Automated stimulus-response mapping of high-electrode-count neural implants
| Autor: | Richard A. Normann, Gregory A. Clark, Nicholas A. T. Brown, A.M. Wilder, B R Dowden, Scott Hiatt |
|---|---|
| Rok vydání: | 2009 |
| Předmět: |
Engineering
Automatic control Biomedical Engineering Action Potentials Electric Stimulation Therapy Application software computer.software_genre Sensitivity and Specificity Software Internal Medicine Electronic engineering Torque Functional electrical stimulation Animals Peripheral Nerves Muscle Skeletal Impulse response Brain–computer interface business.industry General Neuroscience Rehabilitation Reproducibility of Results Equipment Design Electrodes Implanted Equipment Failure Analysis Brain implant Therapy Computer-Assisted Cats business computer Computer hardware Algorithms |
| Zdroj: | IEEE transactions on neural systems and rehabilitation engineering : a publication of the IEEE Engineering in Medicine and Biology Society. 17(5) |
| ISSN: | 1558-0210 |
| Popis: | Over the past decade, research in the field of functional electrical stimulation (FES) has led to a new generation of high-electrode-count (HEC) devices that offer increasingly selective access to neural populations. Incorporation of these devices into research and clinical applications, however, has been hampered by the lack of hardware and software platforms capable of taking full advantage of them. In this paper, we present the first generation of a closed-loop FES platform built specifically for HEC neural interface devices. The platform was designed to support a wide range of stimulus-response mapping and feedback-based control routines. It includes a central control module, a 1100-channel stimulator, an array of biometric devices, and a 160-channel data recording module. To demonstrate the unique capabilities of this platform, two automated software routines for mapping stimulus-response properties of implanted HEC devices were implemented and tested. The first routine determines stimulation levels that produce perithreshold muscle activity, and the second generates recruitment curves (as measured by peak impulse response). Both routines were tested on 100-electrode Utah slanted electrode arrays (USEAs) implanted in cat hindlimb nerves using joint torque or EMG as muscle output metric. Mean time to map perithreshold stimulus level was 16.4 s for electrodes that evoked responses (n = 3200), and 3.6 s for electrodes that did not evoke responses (n = 1800). Mean time to locate recruitment curve asymptote for an electrode (n = 155) was 9.6 s , and each point in the recruitment curve required 0.87 s. These results demonstrate the utility of our FES platform by showing that it can be used to completely automate a typically time- and effort-intensive procedure associated with using HEC devices. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|