Implantable Microsystem Records Neural Activity

To both understand neural functions and realize practical neural prostheses, most experts have concentrated on high-density arrays of silicon-based microelectrodes for recording neural activity in the central nervous system using a single channel. Recently, an implantable wireless microsystem that can simultaneously record neural activity on 64 channels was developed by Amir M. Sodagar, Gayatri E. Perlin, Ying Yao, Khalil Najafi, and Kensall D. Wise from the University of Michigan's Engineering Research Center for Wireless Integrated MicroSystems. This system wirelessly transmits spike occurrences to an external interface. In addition, it allows the user to examine the spike waveforms on any channel with 8-b resolution. Signals are amplified by 60 dB with programmable bandwidths from under 100 Hz to 10 kHz.

Using a 2-MHz clock, the microsystem performs channel scanning for spike detection at a rate of 62.5 kSamples/s. It consumes only 14.4 mW power at 1.8 V. The 1.4-x-1.55-cm microsystem comprises a recording front end, neural processor, and bidirectional telemetry chip. Single-unit cortical activity is sensed by two- and three-dimensional electrode arrays comprising 64 sites. In addition, four 16-channel signal-preconditioning chips form a bank of 64 signal-preconditioning blocks. A 64-channel neural processing unit (NPU) performs spike thresholding or waveform digitization.

This system receives power, programming data, and a synchronized clock from the external setup via an inductive link. Its neural processor is based on a modular architecture, which can respond to developing needs in the future. See "An Implantable 64-Channel Wireless Microsystem for Single-Unit Neural Recording," IEEE Journal Of Solid-State Circuits, September 2009, p. 2591.

TAGS: Content
Hide comments


  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.