Abstract:Epilepsy is a common clinical disease. Severe epilepsy can be life-threatening in certain unexpected conditions, so it is important to detect seizures instantly with a wearable device and to provide treatment within the golden window. The observation of the electroencephalography (EEG) signal is an imperative method to assist correct epilepsy diagnosis. To detect and classify EEG signals, a convolutional neural network (CNN) is an intuitive and appropriate method that borrows expertise from neurologists. However, the computational cost of training and inference on artificial intelligence (AI)-based solutions make software-only and hardware-only solutions incompetent for real-time monitoring on embedded devices. Hence, this study proposes three key contributions for the challenge, namely, an algorithm framework to provide real-time epilepsy detection, a dedicated coprocessor chip implementing this framework to enable real time epilepsy detection to offload and accelerate detection algorithm, and a custom interface with the coprocessor and reduced instruction set computer-V (RISC-V) instructions to reconfigure the coprocessor and transfer data. The epilepsy detection framework is implemented in 11-layer CNN. The proposed epilepsy detection algorithm performs 97.8% accuracy for floating-point and 93.5% for fixed-point operations through animal experiments with lab rats. The RISC-V CNN coprocessor is fabricated in the TSMC 0.18-μm CMOS process. For each classification, the coprocessor consumes 51 nJ/class. and 0.9 µJ/class. energy on data transfer and inference, respectively. The detection latency on the chip is 0.012 s. With the integration of the hardware coprocessor, AI algorithms can be applied to epilepsy detection for real-time monitoring.

An Effective Chip Implementation of a Real-Time Eight-Channel EEG Signal Processor Based on On-Line Recursive ICA Algorithm

Design of a Hippocampal Cognitive Prosthesis Chip

ASIC Implementation of a Nonlinear Dynamical Model for Hippocampal Prosthesis

Real-Time Adaptive EEG Source Separation Using Online Recursive Independent Component Analysis

An 8-Channel Wearable EEG Acquisition Front-End IC with Integrated Multi-Functions

Fast Automated On-Chip Artefact Removal of EEG for Seizure Detection Based on ICA-R Algorithm and Wavelet Denoising

A 0.99-to-4.38 Uj/class Event-Driven Hybrid Neural Network Processor for Full-Spectrum Neural Signal Analyses

An Area Efficient Low Power ECoG Front-End Chip for Digitalized Subdural Grid

An Efficient EEGNet Processor Design for Portable EEG-Based BCIs

A 0.5-V multi-channel low-noise readout front-end for portable EEG acquisition

RISC-V CNN Coprocessor for Real-Time Epilepsy Detection in Wearable Application

An Optimized EEGNet Processor for Low-Power and Real-Time EEG Classification in Wearable Brain-Computer Interfaces

A low-cost and energy-efficient EEG processor for continuous seizure detection using wavelet transform and AdaBoost

A Low-Noise Low-Power 0.001Hz–1kHz Neural Recording System-on-Chip with Sample-Level Duty-Cycling

An event-driven SoC for neural recording.

An Energy-Efficient ECG Processor with Ultra-Low-Parameter Multi-Stage Neural Network and Optimized Power-of-Two Quantization

A Miniature On-Chip Multi-Functional ECG Signal Processor with 30 &#x00b5;w Ultra-Low Power Consumption

A Compact Online-Learning Spiking Neuromorphic Biosignal Processor

A 746 Nw ECG Processor ASIC Based on Ternary Neural Network

A 1.8 Μw 32 Nv/√hz Current-Reuse Capacitively-Coupled Instrumentation Amplifier for EEG Detection

A 1.5 mm2 4-Channel EEG/BIOZ Acquisition ASIC With 15.2-Bit 3-Step ADC Based on a Signal-Dependent Low-Power Strategy