for self-driving vehicles in complex conditions. The multiplexed RTF coding scheme demonstrates the feasibility of developing highly efficient spike-based neuromorphic hardware.
Asunto(s)
Potenciales de Acción , Redes Neurales de la Computación , Neuronas , Percepción Visual , Humanos , Neuronas/fisiología , Potenciales de Acción/fisiología , Percepción Visual/fisiología , Modelos NeurológicosRESUMEN
Cloud-based training and edge-based inference modes for Artificial Intelligence of Medical Things (AIoMT) applications suffer from accuracy degradation due to physiological signal variations among patients. On-chip learning can overcome this issue by online adaptation of neural network parameters for user-specific tasks. However, existing on-chip learning processors have limitations in terms of versatility, resource utilization, and energy efficiency. We propose HybMED, which is a novel neural signal processor that supports on-chip hybrid neural network training using a composite direct feedback alignment-based paradigm. HybMED is suitable for general-purpose health monitoring AIoMT devices. It improves resource utilization and area efficiency by the reconfigurable homogeneous core with heterogeneous data flow and enhances energy efficiency by exploiting sparsity at different granularities. The chip was fabricated by TSMC 40nm process and tested in multiple physiological signal processing tasks, demonstrating an average improvement in accuracy of 41.16% following online few-shot learning. The chip demonstrates an area efficiency of 1.17 GOPS/mm2 and an energy efficiency of 1.58 TOPS/W. Compared to the previous state-of-the-art physiological signal processors with on-chip learning, the chip achieves a 65× improvement in area efficiency and 1.48× improvement in energy efficiency, respectively.
RESUMEN
Implantable neuromodulation devices have significantly advanced treatments for neurological disorders such as Parkinson's disease, epilepsy, and depression. Traditional open-loop devices like deep brain stimulation (DBS) and spinal cord stimulators (SCS) often lead to overstimulation and lack adaptive precision, raising safety and side-effect concerns. Next-generation closed-loop systems offer real-time monitoring and on-device diagnostics for responsive stimulation, presenting a significant advancement for treating a range of brain diseases. However, the high false alarm rates of current closed-loop technologies limit their efficacy and increase energy consumption due to unnecessary stimulations. In this study, we introduce an artificial intelligence-integrated circuit co-design that targets these issues and using an online demonstration system for closed-loop seizure prediction to showcase its effectiveness. Firstly, two neural network models are obtained with neural-network search and quantization strategies. A binary neural network is optimized for minimal computation with high sensitivity and a convolutional neural network with a false alarm rate as low as 0.1/h for false alarm rejection. Then, a dedicated low-power processor is fabricated in 55 nm technology to implement the two models. With reconfigurable design and event-driven processing feature the resulting application-specific integrated circuit (ASIC) occupies only 5mm2 silicon area and the average power consumption is 142 µW. The proposed solution achieves a significant reduction in both false alarm rates and power consumption when benchmarked against state-of-the-art counterparts.
RESUMEN
This paper presents a low-power frequency-domain functional near-infrared spectroscopy (FD-fNIRS) readout circuit for the absolute value measurement of tissue optical characteristics. The paper proposes a mixer-first analog front-end (AFE) structure and a 1-bit Σ-Δ phase-to-digital converter (PDC) to reduce the required circuit bandwidth and the laser modulation frequency, thereby saving power while maintaining high resolution. The proposed chip achieves sub-0.01° phase resolution and consumes 6.8 mW of power. Nine optical solid phantoms are produced to evaluate the chip. Compared to a self-built high-precision measurement platform that combines a network analyzer with an avalanche photodiode (APD) module, the maximum measuring errors of the absorption coefficient and reduced scattering coefficient are 10.6% and 12.3%, respectively.
RESUMEN
Introduction: Language impairments often result from severe neurological disorders, driving the development of neural prosthetics utilizing electrophysiological signals to restore comprehensible language. Previous decoding efforts primarily focused on signals from the cerebral cortex, neglecting subcortical brain structures' potential contributions to speech decoding in brain-computer interfaces. Methods: In this study, stereotactic electroencephalography (sEEG) was employed to investigate subcortical structures' role in speech decoding. Two native Mandarin Chinese speakers, undergoing sEEG implantation for epilepsy treatment, participated. Participants read Chinese text, with 1-30, 30-70, and 70-150 Hz frequency band powers of sEEG signals extracted as key features. A deep learning model based on long short-term memory assessed the contribution of different brain structures to speech decoding, predicting consonant articulatory place, manner, and tone within single syllable. Results: Cortical signals excelled in articulatory place prediction (86.5% accuracy), while cortical and subcortical signals performed similarly for articulatory manner (51.5% vs. 51.7% accuracy). Subcortical signals provided superior tone prediction (58.3% accuracy). The superior temporal gyrus was consistently relevant in speech decoding for consonants and tone. Combining cortical and subcortical inputs yielded the highest prediction accuracy, especially for tone. Discussion: This study underscores the essential roles of both cortical and subcortical structures in different aspects of speech decoding.
RESUMEN
Optical-based imaging has improved from early single-location research to further sophisticated imaging in 2D topography and 3D tomography. These techniques have the benefit of high specificity and non-radiative safety for brain detection and therapy. However, their performance is limited by complex tissue structures. To overcome the difficulty in successful brain imaging applications, we conducted a simulation using 16 optical source types within a brain model that is based on the Monte Carlo method. In addition, we propose an evaluation method of the optical propagating depth and resolution, specifically one based on the optical distribution for brain applications. Based on the results, the best optical source types were determined in each layer. The maximum propagating depth and corresponding source were extracted. The optical source propagating field width was acquired in different depths. The maximum and minimum widths, as well as the corresponding source, were determined. This paper provides a reference for evaluating the optical propagating depth and resolution from an optical simulation aspect, and it has the potential to optimize the performance of optical-based techniques.
RESUMEN
Alzheimer's disease (AD) progressively impairs the memory and thinking skills of patients, resulting in a significant global economic and social burden each year. However, diagnosis of this neurodegenerative disorder can be challenging, particularly in the early stages of developing cognitive decline. Current clinical techniques are expensive, laborious, and invasive, which hinders comprehensive studies on Alzheimer's biomarkers and the development of efficient devices for Point-of-Care testing (POCT) applications. To address these limitations, researchers have been investigating various biosensing techniques. Unfortunately, these methods have not been commercialized due to several drawbacks, such as low efficiency, reproducibility, and the lack of accurate identification of AD markers. In this review, we present diverse promising hallmarks of Alzheimer's disease identified in various biofluids and body behaviors. Additionally, we thoroughly discuss different biosensing mechanisms and the associated challenges in disease diagnosis. In each context, we highlight the potential of realizing new biosensors to study various features of the disease, facilitating its early diagnosis in POCT. This comprehensive study, focusing on recent efforts for different aspects of the disease and representing promising opportunities, aims to conduct the future trend toward developing a new generation of compact multipurpose devices that can address the challenges in the early detection of AD.
RESUMEN
Superantigens are virulence factors secreted by microorganisms that can cause various immune diseases, such as overactivating the immune system, resulting in cytokine storms, rheumatoid arthritis, and multiple sclerosis. Some studies have demonstrated that superantigens do not require intracellular processing and instated bind as intact proteins to the antigen-binding groove of major histocompatibility complex II on antigen-presenting cells, resulting in the activation of T cells with different T-cell receptor Vß and subsequent overstimulation. To combat superantigen-mediated diseases, researchers have employed different approaches, such as antibodies and simulated peptides. However, due to the complex nature of superantigens, these approaches have not been entirely successful in achieving optimal therapeutic outcomes. CD28 interacts with members of the B7 molecule family to activate T cells. Its mimicking peptide has been suggested as a potential candidate to block superantigens, but it can lead to reduced T-cell activity while increasing the host's infection risk. Thus, this review focuses on the use of drug delivery methods to accurately target and block superantigens, while reducing the adverse effects associated with CD28 mimic peptides. We believe that this method has the potential to provide an effective and safe therapeutic strategy for superantigen-mediated diseases.
Asunto(s)
Anticuerpos , Antígenos CD28 , Células Presentadoras de Antígenos , Péptidos , SuperantígenosRESUMEN
Brain-machine interfaces (BMI) are widely adopted in neuroscience investigations and neural prosthetics, with sensing channel counts constantly increasing. These Investigations place increasing demands for high data rates and low-power implantable devices despite high tissue losses. The Impulse radio ultra-wideband (IR-UWB), a revived wireless technology for short-range radios, has been widely used in various applications. Since the requirements and solutions are application-oriented, in this review paper we focus on neural recording implants with high-data rates and ultra-low power requirements. We examine in detail the working principle, design methodology, performance, and implementations of different architectures of IR-UWB transceivers in a quantitative manner to draw a deep comparison and extract the bottlenecks and possible solutions concerning the dedicated application. Our analysis shows that current solutions rely on enhanced or combined modulation techniques to improve link margin. An in-depth study of prior-art publications that achieved Gbps data rates concludes that edge-combination architecture and non-coherent detectors are remarkable for transmitter and receiver, respectively. Although the aim to minimize power and improve data rate - defined as energy efficiency (pJ/b) - extending communication distance despite high tissue losses and limited power budget, good narrow-band interference (NBI) tolerance coexisted in the same frequency band of UWB systems, and compatibility with energy harvesting designs are among the critical challenges remained unsolved. Furthermore, we expect that the combination of artificial intelligence (AI) and the inherent advantages of UWB radios will pave the way for future improvements in BMIs.
Asunto(s)
Interfaces Cerebro-Computador , Prótesis Neurales , Inteligencia Artificial , Prótesis e Implantes , Tecnología InalámbricaRESUMEN
Connectivity analyses of intracranial electroencephalography (iEEG) could guide surgical planning for epilepsy surgery by improving the delineation of the seizure onset zone. Traditional approaches fail to quantify important interactions between frequency components. To assess if effective connectivity based on cross-bispectrum -a measure of nonlinear multivariate cross-frequency coupling- can quantitatively identify generators of seizure activity, cross-bispectrum connectivity between channels was computed from iEEG recordings of 5 patients (34 seizures) with good postsurgical outcome. Personalized thresholds of 50% and 80% of the maximum coupling values were used to identify generating electrode channels. In all patients, outflow coupling between α (8-15 Hz) and ß (16-31 Hz) frequencies identified at least one electrode inside the resected seizure onset zone. With the 50% and 80% thresholds respectively, an average of 5 (44.7%; specificity = 82.6%) and 2 (22.5%; specificity = 99.0%) resected electrodes were correctly identified. Results show promise for the automatic identification of the seizure onset zone based on cross-bispectrum connectivity analysis.
Asunto(s)
Electrocorticografía , Epilepsia , Humanos , Encéfalo/diagnóstico por imagen , Electroencefalografía/métodos , Convulsiones/diagnósticoRESUMEN
Tamm Plasmon Polariton (TPP) is a nanophotonic phenomenon that has attracted much attention due to its spatial strong field confinement, ease of mode excitation, and polarization independence. TPP has applications in sensing, storage, lasing, perfect absorber, solar cell, nonlinear optics, and many others. In this work, we demonstrate a biosensing platform based on TPP resonant mode. Both theoretical analyses based on the transfer matrix method and experimental validation through nonspecific detection of liquids of different refractive indices and specific detection of SARS-CoV-2 nucleocapsid protein (N-protein) are presented. Results show that the TPP biosensor has high sensitivity and good specificity. For N-protein detection, the sensitivity can be up to 1.5 nm/(µg/mL), and the limit of detection can reach down to 7 ng/mL with a spectrometer of 0.01 nm resolution in wavelength shift. Both nonspecific detection of R.I. liquids and specific detection of N-protein have been simulated and compared with experimental results to demonstrate consistency. This work paves the way for design, optimization, fabrication, characterization, and performance analysis of TPP based biosensors.
Asunto(s)
Técnicas Biosensibles , Silicio , Porosidad , Técnicas Biosensibles/métodos , Refractometría , Óptica y FotónicaRESUMEN
A hypoglossal nerve stimulator (HGNS) is an invasive device that is used to treat obstructive sleep apnea (OSA) through electrical stimulation. The conventional implantable HGNS device consists of a stimuli generator, a breathing sensor, and electrodes connected to the hypoglossal nerve via leads. However, this implant is bulky and causes significant trauma. In this paper, we propose a minimally invasive HGNS based on an electrocardiogram (ECG) sensor and wireless power transfer (WPT), consisting of a wearable breathing monitor and an implantable stimulator. The breathing external monitor utilizes an ECG sensor to identify abnormal breathing patterns associated with OSA with 88.68% accuracy, achieved through the utilization of a convolutional neural network (CNN) algorithm. With a skin thickness of 5 mm and a receiving coil diameter of 9 mm, the power conversion efficiency was measured as 31.8%. The implantable device, on the other hand, is composed of a front-end CMOS power management module (PMM), a binary-phase-shift-keying (BPSK)-based data demodulator, and a bipolar biphasic current stimuli generator. The PMM, with a silicon area of 0.06 mm2 (excluding PADs), demonstrated a power conversion efficiency of 77.5% when operating at a receiving frequency of 2 MHz. Furthermore, it offers three-voltage options (1.2 V, 1.8 V, and 3.1 V). Within the data receiver component, a low-power BPSK demodulator was ingeniously incorporated, consuming only 42 µW when supplied with a voltage of 0.7 V. The performance was achieved through the implementation of the self-biased phase-locked-loop (PLL) technique. The stimuli generator delivers biphasic constant currents, providing a 5 bit programmable range spanning from 0 to 2.4 mA. The functionality of the proposed ECG- and WPT-based HGNS was validated, representing a highly promising solution for the effective management of OSA, all while minimizing the trauma and space requirements.
Asunto(s)
Terapia por Estimulación Eléctrica , Apnea Obstructiva del Sueño , Humanos , Terapia por Estimulación Eléctrica/métodos , Nervio Hipogloso , Apnea Obstructiva del Sueño/terapia , Prótesis e Implantes , ElectrocardiografíaRESUMEN
Surface-enhanced Raman scattering (SERS) is an ultrasensitive spectroscopic technique that can identify materials and chemicals based on their inelastic light-scattering properties. In general, SERS relies on sub-10 nm nanogaps to amplify the Raman signals and achieve ultralow-concentration identification of analytes. However, large-sized analytes, such as proteins and viruses, usually cannot enter these tiny nanogaps, limiting the practical applications of SERS. Herein, we demonstrate a universal SERS platform for the reliable and sensitive identification of a wide range of analytes. The key to this success is the prepared "slot-under-groove" nanoarchitecture arrays, which could realize a strongly coupled field enhancement with a large spatial mode distribution via the hybridization of gap-surface plasmons in the upper V-groove and localized surface plasmon resonance in the lower slot. Therefore, our slot-under-groove platform can simultaneously deliver high sensitivity for small-sized analytes and the identification of large-sized analytes with a large Raman gain.
RESUMEN
Chips-based platforms intended for single-cell manipulation are considered powerful tools to analyze intercellular interactions and cellular functions. Although the conventional cell co-culture models could investigate cell communication to some extent, the role of a single cell requires further analysis. In this study, a precise intercellular interaction model was built using a microelectrode array [microelectrode array (MEA)]-based and dielectrophoresis-driven single-cell manipulation chip. The integrated platform enabled precise manipulation of single cells, which were either trapped on or transferred between electrodes. Each electrode was controlled independently to record the corresponding cellular electrophysiology. Multiple parameters were explored to investigate their effects on cell manipulation including the diameter and depth of microwells, the geometry of cells, and the voltage amplitude of the control signal. Under the optimized microenvironment, the chip was further evaluated using 293T and neural cells to investigate the influence of electric field on cells. An examination of the inappropriate use of electric fields on cells revealed the occurrence of oncosis. In the end of the study, electrophysiology of single neurons and network of neurons, both differentiated from human induced pluripotent stem cells (iPSC), was recorded and compared to demonstrate the functionality of the chip. The obtained preliminary results extended the nature growing model to the controllable level, satisfying the expectation of introducing more elaborated intercellular interaction models.
RESUMEN
The effective control of infectious diseases, including Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) infection, depends on the availability of rapid and accurate monitoring techniques. However, conventional SARS-CoV-2 detection technologies do not support continuous self-detection and may lead to cross-infection when utilized in medical institutions. In this study, we introduce a prototype of a mask biosensor designed for the long-term collection and self-detection of SARS-CoV-2. The biosensor utilizes the average resonance Rayleigh scattering intensity of Au nanocluster-aptamers. The inter-mask surface serves as a medium for the long-term collection and concentration enhancement of SARS-CoV-2, while the heterogeneous-nucleation nanoclusters (NCs) contribute to the exceptional stability of Au NCs for up to 48 h, facilitated by the adhesion of Ti NCs. Additionally, the biosensors based on Au NC-aptamers exhibited high sensitivity for up to 1 h. Moreover, through the implementation of a support vector machine classifier, a significant number of point signals can be collected and differentiated, leading to improved biosensor accuracy. These biosensors offer a complementary wearable device-based method for diagnosing SARS-CoV-2, with a limit of detection of 103 copies. Given their flexibility, the proposed biosensors possess tremendous potential for the continuous collection and sensitive self-detection of SARS-CoV-2 variants and other infectious pathogens.
RESUMEN
We describe a machine learning (ML) approach to processing the signals collected from a COVID-19 optical-based detector. Multilayer perceptron (MLP) and support vector machine (SVM) were used to process both the raw data and the feature engineering data, and high performance for the qualitative detection of the SARS-CoV-2 virus with concentration down to 1 TCID50/mL was achieved. Valid detection experiments contained 486 negative and 108 positive samples, and control experiments, in which biosensors without antibody functionalization were used to detect SARS-CoV-2, contained 36 negative samples and 732 positive samples. The data distribution patterns of the valid and control detection dataset, based on T-distributed stochastic neighbor embedding (t-SNE), were used to study the distinguishability between positive and negative samples and explain the ML prediction performance. This work demonstrates that ML can be a generalized effective approach to process the signals and the datasets of biosensors dependent on resonant modes as biosensing mechanism.
RESUMEN
A novel aptasensor has been designed for quantitative monitoring of epinephrine (EP) based on cerium metal-organic framework (CeMOF) loaded gold nanoparticles (AuNPs). The aptamer, specific to EP, is immobilized on a flexible screen-printed electrode modified with AuNPs@CeMOF, enabling highly selective binding to the target biomolecule. Under optimized operational conditions, the peak currents using voltammetric detection measured at voltage of 83 mV (vs. Ag/AgCl) for epinephrine exhibit a linear increase within concentration in the range 1 pM-10 nM. Following this detection strategy, a boasted limit of detection of 0.3 pM was achieved, surpassing the sensitivity of most reported methods. The developed biosensor showcased exceptional performance in detection of EP in spiked serum sample, with remarkable recovery range of 95.8-113% and precision reflected by low relative standard deviation (RSD) ranging from 2.23 to 6.19%. These results indicate the potential utility of this biosensor as a valuable clinical diagnostic tool. Furthermore, in vitro experiments were carried out using the presented biosensor to monitor the release of epinephrine from PC12 cells upon extracellular stimulation with K+ ions.
Asunto(s)
Aptámeros de Nucleótidos , Nanopartículas del Metal , Estructuras Metalorgánicas , Oro/química , Nanopartículas del Metal/química , Aptámeros de Nucleótidos/química , Estructuras Metalorgánicas/química , EpinefrinaRESUMEN
Machine learning on electromyography (EMG) has recently achieved remarkable success on various tasks, while such success relies heavily on the assumption that the training and future data must be of the same data distribution. However, this assumption may not hold in many real-world applications. Model calibration is required via data re-collection and label annotation, which is generally very expensive and time-consuming. To address this issue, transfer learning (TL), which aims to improve target learners' performance by transferring knowledge from related source domains, is emerging as a new paradigm to reduce the amount of calibration effort. This survey assesses the eligibility of more than fifty published peer-reviewed representative transfer learning approaches for EMG applications. Unlike previous surveys on purely transfer learning or EMG-based machine learning, this survey aims to provide insight into the biological foundations of existing transfer learning methods on EMG-related analysis. Specifically, we first introduce the muscles' physiological structure, the EMG generating mechanism, and the recording of EMG to provide biological insights behind existing transfer learning approaches. Further, we categorize existing research endeavors into data based, model based, training scheme based, and adversarial based. This survey systematically summarizes and categorizes existing transfer learning approaches for EMG related machine learning applications. In addition, we discuss possible drawbacks of existing works and point out the future direction of better EMG transfer learning algorithms to enhance practicality for real-world applications.