Computationally-Efficient Algorithm for Real-Time Absence Seizure Detection in Wearable Electroencephalography

Abstract

Advances in electroencephalography (EEG) equipment now allow monitoring of people with epilepsy in their daily-life environment. The large volumes of data that can be collected from long-term out-of-clinic monitoring require novel algorithms to process the recordings on board of the device to identify and log or transmit only relevant data epochs. Existing seizure-detection algorithms are generally designed for post-processing purposes, so that memory and computing power are rarely considered as constraints. We propose a novel multi-channel EEG signal processing method for automated absence seizure detection which is specifically designed to run on a microcontroller with minimal memory and processing power. It is based on a linear multi-channel filter that is precomputed offline in a data-driven fashion based on the spatial-temporal signature of the seizure and peak interference statistics. At run-time, the algorithm requires only standard linear filtering operations, which are cheap and efficient to compute, in particular on microcontrollers with a multiply-accumulate unit (MAC). For validation, a dataset of eight patients with juvenile absence epilepsy was collected. Patients were equipped with a 20-channel mobile EEG unit and discharged for a day-long recording. The algorithm achieves a median of 0.5 false detections per day at 95% sensitivity. We compare our algorithm with state-of-the-art absence seizure detection algorithms and conclude it performs on par with these at a much lower computational cost.

Absence Seizure Detection Algorithm for Portable EEG Devices

“…In a recent study, Dan et al presented an absence seizure detector based on a linear multichannel filter that was precomputed offline in a data-driven fashion based on the spatial-temporal signature of the seizure and peak interference statistics ( 21 ). The performance of this detector depends on the number of channels (from 3 to 18) used in the calculations.…”

“…The need for automatic and reliable detection of absence seizures has long been recognized (8). Diverse algorithms have been proposed so far to detect seizures in animal models of epilepsy (9)(10)(11)(12) or in human EEG (13)(14)(15)(16)(17)(18)(19)(20)(21). Herein, we present a novel approach to absence seizure detection, which is applicable both to clinical EEGs and recordings made with portable EEG devices with a small number of channels.…”

Epileptic Seizure Detection in EEG via Fusion of Multi-View Attention-Gated U-Net Deep Neural Networks

“…The filter w is optimized in a data-driven fashion to maximize the SNR of o(t) over a training set, where the target signal corresponds to the seizure epochs. A detailed description of this artifact selection procedure is described in [25].…”

Wearable sensors to improve activities in individuals with cerebral palsy

“…However, the expectation from wearable devices is not the same as with non‐wearable ones; rather, they are expected to obtain additional information that could not otherwise be captured. For example, wearable seizure detection devices can measure the number, duration, and timing of seizures continuously, with the potential of adding precision to patient self‐report diaries in order to adapt antiepileptic treatment . These devices may also provide insights into seizure patterns, which could lead to effective seizure prediction and prevention treatment.…”

“…For example, wearable seizure detection devices can measure the number, duration, and timing of seizures continuously, with the potential of adding precision to patient self-report diaries in order to adapt antiepileptic treatment. 1 These devices may also provide insights into seizure patterns, which could lead to effective seizure prediction and prevention treatment.…”