Automatic annotation correction for wearable EEG based epileptic seizure detection

Single-channel seizure detection with clinical confirmation of seizure locations using CHB-MIT dataset

Introduction

Long-term electroencephalography (EEG) monitoring is advised to patients with refractory epilepsy who have a failure of anti-seizure medication and therapy. However, its real-life application is limited mainly due to the use of multiple EEG channels. We proposed a patient-specific deep learning-based single-channel seizure detection approach using the long-term scalp EEG recordings of the Children's Hospital Boston-Massachusetts Institute of Technology (CHB-MIT) dataset, in conjunction with neurologists' confirmation of spatial seizure characteristics of individual patients.

Methods

We constructed 18-, 4-, and single-channel seizure detectors for 13 patients. Neurologists selected a specific channel among four channels, two close to the behind-the-ear and two at the forehead for each patient, after reviewing the patient's distinctive seizure locations with seizure re-annotation.

Results

Our multi- and single-channel detectors achieved an average sensitivity of 97.05-100%, false alarm rate of 0.22-0.40/h, and latency of 2.1-3.4 s for identification of seizures in continuous EEG recordings. The results demonstrated that seizure detection performance of our single-channel approach was comparable to that of our multi-channel ones.

Discussion

We suggest that our single-channel approach in conjunction with clinical designation of the most prominent seizure locations has a high potential for wearable seizure detection on long-term EEG recordings for patients with refractory epilepsy.

Seizure detection with deep neural networks for review of two-channel electroencephalogram

Ultra-long-term electroencephalographic (EEG) registration using minimally invasive low-channel devices is an emerging technology to assess sporadic seizure events. Highly sensitive automatic seizure detection algorithms are needed for semiautomatic evaluation of these prolonged recordings. We describe the design and validation of a deep neural network for two-channel seizure detection. The model is trained using EEG recordings from 590 patients in a publicly available seizure database. These recordings are based on the full 10-20 electrode system and include seizure annotations created by reviews of the full set of EEG channels. Validation was performed using 48 scalp EEG recordings from an independent epilepsy center and consensus seizure annotations from three neurologists. For each patient, a three-electrode subgroup (two channels with a common reference) of the full montage was selected for validation of the two-channel model. Mean sensitivity across patients of 88.8% and false positive rate across patients of 12.9/day were achieved. The proposed training approach is of great practical relevance, because true recordings from low-channel devices are currently available only in small numbers, and the generation of gold standard seizure annotations in two EEG channels is often difficult. The study demonstrates that automatic seizure detection based on two-channel EEG data is feasible and review of ultra-long-term recordings can be made efficient and effective.

Unsupervised domain adaptation for cross-patient seizure classification

Objective. Epileptic seizure is a chronic neurological disease affecting millions of patients. Electroencephalogram (EEG) is the gold standard in epileptic seizure classification. However, its low signal-to-noise ratio, strong non-stationarity, and large individual difference nature make it difficult to directly extend the seizure classification model from one patient to another. This paper considers multi-source unsupervised domain adaptation for cross-patient EEG-based seizure classification, i.e. there are multiple source patients with labeled EEG data, which are used to label the EEG trials of a new patient.Approach. We propose an source domain selection (SDS)-global domain adaptation (GDA)-target agent subdomain adaptation (TASA) approach, which includes SDS to filter out dissimilar source domains, GDA to align the overall distributions of the selected source domains and the target domain, and TASA to identify the most similar source domain to the target domain so that its labels can be utilized.Main results. Experiments on two public seizure datasets demonstrated that SDS-GDA-TASA outperformed 13 existing approaches in unsupervised cross-patient seizure classification.Significance. Our approach could save clinicians plenty of time in labeling EEG data for epilepsy patients, greatly increasing the efficiency of seizure diagnostics.

SeizFt: Interpretable Machine Learning for Seizure Detection Using Wearables

This work presents SeizFt-a novel seizure detection framework that utilizes machine learning to automatically detect seizures using wearable SensorDot EEG data. Inspired by interpretable sleep staging, our novel approach employs a unique combination of data augmentation, meaningful feature extraction, and an ensemble of decision trees to improve resilience to variations in EEG and to increase the capacity to generalize to unseen data. Fourier Transform (FT) Surrogates were utilized to increase sample size and improve the class balance between labeled non-seizure and seizure epochs. To enhance model stability and accuracy, SeizFt utilizes an ensemble of decision trees through the CatBoost classifier to classify each second of EEG recording as seizure or non-seizure. The SeizIt1 dataset was used for training, and the SeizIt2 dataset for validation and testing. Model performance for seizure detection was evaluated using two primary metrics: sensitivity using the any-overlap method (OVLP) and False Alarm (FA) rate using epoch-based scoring (EPOCH). Notably, SeizFt placed first among an array of state-of-the-art seizure detection algorithms as part of the Seizure Detection Grand Challenge at the 2023 International Conference on Acoustics, Speech, and Signal Processing (ICASSP). SeizFt outperformed state-of-the-art black-box models in accurate seizure detection and minimized false alarms, obtaining a total score of 40.15, combining OVLP and EPOCH across two tasks and representing an improvement of ~30% from the next best approach. The interpretability of SeizFt is a key advantage, as it fosters trust and accountability among healthcare professionals. The most predictive seizure detection features extracted from SeizFt were: delta wave, interquartile range, standard deviation, total absolute power, theta wave, the ratio of delta to theta, binned entropy, Hjorth complexity, delta + theta, and Higuchi fractal dimension. In conclusion, the successful application of SeizFt to wearable SensorDot data suggests its potential for real-time, continuous monitoring to improve personalized medicine for epilepsy.

Wearable electroencephalography and multi-modal mental state classification: A systematic literature review

BACKGROUND

Wearable multi-modal time-series classification applications outperform their best uni-modal counterparts and hold great promise. A modality that directly measures electrical correlates from the brain is electroencephalography. Due to varying noise sources, different key brain regions, key frequency bands, and signal characteristics like non-stationarity, techniques for data pre-processing and classification algorithms are task-dependent.

METHOD

Here, a systematic literature review on mental state classification for wearable electroencephalography is presented. Four search terms in different combinations were used for an in-title search. The search was executed on the 29th of June 2022, across Google Scholar, PubMed, IEEEXplore, and ScienceDirect. 76 most relevant publications were set into context as the current state-of-the-art in mental state time-series classification.

RESULTS

Pre-processing techniques, features, and time-series classification models were analyzed. Across publications, a window length of one second was mainly chosen for classification and spectral features were utilized the most. The achieved performance per time-series classification model is analyzed, finding linear discriminant analysis, decision trees, and k-nearest neighbors models outperform support-vector machines by a factor of up to 1.5. A historical analysis depicts future trends while under-reported aspects relevant to practical applications are discussed.

CONCLUSIONS

Five main conclusions are given, covering utilization of available area for electrode placement on the head, most often or scarcely utilized features and time-series classification model architectures, baseline reporting practices, as well as explainability and interpretability of Deep Learning. The importance of a 'test battery' assessing the influence of data pre-processing and multi-modality on time-series classification performance is emphasized.