Real-time P300-based BCI in mechatronic control by using a multi-dimensional approach

• Published in: IET software Vol. 12; no. 5; pp. 418 - 424
• Main Authors: De Venuto, Daniela, Annese, Valerio F, Mezzina, Giovanni
• Format: Journal Article
• Language: English
• Published: The Institution of Engineering and Technology 01.10.2018
• Subjects: 32-channel electroencephalography headset; acquisition unit; analyses all the binary discrimination scenarios; BCI architecture; bioelectric potentials; biomedical electronics; biomedical telemetry; brain-computer interfaces; brain–computer interface; classification accuracy; classification stage; complete classification chain; custom algorithm; data gathering; dimensionality reduction step; electroencephalography; feature extraction; functional approach; handicapped aids; learning (artificial intelligence); machine learning algorithm; mechatronic control; mechatronic device driving; mechatronics; medical signal detection; medical signal processing; ML stage; multiclass classification problem; multidimensional approach; navigation unit; neurophysiology; P300 reference features; parietal-cortex area; pattern classification; physical control; processing unit; real-time classification; real-time control; signal classification; spatio-temporal characterisation; Special Issue: Software Architecture for the Web of Things (SAWoT); time 8.16 ms; time-domain features extraction; tuned-residue iteration decomposition; user intention interpretation; handicapped aids; analyses all the binary discrimination scenarios; brain-computer interfaces; ML stage; real-time control; bioelectric potentials; acquisition unit; classification accuracy; classification stage; 32-channel electroencephalography headset; feature extraction; mechatronics; data gathering; physical control; biomedical electronics; navigation unit; learning (artificial intelligence); P300 reference features; user intention interpretation; real-time classification; spatio-temporal characterisation; pattern classification; electroencephalography; multidimensional approach; tuned-residue iteration decomposition; complete classification chain; time 8.16 ms; dimensionality reduction step; multiclass classification problem; brain–computer interface; medical signal detection; biomedical telemetry; signal classification; processing unit; medical signal processing; functional approach; custom algorithm; BCI architecture; parietal-cortex area; mechatronic device driving; time-domain features extraction; neurophysiology; machine learning algorithm; mechatronic control
• ISSN: 1751-8806; 1751-8814; 1751-8814
• DOI: 10.1049/iet-sen.2017.0340

This study presents a P300-based brain-computer interface (BCI) for mechatronic device driving, i.e. without the need for any physical control. The technique is based on a machine learning (ML) algorithm, which exploits a spatio-temporal characterization of the P300, analyses all the discrimination scenarios through a multiclass classification problem. The BCI is composed of the acquisition unit, the processing unit and the navigation unit. The acquisition unit is a wireless 32-channel electroencephalography headset collecting data from six electrodes. The processing unit is a dedicated µPC performing stimuli delivery, ML and classification, leading to the user intention interpretation. The ML stage is based on a custom algorithm (tuned residue iteration decomposition) which trains the classifier on the user-tuned P300 features. The extracted features undergo a dimensionality reduction and are used to define decision boundaries for the real-time classification. The real-time classification performs a functional approach for the features extraction, reducing the amount of data to be analyzed. The Raspberry-based navigation unit actuates the received commands, supporting the wheelchair motion. The experimental results, based on a dataset of seven subjects, demonstrate that the classification chain is performed in 8.16 ms with an accuracy of 84.28 ± 0.87%, allowing the real-time control of the wheelchair.