Improving Long Term Myoelectric Decoding, Using an Adaptive Classifier with Label Correction.

This study presents a novel adaptive myoelectric decoding algorithm for control of upper limb prosthesis. Myoelectric decoding algorithms are inherently subject to decay in decoding accuracy over time, which is caused by the changes occurring in the muscle signals. The proposed algorithm relies on an unsupervised and on demand update of the training set, and has been designed to adapt to both the slow and fast changes that occur in myoelectric signals. An update in the training data is used to counter the slow changes, whereas an update with label correction addresses the fast changes in the signals. We collected myoelectric data from an able bodied user for over four and a half hours, while the user performed repetitions of eight wrist movements. The major benefit of the proposed algorithm is the lower rate of decay in accuracy; it has a decay rate of 0.2 per hour as opposed to 3.3 for the non adaptive classifier. The results show that, long term decoding accuracy in EMG signals can be maintained over time, improving the performance and reliability of myoelectric prosthesis.

Slip detection and grip adjustment using optical tracking in prosthetic hands.

We have designed a closed loop control system that adjusts the grasping force of a prosthetic hand based on the amount of object slip detected by an optical tracking sensor. The system was designed for the i-Limb (a multi-fingered prosthetic hand from Touch Bionics Inc.) and is comprised of an optical sensor embedded inside a silicone prosthetic glove and a control algorithm. In a proof of concept study to demonstrate the effectiveness of optical tracking in slip sensing, we record slip rate while increasing the weight held in the grasp of the hand and compare two cases: grip adjustment on and grip adjustment off. The average slip rate was found to be 0.314 slips/(s · oz) without grip adjustment and 0.0411 slips/(s · oz) with grip adjustment. This paper discusses the advantages of the optical approach in slip detection and presents the experiment and results utilizing the optical sensor and grip control algorithm.

Radio frequency identification--an innovative solution to guide dexterous prosthetic hands.

Dexterous manipulation of a multi-fingered prosthetic hand requires far more cognitive effort compared to typical 1 degree of freedom hands, which hinders their acceptance clinically. This paper presents a Myoelectrically-Operated Radio Frequency Identification (RFID) Prosthetic Hand (MORPH); an implementation of RFID with a myoelectric prosthetic hand as a means to amplify the controllable degrees of freedom. Contextual information from an object equipped with an RFID tag allows automatic preshaping along with dexterous control in an attempt to reduce the cognitive effort required to operate the terminal device. The myoelectric-RFID hybrid has been demonstrated in a proof-of-concept case study where an amputee was fitted with the device and subjected to activities adapted from the Jebsen Hand Function Test and the Smith Hand Function Evaluation with RFID-tagged and untagged items. Evaluation tests revealed that the MORPH system performed significantly better in 4 of the 8 tasks, and comparable to the control in the remainder.

Ensemble fractional sensitivity: a quantitative approach to neuron selection for decoding motor tasks.

A robust method to help identify the population of neurons used for decoding motor tasks is developed. We use sensitivity analysis to develop a new metric for quantifying the relative contribution of a neuron towards the decoded output, called "fractional sensitivity." Previous model-based approaches for neuron ranking have been shown to largely depend on the collection of training data. We suggest the use of an ensemble of models that are trained on random subsets of trials to rank neurons. For this work, we tested a decoding algorithm on neuronal data recorded from two male rhesus monkeys while they performed a reach to grasp a bar at three orientations (45 degrees, 90 degrees, or 135 degrees). An ensemble approach led to a statistically significant increase of 5% in decoding accuracy and 25% increase in identification accuracy of simulated noisy neurons, when compared to a single model. Furthermore, ranking neurons based on the ensemble fractional sensitivities resulted in decoding accuracies 10%-20% greater than when randomly selecting neurons or ranking based on firing rates alone. By systematically reducing the size of the input space, we determine the optimal number of neurons needed for decoding the motor output. This selection approach has practical benefits for other BMI applications where limited number of electrodes and training datasets are available, but high decoding accuracies are desirable.

Spectral modulation of LFP activity in M1 during dexterous finger movements.

Recent studies have shown that cortical local field potentials (LFP) contain information about planning or executing hand movement. While earlier research has looked at gross motor movements, we investigate the spectral modulation of LFP activity and its dependence on recording location during dexterous motor actions. In this study, we recorded LFP activity from the primary motor cortex of a primate as it performed a fine finger manipulation task involving different switches. The event-related spectral perturbations (ERSP) in four different frequency bands were considered for the analysis; 4 Hz, 6-15 Hz, 17-40 Hz and 75-170 Hz. LFPs recorded from electrodes in the hand area showed the largest change in ERSP for the highest frequency band (75-170 Hz) (p 0.05), while LFPs recorded from electrodes placed more medially in the arm area showed the largest change in ERSP for the lowest frequency band (4 Hz) (p 0.05). Furthermore, the spectral information from the <4 Hz and 75-150 Hz frequency bands was used to successfully decode the three dexterous grasp movements with an average accuracy of up to 81%. Although previous research has shown that multi-unit neuronal activity can be used to decode fine motor movements, these results demonstrate that LFP activity can also be used to decode dexterous motor tasks. This has implications for future neuroprosthetic devices due to the robustness of LFP signals for chronic recording.

Towards closed-loop decoding of dexterous hand movements using a virtual integration environment.

It has been shown that Brain-Computer Interfaces (BCIs) involving closed-loop control of an external device, while receiving visual feedback, allows subjects to adaptively correct errors and improve the accuracy of control. Although closed-loop cortical control of gross arm movements has been demonstrated, closed-loop decoding of more dexterous movements such as individual fingers has not been shown. Neural recordings were obtained from rhesus monkeys in three different experiments involving individuated flexion/extension of each finger, wrist rotation, and dexterous grasps. Separate decoding filters were implemented in Matlab's Simulink environment to independently decode this suite of dexterous movements in real-time. Average real-time decoding accuracies of 80% was achieved for all dexterous tasks with as few as 15 neurons for individual finger flexion/extension, 41 neurons for wrist rotation, and 79 neurons for grasps. In lieu of the availability of advanced multi-fingered prosthetic hands, real-time visual feedback of the decoded output was provided through actuation of a virtual prosthetic hand in a Virtual Integration Environment. This work lays the foundation for future closed-loop experiments with monkeys in the loop and dexterous control of an actual prosthetic limb.

Towards control of dexterous hand manipulations using a silicon Pattern Generator.

This work demonstrates how an in silico Pattern Generator (PG) can be used as a low power control system for rhythmic hand movements in an upper-limb prosthesis. Neural spike patterns, which encode rotation of a cylindrical object, were implemented in a custom Very Large Scale Integration chip. PG control was tested by using the decoded control signals to actuate the fingers of a virtual prosthetic arm. This system provides a framework for prototyping and controlling dexterous hand manipulation tasks in a compact and efficient solution.

Including planning activity in feature space distributes activation over a broader neuron population.

In neuroprosthetic systems, decoding based on a sparse population of task-related neurons is impractical because micro-electrode arrays often drift gradually in the cortex. Since the neuronal population being recorded from is dynamic, it is favorable to have a larger number of neurons containing information relevant to movement decoding and to decrease the relative importance of individual neurons. We have shown that a feature space comprised of neural firing rates from planning as well as movement periods exists in a broader distribution of neurons, as compared to a feature space that is derived from the movement period alone. For this study, spike train data from 297 neurons located in M1 and PM areas was analyzed to validate the hypothesis. The data was from a rhesus monkey performing reach to grasp task with measured wrist supination/pronation. Artificial neural networks were used to model encoding of wrist angle, and a sensitivity analysis was performed to attribute the relative importance of the input neurons. A classifier trained on only the least important neurons, as determined by their relative contribution to the decoded variable, had an average 20% better decoding accuracy when the new method of feature selection was used. This indicates that there is valuable information content within the distributed neuronal population.