End User Community

We are now working to disseminate this work - seeking to partner with any labs interested in using the NGNI-v1 platform for experimental work. If you are interested, please contact: t.constandinou@imperial.ac.uk

NGNI v1 Platform Brochure NGNI v1 Technical Resources (Code, Designs and Docs)
NGNI v1 Platform
NGNI v1 platform showing: USB3.0 SuperSpeed bridge (left); 32-channel spike sorting headstage (centre); and software environment for configuration and real-time visualization (right).

Overview

Spike Sorting is the process of deinterleaving a recorded neural signal in order to determine the firing patterns of individual neurons from the aggregate spike stream.

The NGNI platform is an end-to-end solution for on-node, real-time spike sorting. By using a compact, onboard (template based) spike sorting engine, together with offline training (WaveClus-based), a low power real-time solution is achievable.

Features

  • 32-channel neural recording/streaming
  • On-node, realtime template-based spike sorting
  • Proprietary template building engine (based on WaveClus)
  • Onboard template memory, 18.4kbit (4 templates per channel)
  • Low latency (0.3ms) SPI output
  • Low output data-rate - suitable for wireless communication
  • MicroSD logging and control module for standalone deployment (no PC or tether required).
Standalone configuration of the NGNIv1 platform
NGNI v1 platform for standalone deployment. Shown are: Controller with MicroSD logger (left); standalone configuration for chronic recordings (right).

Applications

  • Signal acquisition systems for electrophysiology
  • Large-scale recording applications (multi-probe, multi-channel)
  • Realtime brain machine interface applications
  • Closed loop low-latency biofeedback

References