This site may earn affiliate commissions from the links on this folio. Terms of utilise.

Neurons

IBM has appear an exciting first challenge for its upcoming TrueNorth calculator chips, so-chosen "neuromorphic" computers physically structured like the brain: Look at a mass of data collected from real homo brains, and make sense of it. That sort of pattern finding is something conventional computers are bad at doing, but IBM hopes with new technology information technology will be able to watch the homo brain in real-fourth dimension — and potentially intervene.

Big, truly unsafe epileptic seizures are a effect of big-calibration storms of electrical activity raging dorsum and forth between the ii hemispheres of the brain. Doctors have had a hard time finding reliable signatures of these storms that can be detected early plenty to allow effective counter-measures, but neural networks are perfectly suited to finding such complex patterns. IBM is feeding its new, neurally inspired TrueNorth chips reams of electro-encepholagram (EEG) readings from epileptics in the promise that they tin detect patterns associated with major seizures. But the real potential lies in application — and the unique advantages of neuromorphic architecture.

IBM's TrueNorth chip, and a few friends, in an SMP setup

IBM's TrueNorth flake, and a few friends, in an SMP setup

IBM in one case networked regular digital computers together into a simulation of a human encephalon. It had the full level of complexity, but ran nigh i,500 times slower than the real thing. The researchers at IBM pointed out the real limiting cistron is not computational power, merely electrical power. Because if we took the regular digital algorithms used to run the wearisome simulation, and but ran it 1,500 times faster, we would cease up using something like 12 gigawatts of ability to practise so. If you don't feel similar ringing your lab in nuclear power plants, you'll have to come up with something significantly amend.

Thankfully, we accept a blueprint: the homo brain. Later on all, a human brain runs the complexity of a human being brain in existent time, and it does so for as little as 20 watts, or enough to ability a small lightbulb. IBM's neuromorphic chips, called TrueNorth, have a portion of that increment in free energy efficiency.

Aria deus ex wearable

Your next article of clothing might have a silicon brain ticking away inside…

What that ways is that the chips best suited to running information-mining code, those that are physically structured similar a neural network, are also the chips with the best applied ability to employ those programs in the real world. Complex, always-on data mining hardware could run down even an enormous battery in just a couple of hours using conventional hardware; with TrueNorth or a similar architecture, it might be possible to stay on our bodies and keep learning all day, or even all calendar week.

The ultimate goal is to use TrueNorth to find seizures in real-time through an implant or wearable solution. A chip, power-efficient enough to last, could monitor and sift EEG readings from a wearer to identify an oncoming seizure. It can then warning you via your smartphone or, hopefully, ane-day contact medical services directly or even administer medication if needed.

Habiliment tech needs either a power revolution, or a ability efficiency revolution. If we're ever going to power real devices with electricity harvested from your clothes, for case, we're going to demand fries that can run on such paltry amounts of ability. Neuromorphic chips could be one such applied science. Monitoring epilepsy is simply the beginning of the applications.