A systematic rank of smart training environment applications with motor imagery brain-computer interface

Brain-Computer Interface (BCI) research is considered one of the significant interdisciplinary fields. It assists people with severe motor disabilities to recover and improve their motor actions through rehabilitation sessions using Motor Imagery (MI) based BCI systems. Several smart criteria, such as virtual reality, plays a significant role in training people for motor recovery in a virtual environment. Accordingly, Smart Training Environments (STEs) based on virtual reality for MI-BCI users provide a safe environment. They are cost-effective for real-life conditions and scenarios with severe motor disabilities. Fundamentally, the literature presents a lack of comparison of the STE applications considering the smart and effective criteria of the developed applications. Accordingly, three key issues faced the comparison process: importance, multi-evaluation criteria, and data variation, which falls under complex Multi-Criteria Decision Making (MCDM). Performance issues increased comparison complexity caused by the rapidly changing market demands of the MI-BCI. Therefore, this study developed two methodology phases for evaluating and benchmarking the STE applications for the MI-BCI community; making effective decisions is vital. In the first phase, formulate the STE Decision Matrix (DM) based on two main dimensions: the evaluation of ten smart criteria of STE and the alternatives (27 STE applications) developed in the literature for MI-BCI. In the second phase, integration methods of MCDM have been formulated: Analytic Hierarchy Process (AHP) for weighting the ten smart criteria and Fuzzy Decision by Opinion Score Method (FDOSM) for benchmarking STE applications based on constructed AHP weights. The evaluation results show importunity in the obtained weights among the ten STE criteria to distinguish the greatest and lowest important weights. Through the benchmarking performance, FDOSM processes prioritized all STE applications. The ranking results were objectively validated based on five groups of alternatives, and the results were systematically ranked. Finally, this study argued three important summary points concerning the STE dataset, formulated a DM of STE applications, and smart criteria for STE applications to support the MI-BCI community and market. Developing the appropriate STE application for MI-BCI is a better choice to support a large BCI community by identifying the ten smart criteria and considering the presented methodology to establish a robust, practical, cost-efficient, and reliable BCI system.