Peripheral nerves provide access to highly processed and segregated neural command signals from the brain that control skeletal muscles. Detecting these signals could provide lifelike, intuitive control of high-degree-of-freedom prosthetic limbs. However, detection of individual fascicle neural activity within a nerve has not yet been accomplished without compromising the nerve. The purpose of this study was to detect fascicular-level neural activity in freely moving animals with nonpenetrating nerve cuff electrodes. Three dogs were implanted with 16-contact flat interface nerve electrodes (FINEs) on the sciatic nerve with total duration of the implant ranging from four to nine months with functional recordings lasting 2.2-7.5 months. The recorded neural activity during normal treadmill walking was used to localize fascicular sources within the nerve and their recovered activity was compared with the ankle movements using binary classification against gait phase. The signal-to-noise ratio (SNR) of the neural signals obtained from the FINE was between 3.65 and 7.59 dB. Postmortem analysis showed that the focal points of the recovered fascicular activity were located within 0.75 + 0.38 mm from the nearest major fascicles. The two movement intents recovered from each fascicle matched the actual observed ankle movements with overall accuracy of 70%-80% and false crosstalk rate of <; 10%. The significant outcome of this study is detecting multiple voluntary fascicular activities with consistent accuracy over time from freely moving animals. It demonstrates the potential of using the FINE in the designs of intuitively controlled advanced prostheses and ultimately improving patient's quality of life.

In this study, we present a novel low-frequency electrical stimulation paradigm for mesial temporal lobe epilepsy (MTLE). The paradigm utilizes the hippocampal commissure as a unique stimulation target to simultaneously influence large portions of the bilateral hippocampal network. When applied to an acute rat model of MTLE, animals that received stimulation exhibited an 88% reduction in the signal power of the bilateral epileptiform activity relative to the control group. In addition, the stimulation entrained the hippocampal network's spontaneous epileptiform activity and disrupted its bilateral synchrony.

The theory of phase resetting can reveal important information about the dynamic behavior of a periodic system when a single brief stimulus is applied to that system at the appropriate time. Phase resetting studies have revealed the existence in some biological systems of a vulnerable stimulus window generating desynchronization and suppression of the activity. The objective of this study was to test the hypothesis that a "singular" stimulus could annihilate this activity. Perfusion with the high-K solution produced synchronous, quasi-periodic population bursts with inter-burst interval of ~0.8-1.5 seconds. A single 0.1 ms duration anodic pulse of programmable delay and magnitude was applied to the somatic layer of the CA3 pyramidal cells. Three types of phase-resetting behavior were observed: (1) Weak resetting with little or no effect on the timing of the subsequent burst, (2) Strong resetting where the applied current pulse delayed the next event by one time period, (3) Singular behavior where the applied pulse partially or completely suppressed the subsequent bursting. The singular stimulus parameter window, however, was very narrow making it difficult to generate the singular behavior reliably. Nevertheless, the results indicate that singularities exist for high potassium neural activity and that a well timed pulse applied with the right amplitude can suppress this activity. This study suggests that phase resetting of a population of neurons is possible for quasi-periodic interictal activity and similar techniques could be applied to the control of epileptic seizures.

Epilepsy is a devastating disease affecting /spl sim/1% of the world's population. Although drug therapy is effective in many patients, 25% are not responsive to anticonvulsants. In addition, up to 50% of those receiving regular mediation suffer major side effects. Surgical resection is another treatment also associated with serious complications. An alternative method to control seizure activity is electrical stimulation. Several electrical stimulation protocols have been developed in animal models of epilepsy that can reduce or completely suppress seizures. Moreover, in over 5000 patients worldwide, electrical stimulation has been used to control seizures. The mechanisms underlying some of the techniques of seizure control are not understood. Some stimulation protocols, such as DC stimulation, rely on the effects of fields and currents on the membrane polarization. Other methods using single pulses, such as phase-resetting, desynchronization, and chaos control rely on the modulation of the dynamic properties of the neuronal networks. Both low- and high-frequency periodic stimulation can suppress seizures not only during stimulation, but also by inducing long-term changes in brain function. The purpose of this review is to present these approaches and to discuss their underlying mechanisms and potential for clinical implementation.

The method of linear associative memory (LAM) has recently been applied in nonlinear parameter estimation. In the method of LAM, a model response, nonlinear with respect to the parameters, is approximated linearly by a matrix, which maps inversely from a response vector to a parameter vector. This matrix is determined from a set of initial training parameter vectors and their response vectors according to a given cost function, and can be updated recursively and adaptively with a pair of newly generated parameter-response vector. The advantage of LAM is that it can yield good estimation of the true parameter from a given observed response even if the initial training parameter vectors are far from the true values. In a previous paper, we have significantly improved the LAM method by introducing a weighted linear associative memory (WLAM) approach for nonlinear parameter estimation. In the WLAM approach, the contribution of each pair of parameter-response vector to the cost function is weighted in a way such that if a response vector is closer to the observed one then its pair plays more important role in the cost function. However, in both LAM and WLAM, the linear association is introduced with zero interceptions, which would not give an exact association even if the model function is linear and so will affect the efficiency of the estimations. In this paper, we construct a theory which introduces a linear association memory with a nonzero interception (WLAMB). The results of our estimation tests on two quite different models, Van der Pol equation and somatic shunt cable model, suggest that WLAMB can still significantly improve on WLAM.