A magnetoencephalography dataset during three-dimensional reaching movements for brain-computer interfaces.

Studying the motor-control mechanisms of the brain is critical in academia and also has practical implications because techniques such as brain-computer interfaces (BCIs) can be developed based on brain mechanisms. Magnetoencephalography (MEG) signals have the highest spatial resolution (~3 mm) and temporal resolution (~1 ms) among the non-invasive methods. Therefore, the MEG is an excellent modality for investigating brain mechanisms. However, publicly available MEG data remains scarce due to expensive MEG equipment, requiring a magnetically shielded room, and high maintenance costs for the helium gas supply. In this study, we share the 306-channel MEG and 3-axis accelerometer signals acquired during three-dimensional reaching movements. Additionally, we provide analysis results and MATLAB codes for time-frequency analysis, F-value time-frequency analysis, and topography analysis. These shared MEG datasets offer valuable resources for investigating brain activities or evaluating the accuracy of prediction algorithms. To the best of our knowledge, this data is the only publicly available MEG data measured during reaching movements.

F-Value Time-Frequency Analysis: Between-Within Variance Analysis.

Studies on brain mechanisms enable us to treat various brain diseases and develop diverse technologies for daily life. Therefore, an analysis method of neural signals is critical, as it provides the basis for many brain studies. In many cases, researchers want to understand how neural signals change according to different conditions. However, it is challenging to find distinguishing characteristics, and doing so requires complex statistical analysis. In this study, we propose a novel analysis method, FTF (F-value time-frequency) analysis, that applies the F-value of ANOVA to time-frequency analysis. The proposed method shows the statistical differences among conditions in time and frequency. To evaluate the proposed method, electroencephalography (EEG) signals were analyzed using the proposed FTF method. The EEG signals were measured during imagined movement of the left hand, right hand, foot, and tongue. The analysis revealed the important characteristics which were different among different conditions and similar within the same condition. The FTF analysis method will be useful in various fields, as it allows researchers to analyze how frequency characteristics vary according to different conditions.

Accurate Single-Cell Clustering through Ensemble Similarity Learning.

Single-cell sequencing provides novel means to interpret the transcriptomic profiles of individual cells. To obtain in-depth analysis of single-cell sequencing, it requires effective computational methods to accurately predict single-cell clusters because single-cell sequencing techniques only provide the transcriptomic profiles of each cell. Although an accurate estimation of the cell-to-cell similarity is an essential first step to derive reliable single-cell clustering results, it is challenging to obtain the accurate similarity measurement because it highly depends on a selection of genes for similarity evaluations and the optimal set of genes for the accurate similarity estimation is typically unknown. Moreover, due to technical limitations, single-cell sequencing includes a larger number of artificial zeros, and the technical noise makes it difficult to develop effective single-cell clustering algorithms. Here, we describe a novel single-cell clustering algorithm that can accurately predict single-cell clusters in large-scale single-cell sequencing by effectively reducing the zero-inflated noise and accurately estimating the cell-to-cell similarities. First, we construct an ensemble similarity network based on different similarity estimates, and reduce the artificial noise using a random walk with restart framework. Finally, starting from a larger number small size but highly consistent clusters, we iteratively merge a pair of clusters with the maximum similarities until it reaches the predicted number of clusters. Extensive performance evaluation shows that the proposed single-cell clustering algorithm can yield the accurate single-cell clustering results and it can help deciphering the key messages underlying complex biological mechanisms.

Brain mechanisms in motor control during reaching movements: Transition of functional connectivity according to movement states.

Understanding how the brain controls movements is a critical issue in neuroscience. The role of brain changes rapidly according to movement states. To elucidate the motor control mechanism of brain, it is essential to investigate the changes in brain network in motor-related regions according to movement states. Therefore, the objective of this study was to investigate the brain network transitions according to movement states. We measured whole brain magnetoencephalography (MEG) signals and extracted source signals in 24 motor-related areas. Functional connectivity and centralities were calculated according to time flow. Our results showed that brain networks differed between states of motor planning and movement. Connectivities between most motor-related areas were increased in the motor-planning state. In contrast, only connectivities with cerebellum and basal ganglia were increased while those of other motor-related areas were decreased during movement. Our results indicate that most processes involved in motor control are completed before movement. Further, brain developed network related to feedback rather than motor decision during movements. Our findings also suggest that neural signals during motor planning might be more predictive than neural signals during movement. They facilitate accurate prediction of movement for brain-machine interfaces and provide insight into brain mechanisms in motor control.

Macroscopic Neural Oscillation during Skilled Reaching Movements in Humans.

The neural mechanism of skilled movements, such as reaching, has been considered to differ from that of rhythmic movement such as locomotion. It is generally thought that skilled movements are consciously controlled by the brain, while rhythmic movements are usually controlled autonomously by the spinal cord and brain stem. However, several studies in recent decades have suggested that neural networks in the spinal cord may also be involved in the generation of skilled movements. Moreover, a recent study revealed that neural activities in the motor cortex exhibit rhythmic oscillations corresponding to movement frequency during reaching movements as rhythmic movements. However, whether the oscillations are generated in the spinal cord or the cortical circuit in the motor cortex causes the oscillations is unclear. If the spinal cord is involved in the skilled movements, then similar rhythmic oscillations with time delays should be found in macroscopic neural activity. We measured whole-brain MEG signals during reaching. The MEG signals were analyzed using a dynamical analysis method. We found that rhythmic oscillations with time delays occur in all subjects during reaching movements. The results suggest that the corticospinal system is involved in the generation and control of the skilled movements as rhythmic movements.

A study on a robot arm driven by three-dimensional trajectories predicted from non-invasive neural signals.

BACKGROUND: A brain-machine interface (BMI) should be able to help people with disabilities by replacing their lost motor functions. To replace lost functions, robot arms have been developed that are controlled by invasive neural signals. Although invasive neural signals have a high spatial resolution, non-invasive neural signals are valuable because they provide an interface without surgery. Thus, various researchers have developed robot arms driven by non-invasive neural signals. However, robot arm control based on the imagined trajectory of a human hand can be more intuitive for patients. In this study, therefore, an integrated robot arm-gripper system (IRAGS) that is driven by three-dimensional (3D) hand trajectories predicted from non-invasive neural signals was developed and verified. METHODS: The IRAGS was developed by integrating a six-degree of freedom robot arm and adaptive robot gripper. The system was used to perform reaching and grasping motions for verification. The non-invasive neural signals, magnetoencephalography (MEG) and electroencephalography (EEG), were obtained to control the system. The 3D trajectories were predicted by multiple linear regressions. A target sphere was placed at the terminal point of the real trajectories, and the system was commanded to grasp the target at the terminal point of the predicted trajectories. RESULTS: The average correlation coefficient between the predicted and real trajectories in the MEG case was [Formula: see text] ([Formula: see text]). In the EEG case, it was [Formula: see text] ([Formula: see text]). The success rates in grasping the target plastic sphere were 18.75 and 7.50 % with MEG and EEG, respectively. The success rates of touching the target were 52.50 and 58.75 % respectively. CONCLUSIONS: A robot arm driven by 3D trajectories predicted from non-invasive neural signals was implemented, and reaching and grasping motions were performed. In most cases, the robot closely approached the target, but the success rate was not very high because the non-invasive neural signal is less accurate. However the success rate could be sufficiently improved for practical applications by using additional sensors. Robot arm control based on hand trajectories predicted from EEG would allow for portability, and the performance with EEG was comparable to that with MEG.

A study on decoding models for the reconstruction of hand trajectories from the human magnetoencephalography.

Decoding neural signals into control outputs has been a key to the development of brain-computer interfaces (BCIs). While many studies have identified neural correlates of kinematics or applied advanced machine learning algorithms to improve decoding performance, relatively less attention has been paid to optimal design of decoding models. For generating continuous movements from neural activity, design of decoding models should address how to incorporate movement dynamics into models and how to select a model given specific BCI objectives. Considering nonlinear and independent speed characteristics, we propose a hybrid Kalman filter to decode the hand direction and speed independently. We also investigate changes in performance of different decoding models (the linear and Kalman filters) when they predict reaching movements only or predict both reach and rest. Our offline study on human magnetoencephalography (MEG) during point-to-point arm movements shows that the performance of the linear filter or the Kalman filter is affected by including resting states for training and predicting movements. However, the hybrid Kalman filter consistently outperforms others regardless of movement states. The results demonstrate that better design of decoding models is achieved by incorporating movement dynamics into modeling or selecting a model according to decoding objectives.

High-accuracy brain-machine interfaces using feedback information.

Sensory feedback is very important for movement control. However, feedback information has not been directly used to update movement prediction model in the previous BMI studies, although the closed-loop BMI system provides the visual feedback to users. Here, we propose a BMI framework combining image processing as the feedback information with a novel prediction method. The feedback-prediction algorithm (FPA) generates feedback information from the positions of objects and modifies movement prediction according to the information. The FPA predicts a target among objects based on the movement direction predicted from the neural activity. After the target selection, the FPA modifies the predicted direction toward the target and modulates the magnitude of the predicted vector to easily reach the target. The FPA repeats the modification in every prediction time points. To evaluate the improvements of prediction accuracy provided by the feedback, we compared the prediction performances with feedback (FPA) and without feedback. We demonstrated that accuracy of movement prediction can be considerably improved by the FPA combining feedback information. The accuracy of the movement prediction was significantly improved for all subjects (P<0.001) and 32.1% of the mean error was reduced. The BMI performance will be improved by combining feedback information and it will promote the development of a practical BMI system.

Estimation of the velocity and trajectory of three-dimensional reaching movements from non-invasive magnetoencephalography signals.

OBJECTIVE: Studies on the non-invasive brain-machine interface that controls prosthetic devices via movement intentions are at their very early stages. Here, we aimed to estimate three-dimensional arm movements using magnetoencephalography (MEG) signals with high accuracy. APPROACH: Whole-head MEG signals were acquired during three-dimensional reaching movements (center-out paradigm). For movement decoding, we selected 68 MEG channels in motor-related areas, which were band-pass filtered using four subfrequency bands (0.5-8, 9-22, 25-40 and 57-97 Hz). After the filtering, the signals were resampled, and 11 data points preceding the current data point were used as features for estimating velocity. Multiple linear regressions were used to estimate movement velocities. Movement trajectories were calculated by integrating estimated velocities. We evaluated our results by calculating correlation coefficients (r) between real and estimated velocities. MAIN RESULTS: Movement velocities could be estimated from the low-frequency MEG signals (0.5-8 Hz) with significant and considerably high accuracy (p <0.001, mean r > 0.7). We also showed that preceding (60-140 ms) MEG signals are important to estimate current movement velocities and the intervals of brain signals of 200-300 ms are sufficient for movement estimation. SIGNIFICANCE: These results imply that disabled people will be able to control prosthetic devices without surgery in the near future.