Page 53 - EE Times Europe Magazine | June2020

EE|Times EUROPE — Sensors Insert 53

TRENDS TO WATCH
Neuromorphic Chips Mimic the Human Brain

By Maurizio Di Paolo Emilio

IMAGE: CERYX

he electrical properties of biological cells have long been repair diseased biocircuits and emulate their function with biomedical
studied to understand intracellular dynamics. The difficulty of implants that can adapt to biofeedback.
measuring microscopic parameters that control the dynamics Neuromorphic chips represent a promising technology for
T of ionic currents and the nonlinearity of ionic conductance implanted brain-machine interfaces, with many research projects
have hindered efforts to construct quantitative computational models. now under way. An application example includes solutions to improve
The growing attention paid to implantable bioelectronics for the visual prosthetic systems or deep brain stimulation settings. “Neu-
treatment of chronic diseases is driving technology toward low-power romorphic chips are low-power and compact, and they potentially
solid-state analog devices that accurately mimic biological circuits. can adapt, through on-chip online learning circuits, to the changes
The human brain processes information and stores it instantly that the body undergoes with time,” said Giacomo Indiveri, director
through more than 100 billion neurons. The neurons communicate of the Institute of Neuroinformatics at the University of Zurich and
with each other through more than 100 trillion synapses that are ETH Zurich. “Typically, neuromorphic chips are designed to couple
connected in parallel, allowing the network to perform memory, to the neural circuits they are being interfaced to by using the same
computation, reasoning, and computing simultaneously at low power dynamics and then detecting anomalies in the activity of the neural
(about 20 W). populations they are talking to” — for example, to detect the onset
of a seizure.
Neuromorphic chips work similarly to the human brain, conserving
Many researchers and analysts believe that energy and working only when needed. Many researchers and analysts
neuromorphic chips are likely to be the future believe that these chips are likely to be the future not only for artificial
not only for AI but also for developing low- intelligence but also for developing low-energy cryptographic evalua-
tion systems.
energy cryptographic evaluation systems. “The chips would meet the needs of [patients with] degenerative
neuron disease by substituting diseased biocircuits with synthetic ones,”
said Alain Nogaret, a professor of physics at the University of Bath. Nog-
Neurons determine signals as part of networks that produce collec- aret is part of a research team that worked with cardiologists to show
tive oscillation patterns that are extremely sensitive to the neurons’ that “neuron chips can reverse the effects of heart failure by restoring
properties. One objective of neuromorphic chips is to be able to inte- the function of … respiratory neurons” at the base of the brain.
grate nonlinear electrical characteristics and offer shallow power with In a paper on their findings, the authors state their reluctance to
1
the ability to process a considerable volume of signals in real time. extrapolate the results to other diseases, “as the only extensive trials
Solid-state neurons implemented through a microelectronic layout we have conducted on animal models of disease so far are animal
respond almost identically to biological neurons under stimulation models of heart failure.” But according to Nogaret, “diseases that come
from a wide range of current injection protocols. The optimization to mind” as candidates for the approach include “Alzheimer’s disease
of nonlinear-equation models demonstrates an effective method for and diseases of ion channels in the neuron membrane (channelopa-
programming analog electronic circuits. This approach offers a way to thies). Epilepsy [patients] could also benefit, as some forms of epilepsy

www.eetimes.eu | JUNE 2020