On electromagnetic head digitization in MEG and EEG

Accurate digitization of EEG electrode locations by electromagnetic tracking system: The proposed head rotation method and comparison against optical system

Electroencephalogram (EEG) electrode digitization is crucial for accurate EEG source estimation, and several commercial systems are available for this purpose. The present study aimed to evaluate the digitizing accuracy of electromagnetic and optical systems. Additionally, we introduced a novel rotation method for the electromagnetic system and compared its accuracy with the conventional method of electromagnetic and optical systems. In the conventional method, the operator moves around a stationary participant to digitize, while the participant does not move their head or body. In contrast, in our proposed rotation method with an electromagnetic system, the operator rotates the participant sitting on a swivel chair to digitize in a consistent position. We showed high localization accuracy in both the optical and electromagnetic systems, with an average localization error of less than 3.6 mm. Comparisons of the digitization methods revealed that the electromagnetic system demonstrates superior digitizing accuracy compared to the optical system. Notably, the proposed rotational method is the most accurate among the three methods, which can be attributed to the consistent positioning of EEG electrode digitization within the electromagnetic field. Considering the affordability of the electromagnetic system, our findings provide valuable insights for researchers aiming for precise EEG source estimation.•The study compares the accuracy of electromagnetic and optical systems for EEG electrode digitization, introducing a novel rotation method for improved consistency and precision.•The electromagnetic system, especially with the proposed rotation method, achieves superior digitizing accuracy over the optical system.•Highlighting the cost-effectiveness and precision of the electromagnetic system with the rotation method, this research offers significant insights for achieving precise EEG source estimation.

Exploring Convolutional Neural Network Architectures for EEG Feature Extraction

The main purpose of this paper is to provide information on how to create a convolutional neural network (CNN) for extracting features from EEG signals. Our task was to understand the primary aspects of creating and fine-tuning CNNs for various application scenarios. We considered the characteristics of EEG signals, coupled with an exploration of various signal processing and data preparation techniques. These techniques include noise reduction, filtering, encoding, decoding, and dimension reduction, among others. In addition, we conduct an in-depth analysis of well-known CNN architectures, categorizing them into four distinct groups: standard implementation, recurrent convolutional, decoder architecture, and combined architecture. This paper further offers a comprehensive evaluation of these architectures, covering accuracy metrics, hyperparameters, and an appendix that contains a table outlining the parameters of commonly used CNN architectures for feature extraction from EEG signals.