Article
Neurosciences
Chang Liu, Jia You, Kun Wang, Shanshan Zhang, Yining Huang, Minpeng Xu, Dong Ming
Summary: Motor imagery-based brain-computer interfaces (MI-BCIs) have great potential in neurological rehabilitation, but their application is limited by a controllable instruction set. To overcome this limitation, we proposed a novel movement-intention encoding paradigm based on sequential finger movement. Our offline and online experiments demonstrated the feasibility of this paradigm and highlighted its potential in extending the instruction set of movement intention-based BCIs.
FRONTIERS IN NEUROSCIENCE
(2023)
Article
Multidisciplinary Sciences
Kaushalya Kumarasinghe, Nikola Kasabov, Denise Taylor
Summary: The study introduces a novel Brain-Inspired Spiking Neural Network (BI-SNN) model for incremental learning of spike sequences, successfully predicting muscle activity and kinematics. By mapping spiking activity from input channels into a high-dimensional source-space, the BI-SNN extends a previously proposed computational model, integrating it with a brain-inspired SNN architecture.
SCIENTIFIC REPORTS
(2021)
Article
Multidisciplinary Sciences
Hagar G. Yamin, Guy Gurevitch, Tomer Gazit, Lavi Shpigelman, Itzhak Fried, Yuval Nir, Yoav Benjamini, Talma Hendler
Summary: By analyzing simultaneous recordings of scalp EEG and unit activity, we found that the average firing activity of two medial temporal lobe areas can be estimated from EEG spectral features. Changes in firing activity in both areas and states can be predicted from scalp EEG frequency modulations.
Article
Engineering, Biomedical
Jiarong Wang, Luzheng Bi, Weijie Fei, Cuntai Guan
Summary: This study investigated the neural signatures and decoding methods for single-hand and both-hand movement direction from EEG signals, revealing significant differences in the negative offset maximums of movement-related cortical potentials between single-hand and both-hand movements. Using EEG potentials with SVM classifier, a recognition accuracy of 70.29% ± 10.85% for six-class classification was achieved, demonstrating the feasibility of decoding both hand movement directions.
IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING
(2021)
Article
Computer Science, Information Systems
Yi-Feng Chen, Ruiqi Fu, Junde Wu, Jongbin Song, Rui Ma, Yi-Chuan Jiang, Mingming Zhang
Summary: This study proposed a novel bimanual brain-computer interface (BCI) paradigm to reconstruct the continuous trajectory of both hands during coordinated movements from electroencephalogram (EEG) signals. The results showed that the proposed model outperformed other commonly-used methods in terms of decoding bimanual trajectories. These findings demonstrated the feasibility of simultaneously decoding bimanual trajectory from EEG, indicating the potential of bimanual control for coordinated tasks.
IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS
(2022)
Article
Computer Science, Artificial Intelligence
Mohammadali Ganjali, Alireza Mehridehnavi, Sajed Rakhshani, Abed Khorasani
Summary: Stable decoding of movement parameters is crucial for the success of brain-machine interfaces (BMIs). This study proposes an automatic unsupervised algorithm that addresses the issue of neural activity instability by aligning manifolds and reducing dimensions. The method shows higher decoding performance compared to a state-of-the-art unsupervised BMI stabilizer, offering a promising solution for achieving stable and accurate movement decoding in BMI applications.
INTERNATIONAL JOURNAL OF NEURAL SYSTEMS
(2023)
Article
Neurosciences
Whitney S. Griggs, Sumner L. Norman, Thomas Deffieux, Florian Segura, Bruno-Felix Osmanski, Geeling Chau, Vasileios Christopoulos, Charles Liu, Mickael Tanter, Mikhail G. Shapiro, Richard A. Andersen
Summary: In this study, a closed-loop ultrasound-based brain-machine interface (BMI) was demonstrated in rhesus macaques. The BMI controlled eight independent movement directions and maintained stability across months without retraining.
NATURE NEUROSCIENCE
(2023)
Article
Engineering, Biomedical
Xiaobo Zhou, Renling Zou, Xiayang Huang
Summary: A wavelet neural network (WNN) was proposed in this study to improve the accuracy of decoding movements from motor imagery EEG signals. Experimental optimization of factors such as mother wavelet, wavelet function, number of channels, and imaging segment time led to an accuracy of 86.27 +/- 6.98%. Comparative experiments with other classifiers showed an improvement in accuracy by 15 to 40%, demonstrating the effectiveness of the WNN method in multi-movement classification.
BIOMEDICAL SIGNAL PROCESSING AND CONTROL
(2021)
Article
Engineering, Biomedical
Eva Calvo Merino, A. Faes, M. M. Van Hulle
Summary: The study aimed to identify electrocorticography (ECoG) frequency features that encode different finger movement states. Using linear regression analysis, two superior frequency features were identified, one for distinguishing movement events from rest and the other for encoding movement dynamics. Combining these two features improved the accuracy of predicting finger movement trajectories. The study also showed the influence of adjusting the frequency range for the local motor potential (LMP) on performance.
JOURNAL OF NEURAL ENGINEERING
(2023)
Article
Engineering, Electrical & Electronic
Mohammad Rowshan, Andreas Burg, Emanuele Viterbo
Summary: This research introduces a one-to-one convolutional transform as a pre-coding step before polar transform, resulting in Polarization-Adjusted Convolutional (PAC) codes. Strategies such as adaptive heuristic metric and tree search constraints are proposed to reduce the complexity of sequential decoding for PAC/polar codes. Efficient computation method for intermediate LLRs and partial sums is provided, contributing to improved decoding efficiency and avoiding intermediate information storage or decoding process restart. Performance, complexity, and resource requirements of three decoding algorithms are compared in the study.
IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY
(2021)
Article
Engineering, Biomedical
Mingming Zhang, Junde Wu, Jongbin Song, Ruiqi Fu, Rui Ma, Yi-Chuan Jiang, Yi-Feng Chen
Summary: This study developed a brain-computer interface paradigm to decode coordinated directions of task-oriented bimanual movements from EEG signals. A deep learning model was proposed and achieved high classification accuracies for different movement directions.
IEEE TRANSACTIONS ON NEURAL SYSTEMS AND REHABILITATION ENGINEERING
(2023)
Article
Automation & Control Systems
Armin W. Thomas, Ulman Lindenberger, Wojciech Samek, Klaus-Robert Mueller
Summary: Research has shown that transfer learning improves the performance of deep learning models in datasets with small sample sizes. In this study, the application of transfer learning to cognitive decoding analysis using functional neuroimaging data is systematically evaluated. Pre-trained deep learning models consistently achieve higher decoding accuracies and require less training time and data compared to models trained from scratch. The benefits of pre-training come from the ability to reuse learned features when training with new data. However, challenges arise when interpreting the decoding decisions of pre-trained models, as they may utilize fMRI data in unforeseen and counterintuitive ways.
JOURNAL OF THE FRANKLIN INSTITUTE-ENGINEERING AND APPLIED MATHEMATICS
(2023)
Article
Neurosciences
Jorge A. Salas, Roza G. Bayrak, Yuankai Huo, Catie Chang
Summary: This study introduces a computational technique to reconstruct continuous low-frequency respiration volume fluctuations from fMRI data alone, which is valuable when physiological recordings are unavailable or of insufficient quality. The predicted RV signals from fMRI data can explain temporal variation patterns in resting-state fMRI data, indicating the potential for enriching existing fMRI datasets with respiratory variations information.
Article
Engineering, Biomedical
Baoguo Xu, Yong Wang, Leying Deng, Changcheng Wu, Wenbing Zhang, Huijun Li, Aiguo Song
Summary: This study successfully decoded hand movement types and kinematic information using EEG signals, achieving high accuracy in controlling neuroprosthesis or rehabilitation devices under different movement parameters.
IEEE TRANSACTIONS ON NEURAL SYSTEMS AND REHABILITATION ENGINEERING
(2021)
Article
Neurosciences
Tyler J. Adkins, Taraz G. Lee
Summary: The study found that people perform actions more quickly and accurately when offered a large reward, which may enhance performance by improving neural representations of actions used in motor planning.