Evaluation of EEG Headset Mounting for Brain-Computer Interface-Based Stroke Rehabilitation by Patients, Therapists, and Relatives

Abstract

Brain-computer interfaces (BCIs) have successfully been used for motor recovery training in stroke patients. However, the setup of BCI systems is complex and may be divided into (1) mounting the headset and (2) calibration of the BCI. One of the major problems is mounting the headset for recording brain activity in a stroke rehabilitation context, and usability testing of this is limited. In this study, the aim was to compare the translational aspects of mounting five different commercially available headsets from a user perspective and investigate the design considerations associated with technology transfer to rehabilitation clinics and home use. No EEG signals were recorded, so the effectiveness of the systems have not been evaluated. Three out of five headsets covered the motor cortex which is needed to pick up movement intentions of attempted movements. The other two were as control and reference for potential design considerations. As primary stakeholders, nine stroke patients, eight therapists and two relatives participated; the stroke patients mounted the headsets themselves. The setup time was recorded, and participants filled in questionnaires related to comfort, aesthetics, setup complexity, overall satisfaction, and general design considerations. The patients had difficulties in mounting all headsets except for a headband with a dry electrode located on the forehead (control). The therapists and relatives were able to mount all headsets. The fastest headset to mount was the headband, and the most preferred headsets were the headband and a behind-ear headset (control). The most preferred headset that covered the motor cortex used water-based electrodes. The patients reported that it was important that they could mount the headset themselves for them to use it every day at home. These results have implications for design considerations for the development of BCI systems to be used in rehabilitation clinics and in the patient's home.

EEG Headset Evaluation for Detection of Single-Trial Movement Intention for Brain-Computer Interfaces

“…It must be stated that some of the headsets were not developed for recording movement-related activity, which was also reflected in the results. However, it was important to include these headsets, since stroke patients in a comparative study preferred them [15].…”

“…It can be fairly time consuming to mount the EEG headset and ensure the cap is correctly placed, and a proper EEG signal quality is obtained. This is especially evident if the patients potentially should mount the EEG headset themselves [15]. These points are related to the usability of the headsets, which became a research topic when new and cheaper headsets and electrode types became available [15][16][17][18].…”

“…This is especially evident if the patients potentially should mount the EEG headset themselves [15]. These points are related to the usability of the headsets, which became a research topic when new and cheaper headsets and electrode types became available [15][16][17][18]. The usability evaluation consists of three factors: effectiveness, efficiency and satisfaction [19], and the former two factors are related to another impeding factor to create a stand-alone BCI system, the need for calibrating the BCI to ensure adequate detection performance.…”

“…Two headsets covered the motor cortex, which is the traditional position to record MRCPs. The other two headsets were placed on the forehead and around the ear, which was preferred by stroke patients in a recent usability study [15]; however, it is not known if MRCPs can be recorded from these positions. We tested whether movement intentions (MRCPs) can be classified with respect to idle activity, which is the scenario that would be used in BCI training for stroke rehabilitation.…”

Design and Characterization of an EEG-Hat for Reliable EEG Measurements

“…In addition, it must provide the least stress possible to the user to minimize interference with the brain activities of interest. Table 1 summarizes the existing methods [ 24 , 25 , 26 , 27 , 28 , 29 , 30 , 31 , 32 , 33 , 34 , 35 , 36 , 37 , 38 , 39 , 40 , 41 ]. Net-type EEG devices are widely used in medical applications because they are capable of acquiring EEG signals from many positions with a good spatial resolution [ 27 ].…”