Jiang Wang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Jiang Wang has a total of 264 co-authors. Here are the top co-authors by number of shared publications:
Bin Deng
School of Electrical and Information Engineering, Tianjin University
39 shared publications- Chen Liu
School of Electrical and Information Engineering, Tianjin Key Laboratory of Process Measurement and Control, Tianjin University, Tianjin, China
32 shared publications - Bin Deng
School of Electrical and Information Engineering, Tianjin University, Tianjin 300072, China
24 shared publications - Xile Wei
School of Electrical and Information Engineering, Tianjin University, Tianjin 300072, China
21 shared publications - Guosheng Yi
School of Electrical and Information Engineering, Tianjin University, Tianjin, China
20 shared publications - Zhen Zhang
State Key Laboratory of Smart Power Distribution Equipment and System, and School of Electrical and Information Engineering, Tianjin University, Tianjin, China
18 shared publications
Latest Publications
- 29 May 2026
- Published by IOP Publishing in Journal of Neural Engineering
- Vol. 23 (3) , 036023
Physiology-aware latent shift adaptation for sEMG-based hand kinematics decoding across able-bodied and amputee populations
Objective . Decoding continuous hand kinematics from surface electromyography (sEMG) across able-bodied and amputee populations remains a significant challenge, as amputation induces systematic alterations in neural drive and muscle activation patterns that reshape the generative relationship between motor intentions and observed sEMG signals. Such physiology-driven changes are not explicitly modeled by purely statistical domain alignment approaches, limiting their ability to achieve robust and interpretable cross-population decoding. Approach . We develop an integrated approach, termed physiology-aware latent shift adaptation (PALSA), which combines biophysical modeling with latent-space domain adaptation to characterize cross-population discrepancies at the level of latent neuromuscular representation. PALSA first learns a multidimensional, physiologically interpretable synergy latent space from able-bodied subjects (Ninapro DB2) by integrating a differentiable, motor-unit-based physiology layer. For amputee adaptation (Ninapro DB3), PALSA freezes both the neural encoder and kinematic regressor while introducing a compact, learnable latent shift vector to compensate for population-specific deviations within the learned neuromuscular manifold. Main results . Since Ninapro DB3 does not provide paired kinematics, we evaluate population-level plausibility rather than subject-specific accuracy, using DB9-derived calibrated joint angles as a standardized kinematic reference. PALSA outperforms the classic UDA algorithms CORAL and DANN in transfer learning for motor intention recognition of amputees while effectively characterizing the physiological differences between the two population groups. Across 40 gestures, PALSA achieves a mean cosine similarity of 0.6238 and a mean joint-angle root mean square error of 22.60 ∘ with respect to this reference, consistently demonstrating higher performance for extrinsic-dominant (macro) movements than for intrinsic-dependent (micro) gestures. Significance . By integrating neuromuscular modeling priors with deep transfer learning algorithms, this paper proposes a biologically meaningful and statistically robust solution for hand motion intention decoding targeted at amputees in cross-population scenarios.
- 29 May 2026
- Published by IOP Publishing in Journal of Neural Engineering
- Vol. 23 (3) , 036023
Physiology-aware latent shift adaptation for sEMG-based hand kinematics decoding across able-bodied and amputee populations
Objective . Decoding continuous hand kinematics from surface electromyography (sEMG) across able-bodied and amputee populations remains a significant challenge, as amputation induces systematic alterations in neural drive and muscle activation patterns that reshape the generative relationship between motor intentions and observed sEMG signals. Such physiology-driven changes are not explicitly modeled by purely statistical domain alignment approaches, limiting their ability to achieve robust and interpretable cross-population decoding. Approach . We develop an integrated approach, termed physiology-aware latent shift adaptation (PALSA), which combines biophysical modeling with latent-space domain adaptation to characterize cross-population discrepancies at the level of latent neuromuscular representation. PALSA first learns a multidimensional, physiologically interpretable synergy latent space from able-bodied subjects (Ninapro DB2) by integrating a differentiable, motor-unit-based physiology layer. For amputee adaptation (Ninapro DB3), PALSA freezes both the neural encoder and kinematic regressor while introducing a compact, learnable latent shift vector to compensate for population-specific deviations within the learned neuromuscular manifold. Main results . Since Ninapro DB3 does not provide paired kinematics, we evaluate population-level plausibility rather than subject-specific accuracy, using DB9-derived calibrated joint angles as a standardized kinematic reference. PALSA outperforms the classic UDA algorithms CORAL and DANN in transfer learning for motor intention recognition of amputees while effectively characterizing the physiological differences between the two population groups. Across 40 gestures, PALSA achieves a mean cosine similarity of 0.6238 and a mean joint-angle root mean square error of 22.60 ∘ with respect to this reference, consistently demonstrating higher performance for extrinsic-dominant (macro) movements than for intrinsic-dependent (micro) gestures. Significance . By integrating neuromuscular modeling priors with deep transfer learning algorithms, this paper proposes a biologically meaningful and statistically robust solution for hand motion intention decoding targeted at amputees in cross-population scenarios.
- 10 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Neural Networks and Learning Systems
- Vol. PP (2162237X) , 1-12
A Push-Pull Network Mechanism Revealed by Describing Function Analysis for Alzheimer’s Pathological Oscillations
Abnormal theta band (4–8 Hz) cortical oscillations have been identified as a pathological hallmark in Alzheimer’s disease (AD), which can be suppressed through high-frequency electrical stimulation (ES). However, their neural origins and mechanistic underpinnings remain poorly understood. In this work, we develop a mathematical model to characterize AD-related cortical theta oscillations. Through linear equivalent transformation of the physiological model and describing function analysis, we elucidate the network mechanism underlying these pathological oscillations. The results demonstrate that the reduced synaptic inhibition from fast inhibitory interneurons to pyramidal neuron populations drives the cortical network into a pathological low-frequency (theta band) oscillatory state. The oscillation frequency and amplitude can be directly determined from the intersection points between the Nyquist curve of the model’s linear component and the negative inverse curve of the nonlinear element’s describing function. Conversely, the separation between these curves driven by external stimulation visually represents how stimulation can shift the cortical network away from abnormal theta oscillations. The describing function analysis not only provides a mechanistic explanation for aberrant oscillations but also enables precise prediction of effective stimulation parameters for suppressing pathological theta activity. This work offers a novel framework to reveal the “push-pull” network effects of AD, providing an intuitive understanding of cortical stimulation mechanisms and a quantitative approach for optimizing neuromodulation strategies in AD.
- 10 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Neural Networks and Learning Systems
- Vol. PP (2162237X) , 1-12
A Push-Pull Network Mechanism Revealed by Describing Function Analysis for Alzheimer’s Pathological Oscillations
Abnormal theta band (4–8 Hz) cortical oscillations have been identified as a pathological hallmark in Alzheimer’s disease (AD), which can be suppressed through high-frequency electrical stimulation (ES). However, their neural origins and mechanistic underpinnings remain poorly understood. In this work, we develop a mathematical model to characterize AD-related cortical theta oscillations. Through linear equivalent transformation of the physiological model and describing function analysis, we elucidate the network mechanism underlying these pathological oscillations. The results demonstrate that the reduced synaptic inhibition from fast inhibitory interneurons to pyramidal neuron populations drives the cortical network into a pathological low-frequency (theta band) oscillatory state. The oscillation frequency and amplitude can be directly determined from the intersection points between the Nyquist curve of the model’s linear component and the negative inverse curve of the nonlinear element’s describing function. Conversely, the separation between these curves driven by external stimulation visually represents how stimulation can shift the cortical network away from abnormal theta oscillations. The describing function analysis not only provides a mechanistic explanation for aberrant oscillations but also enables precise prediction of effective stimulation parameters for suppressing pathological theta activity. This work offers a novel framework to reveal the “push-pull” network effects of AD, providing an intuitive understanding of cortical stimulation mechanisms and a quantitative approach for optimizing neuromodulation strategies in AD.
- Open Access
Weight-Adaptable Disturbance Observer for Continuous-Control-Set Model Predictive Control of NPC-3L-Fed PMSMs
This paper presents a cascaded control strategy for neutral-point-clamped three-level (NPC-3L) inverter-fed permanent magnet synchronous motors (PMSMs), integrating continuous-control-set model-predictive control (CCS-MPC) with mid-point voltage regulation and an online Lyapunov-stable neural-network (NN) disturbance observer. The outer CCS-MPC loop optimizes voltage vector application for accurate current tracking and harmonic suppression, while the inner loop balances mid-point voltage by adjusting the dwell times of P/N small-voltage vectors (VVs). The NN-based disturbance observer compensates parameter mismatches in real time, reducing steady-state dq-axis current errors. To validate the effectiveness of the proposed strategy, experiments are conducted using a three-phase PMSM fed by three-phase NPC-3L inverters. Experimental results demonstrate substantial improvements in mid-point voltage balance, current quality, and robustness against model uncertainties.
Open AccessWeight-Adaptable Disturbance Observer for Continuous-Control-Set Model Predictive Control of NPC-3L-Fed PMSMs
This paper presents a cascaded control strategy for neutral-point-clamped three-level (NPC-3L) inverter-fed permanent magnet synchronous motors (PMSMs), integrating continuous-control-set model-predictive control (CCS-MPC) with mid-point voltage regulation and an online Lyapunov-stable neural-network (NN) disturbance observer. The outer CCS-MPC loop optimizes voltage vector application for accurate current tracking and harmonic suppression, while the inner loop balances mid-point voltage by adjusting the dwell times of P/N small-voltage vectors (VVs). The NN-based disturbance observer compensates parameter mismatches in real time, reducing steady-state dq-axis current errors. To validate the effectiveness of the proposed strategy, experiments are conducted using a three-phase PMSM fed by three-phase NPC-3L inverters. Experimental results demonstrate substantial improvements in mid-point voltage balance, current quality, and robustness against model uncertainties.
- 17 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Industrial Electronics
- Vol. 73 (2) , 2024-2034
Multiple-Channel Coupling Effect Observation for In-Flight Wireless Charging of Drones
Multiple-drone in-flight wireless charging offers a cost-effective solution for energy replenishment in long-range drone operations. Since multiple transmission channels impact each other, the conventional mitigation of single-channel mutual inductance disturbance cannot be directly utilized for in-flight wireless charging of drones, which has been nearly unexplored in previous studies. To address this challenge, this article presents a multiple-coupling-observed multi-CC output control scheme. The proposed scheme leverages a multiresonant LCC/P topology to extract phase information from parallel branch currents on the transmitter side, enabling real-time monitoring of multiple mutual inductances without requiring load information. Using the monitored mutual inductance, the scheme regulates constant current (CC) outputs for multiple receivers from a single transmitter, eliminating the need for control, detection, or communication on the receiver side, which contributes to drone lightweighting. Experimental results demonstrate that the proposed method achieves mutual inductance monitoring errors below 2.5%, maintains steady-state current variations under 5.5% after mutual inductance changes, and effectively suppresses output current fluctuations during continuous mutual inductance disturbances. These findings confirm that the proposed scheme effectively ensures stable CC charging in multidrone in-flight wireless power transfer (WPT) systems.
- 17 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Industrial Electronics
- Vol. 73 (2) , 2024-2034
Multiple-Channel Coupling Effect Observation for In-Flight Wireless Charging of Drones
Multiple-drone in-flight wireless charging offers a cost-effective solution for energy replenishment in long-range drone operations. Since multiple transmission channels impact each other, the conventional mitigation of single-channel mutual inductance disturbance cannot be directly utilized for in-flight wireless charging of drones, which has been nearly unexplored in previous studies. To address this challenge, this article presents a multiple-coupling-observed multi-CC output control scheme. The proposed scheme leverages a multiresonant LCC/P topology to extract phase information from parallel branch currents on the transmitter side, enabling real-time monitoring of multiple mutual inductances without requiring load information. Using the monitored mutual inductance, the scheme regulates constant current (CC) outputs for multiple receivers from a single transmitter, eliminating the need for control, detection, or communication on the receiver side, which contributes to drone lightweighting. Experimental results demonstrate that the proposed method achieves mutual inductance monitoring errors below 2.5%, maintains steady-state current variations under 5.5% after mutual inductance changes, and effectively suppresses output current fluctuations during continuous mutual inductance disturbances. These findings confirm that the proposed scheme effectively ensures stable CC charging in multidrone in-flight wireless power transfer (WPT) systems.
- 11 September 2025
- Published by Springer Nature in Biological Cybernetics
- Vol. 119 (4-6) , 1-17
Effects of transcranial alternating current stimulation on Spike train correlation in two-compartment model neurons
Correlated spiking has been widely found in large population of neurons and been linked to neural coding. Transcranial alternating current stimulation (tACS) is a promising non-invasive brain stimulation technique that can modulate the spiking activity of neurons. Despite its growing application, the tACS effects on the temporal correlation between spike trains are still not fully understood. In this study, we use a pair of unconnected two-compartment model neurons of the integrate-and-fire (IF) type to simulate the correlated spike trains driven by shared fluctuating dendritic inputs and exposed to weak alternating electric fields. Our results show that the output correlation increases with field intensity, but increases and then decreases with field frequency, displaying thus a frequency resonance. Through varying somatic and dendritic morphologies, we demonstrate that morphological differences between the soma and dendrites fundamentally shape the correlation-frequency resonance, with more pronounced differences yielding stronger resonance effects. Moreover, the anti-phase sinusoidal modulations induced by tACS at the soma and dendrite promote this correlation-frequency resonance, particularly when dendritic fluctuations exhibit a large mean value. We further examine the tACS effects on output correlation in biophysically and morphologically realistic pyramidal model neurons, revealing similar patterns to those observed in the two-compartment models. Our findings provide new insights into how tACS modulates the correlated spike trains and highlight the critical role of morphological differences between the soma and dendrites in determining the frequency-dependent output correlation. These predictions should be taken into consideration when understanding the tACS effects on population correlation and population coding.
- 11 September 2025
- Published by Springer Nature in Biological Cybernetics
- Vol. 119 (4-6) , 1-17
Effects of transcranial alternating current stimulation on Spike train correlation in two-compartment model neurons
Correlated spiking has been widely found in large population of neurons and been linked to neural coding. Transcranial alternating current stimulation (tACS) is a promising non-invasive brain stimulation technique that can modulate the spiking activity of neurons. Despite its growing application, the tACS effects on the temporal correlation between spike trains are still not fully understood. In this study, we use a pair of unconnected two-compartment model neurons of the integrate-and-fire (IF) type to simulate the correlated spike trains driven by shared fluctuating dendritic inputs and exposed to weak alternating electric fields. Our results show that the output correlation increases with field intensity, but increases and then decreases with field frequency, displaying thus a frequency resonance. Through varying somatic and dendritic morphologies, we demonstrate that morphological differences between the soma and dendrites fundamentally shape the correlation-frequency resonance, with more pronounced differences yielding stronger resonance effects. Moreover, the anti-phase sinusoidal modulations induced by tACS at the soma and dendrite promote this correlation-frequency resonance, particularly when dendritic fluctuations exhibit a large mean value. We further examine the tACS effects on output correlation in biophysically and morphologically realistic pyramidal model neurons, revealing similar patterns to those observed in the two-compartment models. Our findings provide new insights into how tACS modulates the correlated spike trains and highlight the critical role of morphological differences between the soma and dendrites in determining the frequency-dependent output correlation. These predictions should be taken into consideration when understanding the tACS effects on population correlation and population coding.
- Top Cited Papers
- 22 August 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Biomedical and Health Informatics
- Vol. 30 (3) , 2694-2707
Spatiospectral Representation and Neural Decoding of Somatic Perception of Acupuncture Stimulations
Characterizing the neural representations underlying somatic perception is crucial for neural decoding of external stimulations. Acupuncture is an important therapeutic method of traditional Chinese medicine and can effectively modulate brain activity for the treatment of neural diseases. In this work, we investigate the neural representations based on the power spectral density (PSD) estimated from electroencephalogram (EEG) across the whole brain with deep learning. Frequency and spatial characteristics of PSD can reliably represent the dynamical brain responses to acupuncture with different manipulations, manifesting enhanced alpha power in parietal lobe. By removing aperiodic components, periodic spatial spectrum shows a higher representation ability of different brain states during acupuncture stimulations, and twiring-rotating (TR) manipulation have a more pronounced modulatory effect than lifting-thrusting (LT) manipulation. Moreover, we further infer the low-dimensional feature-disentangled representations with generative adversarial network (GAN), i.e., w-latents of StyleGAN, which can capture the latent features of periodic spatial spectrum and strike a balance between separability and generalizability. The effectiveness of feature-disentangled representations is evaluated by decoding the acupuncture states, which can achieve a highest accuracy of 95.71% with Transformer classifier. Compared with high-dimensional spatial spectrum, low-dimensional latent features can best characterize different brain states, indicating a precise representation of somatic perception of acupuncture stimulations. Taken together, our results highlight the significant role of spatial spectral representation underlying somatic perception and serve as an important benchmark for the evaluation of acupuncture effect on human brain.
- 22 August 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Biomedical and Health Informatics
- Vol. 30 (3) , 2694-2707
Top Cited PapersSpatiospectral Representation and Neural Decoding of Somatic Perception of Acupuncture Stimulations
Characterizing the neural representations underlying somatic perception is crucial for neural decoding of external stimulations. Acupuncture is an important therapeutic method of traditional Chinese medicine and can effectively modulate brain activity for the treatment of neural diseases. In this work, we investigate the neural representations based on the power spectral density (PSD) estimated from electroencephalogram (EEG) across the whole brain with deep learning. Frequency and spatial characteristics of PSD can reliably represent the dynamical brain responses to acupuncture with different manipulations, manifesting enhanced alpha power in parietal lobe. By removing aperiodic components, periodic spatial spectrum shows a higher representation ability of different brain states during acupuncture stimulations, and twiring-rotating (TR) manipulation have a more pronounced modulatory effect than lifting-thrusting (LT) manipulation. Moreover, we further infer the low-dimensional feature-disentangled representations with generative adversarial network (GAN), i.e., w-latents of StyleGAN, which can capture the latent features of periodic spatial spectrum and strike a balance between separability and generalizability. The effectiveness of feature-disentangled representations is evaluated by decoding the acupuncture states, which can achieve a highest accuracy of 95.71% with Transformer classifier. Compared with high-dimensional spatial spectrum, low-dimensional latent features can best characterize different brain states, indicating a precise representation of somatic perception of acupuncture stimulations. Taken together, our results highlight the significant role of spatial spectral representation underlying somatic perception and serve as an important benchmark for the evaluation of acupuncture effect on human brain.
- Top Cited Papers
- 17 January 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Biomedical and Health Informatics
- Vol. 29 (6) , 4147-4160
Neural Manifold Decoder for Acupuncture Stimulations With Representation Learning: An Acupuncture-Brain Interface
Acupuncture stimulations in somatosensory system can modulate spatiotemporal brain activity and improve cognitive functions of patients with neurological disorders. The correlation between these somatosensory stimulations and dynamical brain responses is still unclear. We proposed a deep learning framework using electroencephalographic activity of stimulated subjects to decode the needling processes of various acupuncture manipulations performed on Zusanli acupoint. Contrastive representation learning integrated with domain adaptation strategy was applied to estimate 3D hand postures and hand joint motion trajectories of acupuncturist with video recordings, by which finite dimensional representations of behavior manifolds for needling operations were inferred. Distinct transition dynamics of behavior manifold were observed for acupuncture with lifting-thrusting and twisting-rotating manipulations. Moreover, latent neural manifolds of acupuncture evoked EEG signals were estimated in low dimensional state space of brain activities with unsupervised manifold learning, which can reliably represent acupuncture stimulations. Furthermore, a nonlinear decoder based on neural networks was designed to transform neural manifolds to behavior manifolds and further predict acupuncture manipulation as well as needling process. Experimental results demonstrated a high performance of the proposed decoding framework for four types of acupuncture manipulations with a precision of 92.42%. The EEG decoder provides an acupuncture-brain interface linking somatosensory stimulations with neural representations, an effective scheme for revealing clinical efficacy of acupuncture treatment.
- 17 January 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Biomedical and Health Informatics
- Vol. 29 (6) , 4147-4160
Top Cited PapersNeural Manifold Decoder for Acupuncture Stimulations With Representation Learning: An Acupuncture-Brain Interface
Acupuncture stimulations in somatosensory system can modulate spatiotemporal brain activity and improve cognitive functions of patients with neurological disorders. The correlation between these somatosensory stimulations and dynamical brain responses is still unclear. We proposed a deep learning framework using electroencephalographic activity of stimulated subjects to decode the needling processes of various acupuncture manipulations performed on Zusanli acupoint. Contrastive representation learning integrated with domain adaptation strategy was applied to estimate 3D hand postures and hand joint motion trajectories of acupuncturist with video recordings, by which finite dimensional representations of behavior manifolds for needling operations were inferred. Distinct transition dynamics of behavior manifold were observed for acupuncture with lifting-thrusting and twisting-rotating manipulations. Moreover, latent neural manifolds of acupuncture evoked EEG signals were estimated in low dimensional state space of brain activities with unsupervised manifold learning, which can reliably represent acupuncture stimulations. Furthermore, a nonlinear decoder based on neural networks was designed to transform neural manifolds to behavior manifolds and further predict acupuncture manipulation as well as needling process. Experimental results demonstrated a high performance of the proposed decoding framework for four types of acupuncture manipulations with a precision of 92.42%. The EEG decoder provides an acupuncture-brain interface linking somatosensory stimulations with neural representations, an effective scheme for revealing clinical efficacy of acupuncture treatment.
- 1 December 2024
- Published by IOP Publishing in Journal of Neural Engineering
- Vol. 21 (6) , 066045
Multi-dimensional hybrid bilinear CNN-LSTM models for epileptic seizure detection and prediction using EEG signals
Objective. Automatic detection and prediction of epilepsy are crucial for improving patient care and quality of life. However, existing methods typically focus on single-dimensional information and often confuse the periodic and aperiodic components in electrophysiological signals. Approach. We propose a novel deep learning (DL) framework that integrates temporal, spatial, and frequency information of EEG signals, in which periodic and aperiodic components are separated in the frequency domain. Specifically, we calculated the periodic and aperiodic components in single channel and the synchronization index of each component between channels. A self-attention mechanism is employed to filter single-channel features by selectively focusing on the most distinguishing features. Then, a hybrid bilinear DL network is utilized to capture the spatiotemporal features by combining a convolutional neural network and a long short-term memory network. Finally, a bilinear pooling layer is employed to extract second-order features based on interactions between these spatiotemporal features. Main results. The model achieves exceptional performance, with a detection accuracy of 98.84% on the CHB-MIT dataset, and a prediction accuracy of 98.44% on CHB-MIT and 97.65% on the Kaggle dataset, both with an false positive rate of 0.02. Significance. This work paves the way for developing real-time, wearable epilepsy prediction devices to improve patient care.
- 1 December 2024
- Published by IOP Publishing in Journal of Neural Engineering
- Vol. 21 (6) , 066045
Multi-dimensional hybrid bilinear CNN-LSTM models for epileptic seizure detection and prediction using EEG signals
Objective. Automatic detection and prediction of epilepsy are crucial for improving patient care and quality of life. However, existing methods typically focus on single-dimensional information and often confuse the periodic and aperiodic components in electrophysiological signals. Approach. We propose a novel deep learning (DL) framework that integrates temporal, spatial, and frequency information of EEG signals, in which periodic and aperiodic components are separated in the frequency domain. Specifically, we calculated the periodic and aperiodic components in single channel and the synchronization index of each component between channels. A self-attention mechanism is employed to filter single-channel features by selectively focusing on the most distinguishing features. Then, a hybrid bilinear DL network is utilized to capture the spatiotemporal features by combining a convolutional neural network and a long short-term memory network. Finally, a bilinear pooling layer is employed to extract second-order features based on interactions between these spatiotemporal features. Main results. The model achieves exceptional performance, with a detection accuracy of 98.84% on the CHB-MIT dataset, and a prediction accuracy of 98.44% on CHB-MIT and 97.65% on the Kaggle dataset, both with an false positive rate of 0.02. Significance. This work paves the way for developing real-time, wearable epilepsy prediction devices to improve patient care.
- Open Access
- 5 November 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Neural Systems and Rehabilitation Engineering
- Vol. 32 (15344320) , 3993-4003
Spatiotemporal Dynamics of Periodic and Aperiodic Brain Activity Under Peripheral Nerve Stimulation With Acupuncture
Brain activities are a mixture of periodic and aperiodic components, manifesting in the power spectral density (PSD) as rhythmic oscillations with spectral peaks and broadband fluctuations. Periodic oscillatory properties of brain response to external stimulation are widely studied, while aperiodic component responses remain unclear. Here, we investigate spatiotemporal dynamics of periodic and aperiodic brain activity under peripheral nerve stimulation with acupuncture by parameterization of power spectra of EEG signals. Regarding periodic brain activity, spectral peak in delta band emerges in frontal and central brain regions indicates a typical phenomenon of neural entrainment, which is formed by coupling periodic brain activity to external rhythmic acupuncture stimulation. In addition, the statistical results show that alpha periodic power is an important indicator for characterizing the modulatory effects of acupuncture on periodic brain activity. As for aperiodic brain activity, broadband EEG spectral trend analysis demonstrates a steeper aperiodic slope in left parietal lobe and a stronger negative correlation with the aperiodic offset under acupuncture compared with resting state, with the absolute value of correlation coefficient increasing from 0.27 to 0.50. Based on the two parameters that can best characterize the acupuncture effect, alpha periodic power and aperiodic slope, the accurate decoding of acupuncture manipulation is realized with AUC = 0.87. This work shows the modulatory effect of peripheral nerve stimulation with acupuncture on the brain activity by characterizing the periodic and aperiodic spectrum features of EEG, providing new insights into the comprehensive understanding of the response processes of human brain to acupuncture stimulation.
- 5 November 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Neural Systems and Rehabilitation Engineering
- Vol. 32 (15344320) , 3993-4003
Open AccessSpatiotemporal Dynamics of Periodic and Aperiodic Brain Activity Under Peripheral Nerve Stimulation With Acupuncture
Brain activities are a mixture of periodic and aperiodic components, manifesting in the power spectral density (PSD) as rhythmic oscillations with spectral peaks and broadband fluctuations. Periodic oscillatory properties of brain response to external stimulation are widely studied, while aperiodic component responses remain unclear. Here, we investigate spatiotemporal dynamics of periodic and aperiodic brain activity under peripheral nerve stimulation with acupuncture by parameterization of power spectra of EEG signals. Regarding periodic brain activity, spectral peak in delta band emerges in frontal and central brain regions indicates a typical phenomenon of neural entrainment, which is formed by coupling periodic brain activity to external rhythmic acupuncture stimulation. In addition, the statistical results show that alpha periodic power is an important indicator for characterizing the modulatory effects of acupuncture on periodic brain activity. As for aperiodic brain activity, broadband EEG spectral trend analysis demonstrates a steeper aperiodic slope in left parietal lobe and a stronger negative correlation with the aperiodic offset under acupuncture compared with resting state, with the absolute value of correlation coefficient increasing from 0.27 to 0.50. Based on the two parameters that can best characterize the acupuncture effect, alpha periodic power and aperiodic slope, the accurate decoding of acupuncture manipulation is realized with AUC = 0.87. This work shows the modulatory effect of peripheral nerve stimulation with acupuncture on the brain activity by characterizing the periodic and aperiodic spectrum features of EEG, providing new insights into the comprehensive understanding of the response processes of human brain to acupuncture stimulation.
- 8 October 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 24 (23) , 39101-39110
Real-Time Detection of Parkinson’s States Based on Wearable Sensor With Embedded Machine Learning
This study presents a novel embedded wearable Parkinson’s states detection system. This system integrates data collection using inertial measurement unit (IMU), feature extraction, machine learning classification algorithm, and Parkinson’s states warning cue into a custom printed circuit board. Fifty Parkinson’s disease (PD) patients experiencing lower limb movement disturbance and fifty age matched healthy controls were enrolled in this study, with two IMU sensors attached to the outside of each ankle and asked to perform three tasks used in clinical to evaluate the severity of PD bradykinesia. In the first phase, the performance of seven state-of-the-art machine learning algorithms to classify lower limb movement symptoms from PD patients with healthy controls was evaluated. In the second phase, the support vector machine with radial basis function classifier demonstrated the best classification performance for the left heel tapping under the knee flexion state task, leading to its selection for microcontroller implantation. The proposed wearable Parkinson’s states detection system includes the microcontroller, IMU sensor, charging module, result indicator module, battery, power switch and reset module. The total area of the printed circuit board is nearly 6.1 cm×3.5 cm, the whole detection device weights 28.2 g, the feature extraction and model classification time needed is about 0.103 seconds, and the power consumption during the active mode is 0.195 W. The embedded machine learning on local microcontroller eliminates the need for data offloading or cloud-based interference, allowing for direct and timely inference of detection result for new subjects.
- 8 October 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 24 (23) , 39101-39110
Real-Time Detection of Parkinson’s States Based on Wearable Sensor With Embedded Machine Learning
This study presents a novel embedded wearable Parkinson’s states detection system. This system integrates data collection using inertial measurement unit (IMU), feature extraction, machine learning classification algorithm, and Parkinson’s states warning cue into a custom printed circuit board. Fifty Parkinson’s disease (PD) patients experiencing lower limb movement disturbance and fifty age matched healthy controls were enrolled in this study, with two IMU sensors attached to the outside of each ankle and asked to perform three tasks used in clinical to evaluate the severity of PD bradykinesia. In the first phase, the performance of seven state-of-the-art machine learning algorithms to classify lower limb movement symptoms from PD patients with healthy controls was evaluated. In the second phase, the support vector machine with radial basis function classifier demonstrated the best classification performance for the left heel tapping under the knee flexion state task, leading to its selection for microcontroller implantation. The proposed wearable Parkinson’s states detection system includes the microcontroller, IMU sensor, charging module, result indicator module, battery, power switch and reset module. The total area of the printed circuit board is nearly 6.1 cm×3.5 cm, the whole detection device weights 28.2 g, the feature extraction and model classification time needed is about 0.103 seconds, and the power consumption during the active mode is 0.195 W. The embedded machine learning on local microcontroller eliminates the need for data offloading or cloud-based interference, allowing for direct and timely inference of detection result for new subjects.