Yaoyao Hao

Field-programmable gate array implementation of a probabilistic neural network for motor cortical decoding in rats

A practical brain-machine interface (BMI) requires real-time decoding algorithms to be realised in a portable device rather than a personal computer. In this article, a field-programmable gate array (FPGA) implementation of a probabilistic neural network (PNN) is proposed and developed to decode motor cortical ensemble recordings in rats performing a lever-pressing task for water rewards. A chronic 16-channel microelectrode array was implanted into the primary motor cortex of the rat to record neural activity, and the pressure signal of the lever were recorded simultaneously. To decode the pressure value from neural activity, both Matlab-based and FPGA-based mapping algorithms using a PNN were implemented and evaluated. In the FPGA architecture, training data of the network were stored in random access memory (RAM) blocks and multiply-add operations were realised by on-chip DSP48E slices. In the approximation of the activation function, a Taylor series and a look-up table (LUT) are used to achieve an accurate approximation. The results of FPGA implementation are as accurate as the realisation of Matlab, but the running speed is 37.9 times faster. This novel and feasible method indicates that the performance of current FPGAs is competent for portable BMI applications.

A portable wireless eye movement-controlled Human-Computer Interface for the disabled

Human-Computer Interface (HCI) has become an important area of research and development for the disabled. A portable wireless eye movement-controlled Human-Computer Interface which can be used for the disabled who have motor paralysis and who cannot speak in multiple applications (such as communication aid and smart home applications) is described here. This Interface consists of four major parts: (1) surface electrodes, (2) a two-channel amplifier, (3) a laptop (or a micro-processor), and (4) a ZigBee wireless module. Horizontal and vertical Electro-Oculography (EOG) signals are measured using five surface electrodes placed on the head .The vertical electrodes are placed about 1.0 cm above the right eyebrow and 2.0 cm below the lower lid of the right eye, the horizontal electrodes are placed 2.0 cm lateral to the each side of outer canthi and the last electrode is placed on users forehead to serve as a ground. The two-channel amplifier is comprised of instrumentation amplifiers, band-pass filters and shift circuits. The EOG signals are sampled at the rate of 250Hz and then sent to a laptop or a micro-processor for signal processing which is based on the method of mathematical morphology to recognize the direction of eye movements and voluntary eye blink. The ZigBee wireless communication technology, which is proved to be reliable, low-power and cost-efficient, is used in the portable interface. The subjects can control the wireless device or move a cursor over a screen by using this interface. The delay of this interface is less than 0.5s and errors are very limited. This interface provides a flexible method for the disabled to improve the life quality.

Local-learning-based neuron selection for grasping gesture prediction in motor brain machine interfaces

OBJECTIVE The high-dimensional neural recordings bring computational challenges to movement decoding in motor brain machine interfaces (mBMI), especially for portable applications. However, not all recorded neural activities relate to the execution of a certain movement task. This paper proposes to use a local-learning-based method to perform neuron selection for the gesture prediction in a reaching and grasping task. APPROACH Nonlinear neural activities are decomposed into a set of linear ones in a weighted feature space. A margin is defined to measure the distance between inter-class and intra-class neural patterns. The weights, reflecting the importance of neurons, are obtained by minimizing a margin-based exponential error function. To find the most dominant neurons in the task, 1-norm regularization is introduced to the objective function for sparse weights, where near-zero weights indicate irrelevant neurons. MAIN RESULTS The signals of only 10 neurons out of 70 selected by the proposed method could achieve over 95% of the full recordings decoding accuracy of gesture predictions, no matter which different decoding methods are used (support vector machine and K-nearest neighbor). The temporal activities of the selected neurons show visually distinguishable patterns associated with various hand states. Compared with other algorithms, the proposed method can better eliminate the irrelevant neurons with near-zero weights and provides the important neuron subset with the best decoding performance in statistics. The weights of important neurons converge usually within 10-20 iterations. In addition, we study the temporal and spatial variation of neuron importance along a period of one and a half months in the same task. A high decoding performance can be maintained by updating the neuron subset. SIGNIFICANCE The proposed algorithm effectively ascertains the neuronal importance without assuming any coding model and provides a high performance with different decoding models. It shows better robustness of identifying the important neurons with noisy signals presented. The low demand of computational resources which, reflected by the fast convergence, indicates the feasibility of the method applied in portable BMI systems. The ascertainment of the important neurons helps to inspect neural patterns visually associated with the movement task. The elimination of irrelevant neurons greatly reduces the computational burden of mBMI systems and maintains the performance with better robustness.

Controlling hand-assistive devices: utilizing electrooculography as a substitute for vision

The loss of hand function, due to amputation or neurological injuries, severely debilitates physically and psychosocially. The most evident and critical impairment after upper limb amputation or neurological injury like brachial plexus or spinal cord injury is the loss of prehension, i.e., the ability to perform those movements in which an object is seized and held partially or wholly within the compass of the hand.

Distinct neural patterns enable grasp types decoding in monkey dorsal premotor cortex

OBJECTIVE Recent studies have shown that dorsal premotor cortex (PMd), a cortical area in the dorsomedial grasp pathway, is involved in grasp movements. However, the neural ensemble firing property of PMd during grasp movements and the extent to which it can be used for grasp decoding are still unclear. APPROACH To address these issues, we used multielectrode arrays to record both spike and local field potential (LFP) signals in PMd in macaque monkeys performing reaching and grasping of one of four differently shaped objects. MAIN RESULTS Single and population neuronal activity showed distinct patterns during execution of different grip types. Cluster analysis of neural ensemble signals indicated that the grasp related patterns emerged soon (200-300 ms) after the go cue signal, and faded away during the hold period. The timing and duration of the patterns varied depending on the behaviors of individual monkey. Application of support vector machine model to stable activity patterns revealed classification accuracies of 94% and 89% for each of the two monkeys, indicating a robust, decodable grasp pattern encoded in the PMd. Grasp decoding using LFPs, especially the high-frequency bands, also produced high decoding accuracies. SIGNIFICANCE This study is the first to specify the neuronal population encoding of grasp during the time course of grasp. We demonstrate high grasp decoding performance in PMd. These findings, combined with previous evidence for reach related modulation studies, suggest that PMd may play an important role in generation and maintenance of grasp action and may be a suitable locus for brain-machine interface applications.

Flight control of tethered honeybees using neural electrical stimulation

This paper presents an insect-machine interface for controlling the flight behavior of tethered honeybees. Flight initiation and cessation can be reproducibly generated using electrical pulses between two wire electrodes implanted into the honeybees brain. Experiments are conducted to compare the effect of different stimulus patterns on the honeybees behavior by parameters including success rate, response time and flight duration. The preliminary results enable us to carry out further research works on the control of complex flight behaviors or the neural mechanism of insect flight.

FPGA implementation of hardware processing modules as coprocessors in brain-machine interfaces

Real-time computation, portability and flexibility are crucial for practical brain-machine interface (BMI) applications. In this work, we proposed Hardware Processing Modules (HPMs) as a method for accelerating BMI computation. Two HPMs have been developed. One is the field-programmable gate array (FPGA) implementation of spike sorting based on probabilistic neural network (PNN), and the other is the FPGA implementation of neural ensemble decoding based on Kalman filter (KF). These two modules were configured under the same framework and tested with real data from motor cortex recording in rats performing a lever-pressing task for water rewards. Due to the parallelism feature of FPGA, the computation time was reduced by several dozen times, while the results are almost the same as those from Matlab implementations. Such HPMs provide a high performance coprocessor for neural signal computation.

Predicting Spike Trains from PMd to M1 Using Discrete Time Rescaling Targeted GLM

The computational model for spike train prediction with inputs from other related cerebral cortices is important in revealing the underlying connection among different cortical areas. To evaluate goodness-of-fit of the model, the time rescaling Kolmogorov–Smirnov (KS) statistic is usually applied, of which the calculation is separated from optimization procedure of the model. If the KS statistic could be embedded into objective function of the optimization procedure, precision of the firing probability series generated by the model would be increased directly. This paper presents a linear-nonlinear-Poisson cascade framework for prediction of spike trains, whose objective function is changed from maximizing log-likelihood of the spike trains to minimizing the penalization of discrete time rescaling KS statistic to eliminate the separation between optimization and evaluation of the model. We apply our model on the task of predicting firing probability of neurons from primary motor cortex with spike trains from dorsal premotor cortex as input, which are two cerebral cortices associated with movements planning and executing. The experimental results show that by introducing the goodness-of-fit metric into the objective function, results of the model will gain a significant improvement, which outperforms the state of the art.

Decoding grasp types with high frequency of local field potentials from primate primary dorsal premotor cortex

Recently, local field potentials (LFPs) have been successfully used to extract information of arm and hand movement in some brain-machine interfaces (BMIs) studies, which suggested that LFPs would improve the performance of BMI applications because of its long-term stability. However, the performance of LFPs in different frequency bands has not been investigated in decoding hand grasp types. Here, the LFPs from the monkeys dorsal premotor cortices were collected by microelectrode array when monkey was performing grip-specific grasp task. A K-nearest neighbor classifier performed on the power spectrum of LFPs was used to decode grasping movements. The decoding powers of LFPs in different frequency bands, channels and trials used for training were also studied. The results show that the broad high frequency band (200-400Hz) LFPs achieved the best performance with classification accuracy reaching over 0.9. It infers that high frequency band LFPs in PMd cortex could be a promising source of control signals in developing functional BMIs for hand grasping.

Neural Ensemble Decoding of Rat's Motor Cortex by Kalman Filter and Optimal Linear Estimation

In this paper, rats were trained to press a lever over a threshold to get water as rewards, and neural ensemble activities in primary motor cortex (MI) and pressure signal of the lever were recorded synchronously. Meanwhile, two algorithms, Kalman filter (KF) and Optimal Linear Estimation (OLE), were used to decode neural ensemble activities around the pressing events. After training, the pressure values were extracted from the neural signals by those decoders, and the performances were compared by Correlation Coefficient (CC) and Mean Square Error (MSE) with pressure values recorded from the lever. KF was better than OLE when the bin size 0.7, but it was not applicable in this experiment when the bin size >100ms for excessively smoothed. And OLE performs better when bin size >100ms for diminishing the randomicity of neural firing, which might imply that KF is more applicable in real-time system.