Optimizing channel selection for seizure detection

Abstract

Interpretation of electroencephalogram (EEG) signals can be complicated by obfuscating artifacts. Artifact detection plays an important role in the observation and analysis of EEG signals. Spatial information contained in the placement of the electrodes can be exploited to accurately detect artifacts. However, when fewer electrodes are used, less spatial information is available, making it harder to detect artifacts. In this study, we investigate the performance of a deep learning algorithm, CNN-LSTM, on several channel configurations. Each configuration was designed to minimize the amount of spatial information lost compared to a standard 22-channel EEG. Systems using a reduced number of channels ranging from 8 to 20 achieved sensitivities between 33% and 37% with false alarms in the range of [38, 50] per 24 hours. False alarms increased dramatically (e.g., over 300 per 24 hours) when the number of channels was further reduced. Baseline performance of a system that used all 22 channels was 39% sensitivity with 23 false alarms. Since the 22-channel system was the only system that included referential channels, the rapid increase in the false alarm rate as the number of channels was reduced underscores the importance of retaining referential channels for artifact reduction. This cautionary result is important because one of the biggest differences between various types of EEGs administered is the type of referential channel used.

Towards Interpretable Machine Learning in EEG Analysis

“…Other single channels. T5-O1 channel was already earlier reported as a promising choice [11]. The default setup was tested on the other 20 EEG channels of the TCP montage to confirm the superiority T5 -O1 channel (see Figure 3).…”

EEG-Based Brain-Computer Interfaces (BCIs): A Survey of Recent Studies on Signal Sensing Technologies and Computational Intelligence Approaches and Their Applications

“…One of the main confusing elements for seizure detection is an artefact that appears in several EEG channels that could be misinterpreted due to wave and spike discharges similar to the occurrence of seizure. To optimise channel selection and accuracy for seizure detection with minimal false alarms, Shah et al [184] proposed a CNN-LSTM algorithm to reject artefacts and optimise the framework's performance for seizure detection. It is believed that the implementation of BCIs and real-time EEG signal processing are suitable for standard clinical application and caring for epilepsy patients [185].…”

Towards long term monitoring: Seizure detection with reduced electroencephalogram channels

“…A unique approach by Shah et al focused on domain knowledge to inform the channel selection [48]. They exploited insights on brain hemisphere function, the proximity of a given electrode to other electrodes, electrode position on the scalp, and the region the electrode covered in terms of signal capture.…”