Zurich, Switzerland-based aiCTX developed DynapSE neuromorphic processor which can be used to detect anomalies in a single-lead, ECG-like signal.
In a YouTube video, aiCTX demonstrated the DynapSE neuromorphic processor.
“Here we have set up a reservoir neural network consisting of 384 silicon neurons. 64 of these neurons receive the spiking input signal. The activity stimulates a population of 256 stochastically interconnected excitatory neurons, as well as 64 inhibitory neurons that balance the overall reservoir activity. This activity of the neurons is recorded and then classified, in order to decide whether the signal contains an anomaly,” said the narrator of the video.
Read more Withings Move ECG Debuts at CES 2019
First they generated an ECG signal that is normal for the first few heartbeats, and then exhibits a depressed ST segment. After that there are few more normal heartbeats, and then a different anomaly, a so-called “missing Pwave”, where the peak is missing.
Next, they converted the signal to spikes and send it to the chip. And while it is processing this information, its activity was shown on the screen, where each flashing dot corresponds to one spiking neuron. The oscilloscope also shows the complex dynamics of a single neuron.
After the signal has been processed by the chip it is classified, and the resulting output is displayed on the screen. The result is expected it to be zero wherever there is no anomaly, and larger when there are anomalies.
On the other hand, while the ECG signal was normal, the output was zero as expected.
aiCTX is a leading-edge neuromorphic computing company. It provides dedicated mixed-signal neuromorphic processors which overcome the limitations of legacy von Neumann computers to provide an unprecedented combination of ultra-low power consumption and low-latency performance. aiCTX has a unique technological edge and IP portfolio that comes from over 20 years of experience in mixed-signal neural processor design, advanced neural routing architectures, and neural algorithms.
Their full-custom neuromorphic processors will be used in a variety of artificial intelligence edge computing applications that require ultra-low-power and ultra-low-latency features, including autonomous robots, always-on co-processors for mobile and embedded devices, wearable healthcare systems, security, IoT applications, and computing at the network edge. As a “full-stack” neuromorphic engineering company, aiCTX delivers complete solutions, including custom hardware and software configurations to meet specific application needs.