An Ensemble of Hyperdimensional Classifiers: Hardware-Friendly Short-Latency Seizure Detection with Automatic iEEG Electrode Selection

• Main Authors: Benatti, S. (Author), Benini, L. (Author), Burrello, A. (Author), Rahimi, A. (Author), Schindler, K. (Author)
• Format: Article
• Language: English
• Published: Institute of Electrical and Electronics Engineers Inc. 2021
• Subjects: accuracy; action potential; algorithm; Algorithms; anxiety; Article; artificial neural network; brain; Brain; classifier; diagnostic imaging; dynamics; electrode; Electrode selection; Electrodes; electroencephalography; Electroencephalography; electrostimulation; ensemble of classifiers; Ensemble of classifiers; epilepsy; epileptic patient; Epileptic seizures; human; Humans; hyperdimensional computing; ieeg; latent period; learning algorithm; line length; local binary patterns; Local binary patterns; Low computational complexity; low-latency seizure detection; machine learning; nerve cell network; Robustness to noise; seizure; Seizures; sensitivity and specificity; signal noise ratio; single photon emission computed tomography; Small memory footprint; speech intelligibility; State-of-the-art algorithms; support vector machine; symbolic dynamics; validation process
• Online Access: View Fulltext in Publisher

 We propose a new algorithm for detecting epileptic seizures. Our algorithm first extracts three features, namely mean amplitude, line length, and local binary patterns that are fed to an ensemble of classifiers using hyperdimensional (HD) computing. These features are embedded into prototype vectors representing ictal (during seizures) and interictal (between seizures) brain states are constructed. These vectors can be computed at different spatial scales ranging from a single electrode up to many electrodes. This flexibility allows our algorithm to identify the electrodes that discriminate best between ictal and interictal brain states. We assess our algorithm on the SWEC-ETHZ iEEG dataset that includes 99 short-Time iEEG seizures recorded with 36 to 100 electrodes from 16 drug-resistant epilepsy patients. Using    fold cross-validation and all electrodes, our algorithm surpasses state-of-The-Art algorithms yielding significantly shorter latency (8.81 s vs. 11.57 s) in seizure onset detection, and higher specificity (97.31% vs. 94.84%) and accuracy (96.85% vs. 95.42%). We can further reduce the latency of our algorithm to 3.74 s by allowing a slightly higher percentage of false alarms (2% specificity loss). Using only the top 10% of the electrodes ranked by our algorithm, we still maintain superior latency, sensitivity, and specificity compared to the other algorithms with all the electrodes. We finally demonstrate the suitability of our algorithm to deployment on low-cost embedded hardware platforms, thanks to its robustness to noise/artifacts affecting the signal, its low computational complexity, and the small memory-footprint on a RISC-V microcontroller. © 2013 IEEE.