Telerehabilitation (TR) has gained attention as a promising rehabilitation format. Our study examined how patients responded to TR and whether it provided adequate support for their lifestyle changes and self-care efforts when compared to conventional rehabilitation (CR). Cardiac patients (n = 136) were randomly assigned to a TR or CR group. The TR group was provided with relevant health care technology for a period of three months, and both groups filled in questionnaires on their motivation for lifestyle changes and self-care psychological distress, and quality of life at 0, 3, 6, and 12 months. Patients in both groups were found to be equally motivated for lifestyle changes and self-care (p < 0.05) and they experienced similar levels of psychological distress and quality of life. TR is comparable to conventional rehabilitation in motivating patients, preventing psychological distress and improving quality of life. Although we observed an initial increase in autonomous motivation in the telerehabilitation group, this positive difference in motivation does not last over time. As such, neither rehabilitation format seems able to ensure long-term motivation. Therefore, TR may serve as a viable replacement for conventional rehabilitation when considered relevant. Further research is needed to enhance long-term motivation, and maybe telerehabilitation can help to achieve this.

Objective. Human-computer interfaces (HCI) are potential tools for assisting (movement replacement) and rehabilitating (movement restoration) individuals with spinal cord injury (SCI). HCIs based on electroencephalography (EEG) have limited accuracy and hence control options; this could be improved by exploiting potential residual muscle activity (electromyography, EMG). The study objectives were to determine if combined EEG and EMG improves offline singletrial movement classification. Furthermore, the effect of number of classes and detection latency on the accuracies was investigated. Methods. Ten able-bodied and eight SCI subjects performed elbow flexion/extension at three force levels while EEG and EMG were recorded. Temporal and spectral features were extracted from the EEG and Hudgins time domain features were extracted from the EMG in 1-second time windows. The time window was shifted (200-ms shift) over 5second epochs around the movement onset. Each segment was classified in three scenarios (2, 3 or 7 classes) using linear discriminant analysis. Results. The accuracies obtained with EEG (51.2%) was outperformed by EMG (95.5%) and combined EMG and EEG (96.2%). Immediately after the EMG onset, the accuracies increased and rapidly reached a plateau. High accuracies were obtained for the different number of classes. Conclusion and Significance. EMG was crucial for obtaining high accuracies, and potential residual EMG should be exploited in HCIs to improve the performance. Force proved to be a viable option for SCI subjects with residual EMG to increase the number of classes for HCI control. These findings could assist design considerations of HCIs for SCI individuals.

For individuals with severe motor deficiencies, controlling external devices such as robotic arms or wheelchairs can be challenging, as many devices require some degree of motor control to be operated, e.g. when controlled using a joystick. A brain-computer interface (BCI) relies only on signals from the brain and may be used as a controller instead of muscles. Motor imagery (MI) has been used in many studies as a control signal for BCIs. However, MI may not be suitable for all control purposes, and several people cannot obtain BCI control with MI. In this study, the aim was to investigate the feasibility of decoding covert speech from single-trial EEG and compare and combine it with MI. In seven healthy subjects, EEG was recorded with twenty-five channels during six different actions: Speaking three words (both covert and overt speech), two arm movements (both motor imagery and execution), and one idle class. Temporal and spectral features were derived from the epochs and classified with a random forest classifier. The average classification accuracy was 67 ± 9 % and 75 ± 7 % for covert and overt speech, respectively; this was 5-10 % lower than the movement classification. The performance of the combined movement-speech decoder was 61 ± 9 % and 67 ± 7 % (covert and overt), but it is possible to have more classes available for control. The possibility of using covert speech for controlling a BCI was outlined; this is a step towards a multimodal BCI system for improved usability.

No abstract

Error-related potentials (ErrPs) have been proposed for designing adaptive brain-computer interfaces (BCIs). Therefore, ErrPs must be decoded. The aim of this study was to evaluate ErrP decoding using combinations of different feature types and classifiers in BCI paradigms involving motor execution (ME) and imagination (MI). Fifteen healthy subjects performed 510 (ME) and 390 (MI) trials of right/left wrist extensions and foot dorsiflexions. Sham BCI feedback was delivered with an accuracy of 80% (ME) and 70% (MI). Continuous EEG was recorded and divided into ErrP and NonErrP epochs. Temporal, spectral, discrete wavelet transform (DWT) marginals, and template matching features were extracted, and all combinations of feature types were classified using linear discriminant analysis, support vector machine, and random forest classifiers. ErrPs were elicited for both ME and MI paradigms, and the average classification accuracies were significantly higher than the chance level. The highest average classification accuracy was obtained using temporal features and a combination of temporal+DWT features classified with random forest; 89±9% and 83±9% for ME and MI, respectively. These results generally indicate that temporal features should be used when detecting ErrPs, but there is great intersubject variability, which means that user-specific features should be derived to maximize the performance.