Publications by authors named "Guanglin Li"
In situ structural-functional synchronous dissection of dynamic neuromuscular system via an integrated multimodal wearable patch.
Hang Zhao Weicen Chen Yuanheng Li Hailiang Wang Hanfei Li Tengfei Li Yan Liu Qingsong Li Xijian Liu Fei Li Xiangxin Li Lin Wang Guanglin Li Zhiyuan Liu
Sci Adv
January 2025
Neuromuscular abnormality is the leading cause of disability in adults. Understanding the complex interplay between muscle structure and function is crucial for effective treatment and rehabilitation. However, the substantial deformation of muscles during movement (up to 40%) poses challenges for accurate assessment.
View Article and Find Full Text PDFArticle Synopsis
- The study addresses the challenges neurologists face in analyzing multichannel EEG recordings for epileptic activities, proposing a computer-aided diagnosis system to minimize manual inspection.
- Researchers introduce a novel short-time Fourier transform (STFT) algorithm utilizing taper functions to enhance spectrogram resolution, which improves the interpretability of EEG signals essential for seizure detection.
- The developed Dilated Convolutional Squeeze and Excitation Networks (DCSENets) achieve a high average accuracy in patient-independent seizure classification while incorporating a visual explainer that enhances model interpretability, aiding neurologists in understanding diagnosis decisions.
Neural machine interface technology is a pioneering approach that aims to address the complex challenges of neurological dysfunctions and disabilities resulting from conditions such as congenital disorders, traumatic injuries, and neurological diseases. Neural machine interface technology establishes direct connections with the brain or peripheral nervous system to restore impaired motor, sensory, and cognitive functions, significantly improving patients' quality of life. This review analyzes the chronological development and integration of various neural machine interface technologies, including regenerative peripheral nerve interfaces, targeted muscle and sensory reinnervation, agonist-antagonist myoneural interfaces, and brain-machine interfaces.
View Article and Find Full Text PDFRice panicle architecture is a pivotal trait that strongly contributes to grain yield. Small peptide ligands from the OsEPF/EPFL family synergistically control panicle architecture by recognition of the OsER1 receptor and subsequent activation of the OsMKKK10-OsMKK4-OsMPK6 cascade, indicating that specific ligand-receptor pairs orchestrate rice panicle development. However, how small homologous peptides fine-tune organ morphogenesis by targeting a common receptor remains elusive.
View Article and Find Full Text PDFArticle Synopsis
- Identifying the seizure occurrence period (SOP) in EEG recordings is critical for accurate seizure diagnosis, but current systems mainly differentiate between ictal and interictal states, limiting their clinical use due to reliance on labeled data.
- This study presents an unsupervised learning framework called 1D-CasCAE that analyzes EEG segments, focusing on learning patterns from non-seizure segments to improve detection of seizure (ictal) segments without needing labeled data.
- Experimental results show that the 1D-CasCAE outperforms existing methods in detecting anomalies in EEG recordings, achieving high scores in sensitivity, specificity, and precision, making it a promising tool for clinical applications to reduce the need for manual EEG analysis by neurologists.*
Article Synopsis
- * The AMI surgery involves grafting the soleus muscle and reconnecting nerves to enhance motor function and proprioception in rats.
- * Results showed that AMI-treated rats had better neurological repair and improved walking stability compared to control rats, suggesting that AMI could help regain proprioceptive function in prosthetic limbs.
Article Synopsis
- Terpenoids are key components of floral scent in the Chrysanthemum genus, showing notable differences between wild species and domesticated cultivars.
- A study compared two wild ancestors and two cultivated varieties of C. morifolium, revealing that wild relatives produce more terpenoids than cultivars, with distinct gene expressions influencing this variation.
- The research identified specific terpene synthase genes linked to floral scent compounds, demonstrating that the biochemical functions of these genes differ between wild and cultivated plants, shedding light on the effects of domestication on scent production.
Article Synopsis
- The study looked at a surgery called the agonist-antagonist myoneural interface (AMI) that helps people with prosthetic limbs feel sensations like they have their real limbs back.
- After the surgery, the muscles can get weak because they’re not used much, but electrical muscle stimulation (EMS) can help keep the muscles strong.
- The experiment on rats showed that those treated with EMS had better muscle performance and sensations, making the surgery even more effective than those who didn’t get EMS.
NIS-SLAM: Neural Implicit Semantic RGB-D SLAM for 3D Consistent Scene Understanding.
IEEE Trans Vis Comput Graph
November 2024
Article Synopsis
- Recent advancements in neural implicit representations have highlighted gaps in scene understanding within SLAM systems, prompting the introduction of NIS-SLAM, a new RGB-D SLAM system that utilizes a 2D segmentation network for improved semantic representation.
- NIS-SLAM combines high-frequency multi-resolution features with low-frequency positional encoding to enhance surface reconstruction and consistency across views, addressing inconsistencies from 2D segmentation.
- The system demonstrates robust camera tracking through a confidence-based pixel sampling method and has shown competitive performance in experiments, proving its applicability in both SLAM and augmented reality projects.
[Lower limb joint contact forces and ground reaction forces analysis based on Azure Kinect motion capture].
Sheng Wu Yi Xue Gong Cheng Xue Za Zhi
August 2024
Article Synopsis
- Traditional gait analysis systems are complex, expensive, and not portable, prompting a study to use Azure Kinect for a simpler solution.
- The study involved analyzing skeletal data from 10 participants to calculate lower limb joint angles, forces, and ground reaction forces using a musculoskeletal model.
- Results showed strong correlations with traditional systems, though some inaccuracies persisted in joint angle predictions, indicating potential for improved analysis with Azure Kinect.
Assessment of TENS-Evoked Tactile Sensations for Transradial Amputees via EEG Investigation.
IEEE Trans Neural Syst Rehabil Eng
September 2024
Article Synopsis
- - Current prostheses can restore motor function for limb amputees but often lack natural sensory feedback; Transcutaneous Electrical Nerve Stimulation (TENS) shows promise in providing non-invasive sensation evocation.
- - The study used multi-channel TENS with various stimulus patterns to evoke sensations in both amputees and non-disabled individuals, while collecting EEG data to objectively assess the responses.
- - Results indicated that customized TENS successfully evoked sensations, with EEG analysis revealing relevant sensory-processing components and increased interactions in the brain's sensorimotor areas, suggesting potential for improved sensory feedback in prosthetics.
Article Synopsis
- * The system consists of two parts: the leader side, which uses a Leap Motion Controller to track the motions of a healthy hand, and the follower side, which drives the exoskeleton to assist the user's hand based on this input.
- * Experiments show that the prototype exoskeleton delivers an average force of 16.5 N during hand flexing and has an 82.03% success rate in grasping tasks, demonstrating its potential for enhancing rehabilitation outcomes across different stroke recovery stages.
ABR-Attention: An Attention-Based Model for Precisely Localizing Auditory Brainstem Response.
IEEE Trans Neural Syst Rehabil Eng
September 2024
Article Synopsis
- Auditory Brainstem Response (ABR) is a method that measures how the brainstem responds to sound, particularly useful for diagnosing hearing loss and diseases by analyzing specific waves, especially Wave V latency.
- Traditional methods for analyzing these waves are labor-intensive and time-consuming for clinicians, prompting the development of automated techniques, although these come with their own limitations.
- The study presents a new deep learning model called ABR-Attention that effectively extracts Wave V latency, achieving a high accuracy of 96.76% while reducing clinician workload, offering a promising tool for more efficient auditory diagnostics.
Article Synopsis
- - This study uses high-density surface electromyography (HD-sEMG) to examine neuromuscular activity patterns in patients with unilateral upper limb motor dysfunction compared to healthy controls while performing various hand and wrist movements.
- - It identifies key spatial and temporal indicators, such as symmetry in muscle synergies and coordination of upper limbs, revealing significant changes in these indicators in patients with motor dysfunction.
- - The results highlight specific alterations in muscle synergy patterns, offering insights that can improve assessments and inform rehabilitation strategies for individuals with neurological disorders.
Article Synopsis
- A study proposes a cost-effective method using dual smartphone videos to estimate lower limb mechanics, aiming to replace the expensive motion capture systems often used in biomechanics research.
- Twelve participants were tested while walking and running, gathering data from smartphone videos, force plates, and markers to analyze joint angles and forces.
- The results showed strong correlations between smartphone-derived estimates and traditional methods, indicating this approach could enhance gait analysis in clinical settings while saving time and resources.*
Article Synopsis
- Hydrogels are advantageous for creating flexible sensors due to their soft texture and adjustable chemical properties, although challenges remain in achieving the right combination of mechanical and conductive traits for practical use.
- A new type of hydrogel sensor, named PAM/BA-Ag@PDA, is developed by polymerizing acrylamide with dynamic crosslinked silver-modified polydopamine nanoparticles, enhancing its strength, conductivity, and self-healing capabilities.
- This hydrogel functions effectively as a strain sensor, showing good sensitivity, rapid response time, and stability while being able to adhere to skin to monitor various movements, making it suitable for wearable healthcare applications.*
Microneedle Array Electrodes Fabricated With 3D Printing Technology for High-Quality Electrophysiological Acquisition.
IEEE Trans Neural Syst Rehabil Eng
July 2024
Article Synopsis
- * Microneedle array electrodes (MAEs) offer a potential improvement, with the study utilizing projection stereolithography (P μ SL) 3D printing for precise and cost-effective MAE fabrication.
- * The results showed that the MAEs had favorable electrode-skin interface impedance and performed well for EMG and ECG measurements without causing skin issues, indicating their promising biomedical applications for minimally invasive procedures.
Towards Wearable and Portable Spine Motion Analysis Through Dynamic Optimization of Smartphone Videos and IMU Data.
IEEE J Biomed Health Inform
October 2024
Article Synopsis
- Monitoring spine kinematics is important for health evaluations, but existing methods face challenges due to the small size of vertebrae and their complex motion.
- The study introduces a new method combining smartphone video and Inertia Measurement Units (IMUs) to track spine motion more accurately, validated by testing on healthy participants.
- Results showed that this integrated approach provides better accuracy for spine motion analysis compared to traditional single-method techniques, making it suitable for real-world use in spinal health assessments.
Article Synopsis
- * They are important for plant responses to environmental stress and adaptation.
- * This review discusses the genomics of lncRNAs, their molecular functions and mechanisms in plants, as well as the challenges in studying them and advancements in methodologies for analysis.
The Hedgehog (Hh) signaling pathway is involved in T cell differentiation and development and plays a major regulatory part in different stages of T cell development. A previous study by us suggested that prenatal exposure to staphylococcal enterotoxin B (SEB) changed the percentages of T cell subpopulation in the offspring thymus. However, it is unclear whether prenatal SEB exposure impacts the Hh signaling pathway in thymic T cells.
View Article and Find Full Text PDFArticle Synopsis
- - The study addresses the challenges of interpreting hand motion intentions using surface electromyography (sEMG), emphasizing the need for continuous kinematics estimation that aligns more naturally with real-life movements rather than just classifying discrete actions.
- - It proposes a novel continuous Kalman estimation method that uses sEMG and joint angles to accurately infer the motion of fingers, validating its effectiveness with a significant correlation coefficient of 0.73 from a large dataset.
- - The approach demonstrates impressive computational efficiency, achieving an average processing time of under 0.01 seconds while training on over 45,000 data segments, highlighting its potential for practical applications in finger motion estimation.
Article Synopsis
- Epilepsy is a neurological disorder marked by dangerous seizures, which are monitored using EEG signals; accurate detection relies on recognizing key EEG features.
- This study introduces an Advanced Multi-View Deep Feature Learning (AMV-DFL) framework that uses machine learning to enhance EEG feature detection through traditional and deep learning methods.
- Experimental results show that AMV-DFL outperforms other existing models by improving classification accuracy, aiding clinicians in identifying crucial EEG features and potentially discovering new biomarkers for epilepsy management.
Article Synopsis
- Histone deacetylase Sir2 is crucial for the pathogenicity of a common fungal pathogen, as its absence leads to lower survival rates and tissue damage in mouse models of infection.
- * Deletion of Sir2 impairs the fungus's ability to adhere to host cells and increases its visibility to the immune system, primarily due to changes in cell surface properties rather than gene expression.
- * Sir2 also plays a role in maintaining metabolic activity in low-oxygen environments, aiding the fungus's colonization and overall adaptability, making it a potential target for antifungal drug development.
Article Synopsis
- Green fruit detection is crucial for orchard yield estimation and resource management, but accurate methods are lacking due to challenges like color similarity with backgrounds and limited datasets.
- The study introduces an improved YOLOv5 model (YOLOv5-AT), integrating data augmentation and innovative blocks to enhance detection accuracy of green fruits like green oranges, tomatoes, and persimmons.
- Results showed the YOLOv5-AT achieved detection accuracies of 84.6%, 98.0%, and 85.1% respectively, and demonstrated effective performance even with only 50% of the dataset, highlighting its potential to boost agricultural efficiency.
Article Synopsis
- Fermentation significantly influences the quality of black tea, but traditional monitoring methods are too slow and rely heavily on the subjective judgment of tea makers.
- The study utilized advanced techniques like proton transfer reaction mass spectrometry (PTR-MS) and Fourier transform infrared spectroscopy (FTIR) to analyze volatile organic compounds (VOCs) during fermentation stages, achieving high prediction accuracy with extreme learning machine (ELM) models.
- The results suggest practical applications of these technologies in black tea production, highlighting their individual strengths for better monitoring and quality control.