New technology: Efficient real time brain spike detection
A team of researchers headed by those at the University of Southampton have recently demonstrated a new way to encode and compress neuronal spiking activity in real time. Besides its implications of finding efficient links between biology and electronics, the Southampton team hopes that this research will lead to a new type of autonomous neuroprosthetic implants which could provide a more efficient manner of detecting the brain’s electrical signals. Future neuroprosthetics built upon the knowledge gathered by this Southampton team could provide a more effective treatment to help those with neurological disorders or trauma.
One of the technologies that contributes to the novelty of this research is called a memristor, which is an electrical component that can regulate, retain, and remember data about the flow of an electrical current running through it. The memristor is known for being a similar device to a biological synapse, which makes a well-suited device in this scenario. For this experiment, the team utilized a multi-electrode array of memristors called a memristive integrating sensor (MIS) which acted like a synapse to analyze the neuronal spikes.
Figure 1: Illustration depicting a solid-state titanium-oxide memristive device with a 32×32 array of memristors (Gupta et al., 2016)
One of the biggest advantages of using a memrisitive integrating sensor is the energy efficiency associated with the device. In comparison to current best practices, using the MIS used up to 100 times less energy. This is a significant feature of the MIS because energy efficacy has proved to be one the biggest issues related to creating implantable technology that can process neural input in real time. The inherent energy efficiency and the cost effectiveness of this technology is a promising step forward for brain interfacing equipment.
Figure 2: Signal processing with memristive integrating sensors. A CMOS MEA system is used as external frontend. Acquired neural recordings are then made compatible with the memristors operating characteristics. Consequently, changes in resistive states caused by neural spikes are picked up by the memristors and extracted offline (Gupta et al., 2016)
The scope of this research required the collaboration of many of the top minds from seemingly distant fields. One notable contribution came from Leon Chua, the person who theorized the concept of a memristor in 1971. In addition to the input of Chua, the research was also aided by biologists from the University of Padova and the Max Plank Institute in Germany, while combining efforts from engineers belonging to the Nanoelectronics and Nanotechnology Group at the University of Southampton. Furthermore, funding was sponsored by the FP7 Project (the European Union’s fund for Research and Innovation), and the facilities of the Southampton Nanofabrication Centre were relied upon for experimentation.
The research headed by the University of Southampton has provided a brilliant proof of concept for encoding and compressing neuronal spikes, while – most importantly for future applications of this technology – keeping energy costs low. Further improvements on this new type of recording system could prove to be extremely beneficial for technology that relies on neuronal input, and could revolutionize the cost and energy expenditure of these technologies. Even though the field of brain-chip interfacing is one with tremendous complexity and challenges, this research is a solid step forward to making this technology more feasible for future applications. Authors of these findings also noted that this research has implications in smart data compression and pervasive sensing systems.
Gupta, I., Serb, A., Khiat, A., Zeitler, R., Vassanelli, S., Prodromakis, T. Real-time encoding and compression of neuronal spikes by metal-oxide memristors. Nat. Commun. 7:12805. Doi: 10.1038/ncomms12805 (2016).
Kurzweil Accelerating Intelligence. Synapse-like memristor-based electronic device detects brain spikes in real time [Internet]. 2016 [Accessed October 31, 2016]. Available from: http://www.kurzweilai.net/synapse-like-memristor-based-electronic-device-detects-brain-spikes-in-real-time.
Email us at firstname.lastname@example.org or ring us on +44 (0) 1825 890 858 for any inquiries.