Implantable multichannel wireless electromyography for prosthesis control.

We have developed a prototype implantable device for recording multiple independent channels of EMG and sending those signals wirelessly to an external receiver. This design records multichannel EMG signals for providing simultaneous multi-axis control of prosthetic limbs. This proof-of-concept study demonstrates benchtop performance of the bioamplifier in dry and soaked in saline configurations, as well as system performance in a short-term in vivo study in six dogs. The amplifier was shown to have an input-referred noise of 2.2 µV(RMS), a common mode rejection ratio greater than 55 dB, and neighboring channel isolation averaging 66 dB. The prototype devices were constructed of an amplifier ASIC along with discrete components for wireless function. These devices were coated in silicone and implanted for at least one week in each dog. EMG recorded from each animal as it walked down a hallway had very low noise and swing/stance phases of gait were clearly shown. This study demonstrates this device design can be used to amplify and transmit muscle signals.

Field deployable EEG monitor for nerve agent casualties.

Early recognition and aggressive management of seizure activity is important in the treatment of patients with nerve agent exposure. However, these patients can experience non-convulsive seizures that are difficult to identify without EEG monitoring. In this paper, we discuss the development and testing of a low-cost, field-deployable device that records and displays patient EEG trends over time. The device is optimized for early levels of care for military and mass casualty patients until they can be relocated to medical facilities with more comprehensive monitoring. The device also records pulse oximetry and acceleration information, and patient data are available for later analysis and improvement of treatment protocols.

Simultaneous, proportional, multi-axis prosthesis control using multichannel surface EMG.

Most upper limb prosthesis controllers only allow the individual selection and control of single joints of the limb. The main limiting factor for simultaneous multi-joint control is usually the availability of reliable independent control signals that can intuitively be used. In this paper, a novel method is presented for extraction of individual muscle source signals from surface EMG array recordings, based on EMG energy orthonormalization along principle movement vectors. In cases where independently-controllable muscles are present in residual limbs, this method can be used to provide simultaneous, multi-axis, proportional control of prosthetic systems. Initial results are presented for simultaneous control of wrist rotation, wrist flexion/extension, and grip open/close for two intact subjects under both isometric and non-isometric conditions and for one subject with transradial amputation.

Discrete stimulus estimation from neural responses in the turtle retina.

In this paper, we investigate the decoding of flashed, full-field visual stimuli while recording from a population of retinal ganglion cells. We present a direct statistical method for determining the likelihood that a response was evoked by a particular stimulus, and use this method to estimate stimuli based on microelectrode array recordings in the turtle retina. This method uses the well-known time-varying Poisson model of neural firing, along with extensions to accommodate neural refractory periods. Unlike other approaches commonly used for Poisson processes, the specific formulation presented here is bin free and requires few user-specified parameters. Statistical dependency issues and the effects of stationarity on the estimation method are also discussed.