Neuromorphic deep spiking neural networks for seizure detection

Abstract

The vast majority of studies that process and analyze neural signals are conducted on cloud computing resources, which is often necessary for the demanding requirements of Deep Neural Network (DNN) workloads. However, applications such as epileptic seizure detection stand to benefit from edge devices that can securely analyze sensitive medical data in real-time and personalised manner. In this work, we propose a novel neuromorphic computing approach to seizure detection using a surrogate gradient-based deep Spiking Neural Network (SNN), which consists of a novel Spiking ConvLSTM unit (SPCLU). We have trained, validated, and rigorously tested the proposed SNN model across three publicly accessible datasets, including Boston Children’s Hospital–MIT (CHB-MIT) dataset from the U.S., and the Freiburg (FB) and EPILEPSIAE intracranial EEG (iEEG) datasets from Germany. The average leave-one-out cross-validation AUC score for FB, CHB-MIT, and EPILEPSIAE datasets can reach 92.7%, 89.0%, and 81.1%, respectively, while the computational overhead and energy consumption are significantly reduced when compared to alternative state-of-the-art models, showing the potential for building an accurate hardware-friendly, low-power neuromorphic system. This is the first feasibility study using a deep Spiking Neural Network for seizure detection on several reliable public datasets.

Efficient and generalizable cross-patient epileptic seizure detection through a spiking neural network

“…• LSTMSNN (Yang et al, 2023 ) applies a spiking convLSTM model after using a sliding window of 1 s with 50% overlap to crop the EEG signals with AdamW optimizer under standard training procedure.…”

“…However, these kinds of networks treat EEG signals as image-liked inputs, which may not better utilize biological information. There are also some works using spiking neural networks for epileptic seizure detection due to biological plausibility (Ghosh-Dastidara and Adeli, 2007 ; Ghosh-Dastidar and Adeli, 2009 ) and energy efficiency (Zarrin et al, 2020 ; Shan et al, 2023 ; Yang et al, 2023 ), however, the performance remains lacking compared with ANNs.…”

“…Zarrin et al ( 2020 ) used feedforward spiking convolutional neural network for intracranial electroencephalography (iEEG) seizure detection under patient-specific setting, while Burelo et al ( 2022 ) aimed at detecting epileptic high-frequency oscillations, using a fully connected feedforward spiking neural network under a patient-specific setting. Yang et al ( 2023 ) applied a spiking convLSTM model for epilepsy seizure detection. However, these works are different from our settings, data/tasks, or model architecture.…”

Tiny dLIF: A Dendritic Spiking Neural Network Enabling a Time-Domain Energy-Efficient Seizure Detection System

“…Out-of-Sample Time-Frequency Domain Time Domain Patient ID CNN [27] Conv-LSTM [27] SNN-Conv-LSTM [27] dLIF sample testing, differs from existing methodologies. Importantly, while our model does not demonstrate superiority in all aspects, it offers valuable insights by considering spiking models and temporal input data often overlooked in prior research.…”

Neuromorphic applications in medicine

“…Researchers have used intelligent neuromorphic paradigms to develop real-time detection for epileptic seizures using neuromorphic technology. There is a growing literature that uses typical brainrelated biosignals to separate, identify, and even classify seizure-related markers [106][107][108][109]. We hereby include some representative examples to illustrate the capabilities and potential of these systems.…”