Real-Time Seizure Detection in Microelectrode Array based on Z-Test Spike Detection for Hardware Implementation

Epilepsy, a neurological disorder affecting over 50 million individuals globally, presents a significant clinical challenge due to many cases being resistant to pharmacological treatment. Advancements in neural implants incorporating on-device processing algorithms are transforming epilepsy management by enabling precise seizure detection and control. Such on-device algorithms must optimize computation to reduce power consumption and minimize transmission bandwidth, ensuring long-term viability.

A proposed approach combines a Z-test-based spike detection algorithm with a heuristic decision tree to differentiate between interictal and ictal events. This algorithm operates in real-time to segment local field potentials (LFP) recorded via microelectrode arrays (MEA), aiming for minimal latency and low power requirements in detecting seizure onsets.

When tested on an MEA dataset from brain slices, the algorithm demonstrated robust performance, achieving 98% sensitivity, 93% precision, 92% accuracy, and a 6% false detection rate. It also achieved a seizure detection latency of 0.5 ± 0.6 seconds. This work highlights a low-computation, low-latency approach that is resilient to noise and requires minimal configuration, thereby advancing brain implantable devices for epilepsy treatment.