Demonstrating Analog Inference on the BrainScaleS-2 Mobile System
This work addresses the need for low-power, portable inference systems for medical applications like ECG analysis, though it is incremental as it applies an existing neuromorphic ASIC to a new mobile setup.
The researchers tackled the problem of performing efficient edge inference by demonstrating the BrainScaleS-2 mobile system as a compact analog inference engine, achieving a 93.7% detection rate for atrial fibrillation with 14.0% false positives and a classification time of 276us per sample at 5.6W system power.
We present the BrainScaleS-2 mobile system as a compact analog inference engine based on the BrainScaleS-2 ASIC and demonstrate its capabilities at classifying a medical electrocardiogram dataset. The analog network core of the ASIC is utilized to perform the multiply-accumulate operations of a convolutional deep neural network. At a system power consumption of 5.6W, we measure a total energy consumption of 192uJ for the ASIC and achieve a classification time of 276us per electrocardiographic patient sample. Patients with atrial fibrillation are correctly identified with a detection rate of (93.7${\pm}$0.7)% at (14.0${\pm}$1.0)% false positives. The system is directly applicable to edge inference applications due to its small size, power envelope, and flexible I/O capabilities. It has enabled the BrainScaleS-2 ASIC to be operated reliably outside a specialized lab setting. In future applications, the system allows for a combination of conventional machine learning layers with online learning in spiking neural networks on a single neuromorphic platform.