14 August 2023 in IEEE Transactions on Neural Systems and Rehabilitation Engineering
Sleep stage classification is a fundamental task in diagnosing and monitoring sleep diseases. There are 2 challenges that remain open: (1) Since most methods only rely on input from a single channel, the spatial-temporal relationship of sleep signals has not been fully explored. (2) Lack of sleep data makes models hard to train from scratch. Here, we propose a vision Transformer-based architecture to process multi-channel polysomnogram signals. The method is an end-to-end framework that consists of a spatial encoder, a temporal encoder, and an MLP head classifier. The spatial encoder using a pre-trained Vision Transformer captures spatial information from multiple PSG channels. The temporal encoder utilizing the self-attention mechanism understands transitions between nearby epochs. In addition, we introduce a tailored image generation method to extract features within multi-channel and reshape them for transfer learning. We validate our method on 3 datasets and outperform the state-of-the-art algorithms. Our method fully explores the spatial-temporal relationship among different brain regions and addresses the problem of data insufficiency in clinical environments. Benefiting from reformulating the problem as image classification, the method could be applied to other 1D-signal problems in the future.
27 February 2025 in Movement Disorders
Background: Bursts of exaggerated subthalamic nucleus (STN) beta activity are believed to contribute to clinical impairments in Parkinson's disease (PD). No previous studies have explored burst characteristics and coupling across the sensorimotor cortical‐STN circuit and determined their relationship to dynamic measurements of bradykinesia.Objective: We sought to (1) establish the characteristics of sensorimotor cortical and STN bursts during naturalistic behaviors, (2) determine the predictability of STN bursts from motor cortical recordings, and (3) relate burst features to continuous measurements of bradykinesia using wearable sensors.Methods: We analyzed 1046 h of wirelessly streamed bilateral sensorimotor cortical and STN recordings from 5 PD patients with concurrent measurements of bradykinesia.Results: STN bursts were longer than cortical bursts and had shorter inter‐burst intervals. Long bursts (>200 ms) in both structures displayed temporal overlap (>30%), with cortical bursts tending to lead STN burst onset by 8 ms. Worsening bradykinesia was linked to increased cortical burst rates and durations, whereas STN burst properties had the opposite effect.Conclusion: Cortical beta bursts tend to precede STN beta bursts with short delays and their occurrence relates to worsening bradykinesia in naturalistic settings. © 2025 The Author(s). Movement Disorders published by Wiley Periodicals LLC on behalf of International Parkinson and Movement Disorder Society.
11 August 2021 in IEEE Transactions on Instrumentation and Measurement
A deficiency in proprioception caused by some nerve injuries could result in movement disorders. An effective assessment of proprioception for patients with nerve injuries is essential for their rehabilitation. When visual input is inconsistent with motor input, humans tend to depend heavily on proprioception to perform a task. Based on it, we designed a novel robot-assisted task to identify proprioceptive acuity. We developed a robotic device to provide RM of wrist and employed VR immersing technology to present VM for a user. The RM and VM were programmed randomly to be consistent or not (conflicting) for each trial. The user was required to decide whether what they were perceiving (real motion) and seeing (virtual motion) is consistent or not in a total of 36 trials. We recruited 40 healthy participants (20 males and 20 females) for the assessment validation. The outcome indicated that the novel robot-assisted identification is sensitive enough to distinguish the difference in proprioceptive acuity between males and females. The male participants have better proprioceptive acuity of the wrist than the females ( $p < 0.05$ ), especially in the P/S motion $p < 0.005$ ). In addition, the gender difference was found in the sensitivity of proprioceptive acuity ( $p < 0.05$ ) while no difference was found in the specificity ( $p > 0.05$ ), indicating that female participants had poorer performance on the identification than the males when the RM is consistent with VM. This study indicated that the novel robot-assisted task is effective to identify limb proprioceptive acuity.
20 August 2025 in IEEE Transactions on Neural Systems and Rehabilitation Engineering
Stroke rehabilitation aims to repair neural circuits and dynamics through the remapping of neuronal functions. However, there is currently a gap in understanding the alteration of neural population dynamics-the fundamental computational unit driving functions-under clinical settings. In this study, we introduced a novel method to identify stable low-dimensional structures of neural population dynamics in stroke patients during motor tasks. Using whole-brain EEG recordings from chronic stroke patients performing motor imagery (MI) tasks before and after brain-computer interface (BCI) training, as well as a public EEG dataset of acute stroke patients performing MI tasks, we projected EEG signals from sensor space to voxel space via source localization (eLORETA), simulating neural population activity in regions of interest. By applying dimensionality reduction, we successfully obtained low-dimensional neural manifolds to represent neural population dynamics. Our analysis revealed three key findings: (1) For right-handed patients, task-related low-dimensional dynamics in the related brain regions remain stable across subjects, with their features holding potential as biomarkers for stroke rehabilitation; (2) BCI training promotes global and sustained restoration of neural population dynamics; (3) EEG theta-band oscillations show strong correlation with these dynamics, highlighting their macroscopic nature. This study proposes a new, simple, and powerful tool for comprehension and validation of stroke rehabilitation mechanisms confirming the effectiveness of BCI training in restoring neural dynamics.
4 September 2023 in CNS Neuroscience & Therapeutics
Background: With the development of modern medicine, the Traditional Chinese Medicine (TCM) combined with western medicine began to be produced and applied. Scalp acupuncture (SA) as a Chinese medicine based on neurological theory, has a great advantage compared with TCM in the treatment of nervous system diseases.Method: In this paper, we analyze the physiological and pathological manifestations of sexual dimorphism (SD) to illustrate the necessity of SD treatment. In addition, we review the factors that can affect SD and analyze in physiological structure, function, and pathological neurons. Diseases (pathological basis, pathological manifestations, and incidence) and factors leading to gender differences, which to analyze the possibility of gender differences in SA.Result: Furthermore, we creatively a new insight of SD‐SA and provide the complete SD treatment cases on the basis of the existing SA in different kinds of diseases including stroke, migraine, attention deficit hyperactivity disorder (ADHD), and depression.Conclusion: In summary, we believe that it is feasible to improve the clinical effectiveness of SA, which is able to promote the development of SA, and then provides an actionable evidence for the promotion of precision medicine in the future.