A New Neuromorphic Computing Approach for Epileptic Seizure Prediction

TL;DR

This paper introduces a neuromorphic computing method using spiking neural networks for epileptic seizure prediction, achieving high accuracy with significantly reduced computational complexity suitable for wearable devices.

Contribution

It proposes a novel Spiking-CNN that combines CNN and SNN advantages, using a Gaussian random discrete encoder for efficient EEG sample processing.

Findings

Sensitivity: 95.1%

Specificity: 99.2%

Computational complexity reduced by 98.58%

Abstract

Several high specificity and sensitivity seizure prediction methods with convolutional neural networks (CNNs) are reported. However, CNNs are computationally expensive and power hungry. These inconveniences make CNN-based methods hard to be implemented on wearable devices. Motivated by the energy-efficient spiking neural networks (SNNs), a neuromorphic computing approach for seizure prediction is proposed in this work. This approach uses a designed gaussian random discrete encoder to generate spike sequences from the EEG samples and make predictions in a spiking convolutional neural network (Spiking-CNN) which combines the advantages of CNNs and SNNs. The experimental results show that the sensitivity, specificity and AUC can remain 95.1%, 99.2% and 0.912 respectively while the computation complexity is reduced by 98.58% compared to CNN, indicating that the proposed Spiking-CNN is hardware friendly and of high precision.