A PC-based data acquisition system for arrays of 100 extracellular electrodes

Over the past decade, advances in micromanufacturing and interconnection technologies have begun to make electrode arrays with 100 or more active recording sites widely available for applications in neuroscience. While these technologies are providing new opportunities for studying the parallel processing of the nervous system, they are also presenting new challenges for data acquisition and practical experiment operation. Because most extracellular recording applications only utilize the action potential waveforms, or “spikes”, that comprise a small portion of the recordings, the incoming data can be significantly reduced when spike detection and classification are performed in real time. Consequently, many algorithms and PC-based data acquisition systems have been designed to accomplish this. However, when 100 or more channels are present, little time is available for configuring and supervising each channel and a high level of robustness is required from the online classification system. Because of this complexity and the potential for data loss, the authors have designed a 100-channel data acquisition system based on a simple strategy of online detection and storage of spike waveforms for offline classification. This approach preserves the data of interest while requiring less system complexity and user supervision during experiments than complete online spike classification approaches