Achieving a hybrid brain–computer interface with tactile selective attention and motor imagery

IEEE Trans. Neural Syst. Rehabil. Eng.

et al. 2018

Abstract-We proposed a multi-class tactile brain-computer interface that utilizes stimulus-induced oscillatory dynamics. It was hypothesized that somatosensory attention can modulate tactile induced oscillation changes, which can decode different sensation attention tasks. Subjects performed four tactile attention tasks, prompted by cues presented in random order and while both wrists were simultaneously stimulated: 1) selective sensation on left hand (SS-L), 2) selective sensation on right hand (SS-R), 3) bilateral selective sensation (SS-B), and 4) selective sensation suppressed or idle state (SS-S). The classification accuracy between SS-L and SS-R (79.9±8.7%) was comparable with that of a previous tactile BCI system based on selective sensation. Moreover, the accuracy could be improved to an average of 90.3±4.9% by optimal class-pair and frequency-band selection. Three-class discrimination had accuracy of 75.2±8.3%, with the best discrimination reached for the classes SS-L, SS-R and SS-S. Finally, four classes were classified with accuracy of 59.4±7.3%. These results show that the proposed system is a promising new paradigm for multi-class BCI.

Section: Discussionmentioning

confidence: 72%

“…Four of the investigated subjects attained an accuracy below 70%. Another study on SSSEP showed a mean classification accuracy of 58% for 16 subjects, with 15 out of 16 subjects below the 70% accuracy level [32]. Subsequently, a tactile P300 system, similar to the visual P300 BCI, based on the oddball paradigm, was proposed [33].…”

Section: Introductionmentioning

confidence: 99%

A Multi-Class Tactile Brain–Computer Interface Based on Stimulus-Induced Oscillatory Dynamics

Yao

Chen

IEEE Trans. Neural Syst. Rehabil. Eng.

et al. 2018

“…Without the requirement of eye movement or focus control, tactile BCI opened new avenue for BCI development, which will be especially useful to those locked-in patients but with partly preserved somatosensory functionality. In the existed tactile BCI systems in the literature, the tactile stimulus evoked transient [13] and steady-state potential [33], [64] were utilized for BCI construction, reaching a mean BCI performance of 72%, 70.4% and 58% respectively, with much higher BCI-illiteracy rate of 90% and 93.75%. In contrast, in our proposed tactile selective sensation BCI, the stimulus-induced oscillatory dynamics was utilized for BCI construction, which has demonstrated a significantly enhanced tactile BCI performance when compared with existing tactile BCI modalities in the literature [13], [33], [64], [65], with much higher classification accuracy in a much larger experiment subject group and much lower BCI-illiteracy rate.…”

Section: A Bci-illiteracy Problem In Tactile Bcimentioning

confidence: 99%

“…Early experimental studies have shown that the classification accuracy for this BCI modality ranged from 64% to 84%, with an average accuracy of 70.4% and a high BCI-illiteracy rate [38]. A subsequent study on SSSEP showed a mean classification accuracy of 58% for 16 subjects, with 15 of the subjects resulting in accuracy less than 70% [39]. Finally, a tactile P300 system, similar to the visual P300 BCI, based on the oddball paradigm, has also been proposed [13] and achieved an accuracy of 72% in 11 subjects, when selecting between two targets.…”

mentioning

confidence: 99%

Performance of Brain–Computer Interfacing Based on Tactile Selective Sensation and Motor Imagery

Yao

IEEE Trans. Neural Syst. Rehabil. Eng.

Mrachacz-Kersting

et al. 2018

-A large proportion of users do not achieve adequate control using current non-invasive Brain-computer Interfaces (BCI). This issue has being coined "BCI-Illiteracy", and is observed among BCI modalities. Here, we compare the performance and BCI-illiteracy rate of tactile selective sensation (SS) and motor imagery (MI) BCI, for large subject samples. We analyzed 80 experimental sessions from 57 subjects with two-class SS protocols. For SS, the group average performance was 79.8±10.6%, with 43 out of the 57 subjects (75.4%) exceeding the 70% BCI-illiteracy threshold for left and right hand SS discrimination. When compared to previous results, this tactile BCI outperformed all other tactile BCIs currently available. We also analyzed 63 experiment sessions from 43 subjects with two-class MI BCI protocols, where the group average performance was 77.2±13.3%, with 69.7% of the subjects exceeded the 70% performance threshold for left and right hand MI. For within-subject comparison, the 24 subjects who participated to both the SS and MI experiments, the BCI performance was superior with SS than MI especially in beta frequency band (p<0.05), with enhanced R 2 discriminative information in the somatosensory cortex for the SS modality. Both SS and MI showed a functional dissociation between lower alpha ([8 10] Hz) and upper alpha ([10 13] Hz) bands, with BCI performance significantly better in the upper alpha than the lower alpha (p<0.05) band. In summary, we demonstrated that SS is a promising BCI modality with low BCI illiteracy issue, and has great potential in practical applications reaching large population.

“…The first prototype of a tactile BCI was proposed by Mueller-Putz et al [46], and based on steady-state somatosensory evoked potentials (SSSEP) [47]- [49], reaching an average accuracy of 70.4%. Another similar study on SSSEP showed a mean classification accuracy of 58% for 16 subjects [50]. Subsequently, a tactile P300 system based on the oddball paradigm, was proposed [51], [52].…”

Section: Discussionmentioning

confidence: 90%

Decoding Covert Somatosensory Attention by a BCI System Calibrated With Tactile Sensation

Yao¹,

IEEE Trans. Biomed. Eng.

Mrachacz-Kersting

et al. 2018

Abstract-Objective: We propose a novel calibration strategy to facilitate the decoding of covert somatosensory attention by exploring the oscillatory dynamics induced by tactile sensation. Methods: It was hypothesized that the similarity of the oscillatory pattern between stimulation sensation (SS, real sensation) and somatosensory attentional orientation (SAO) provides a way to decode covert somatic attention. Subjects were instructed to sense the tactile stimulation, which was applied to the left (SS-L) or the right (SS-R) wrist. The BCI system was calibrated with the sensation data and then applied for online SAO decoding. Results: Both SS and SAO showed oscillatory activation concentrated on the contralateral somatosensory hemisphere. Offline analysis showed that the proposed calibration method led to greater accuracy than the traditional calibration method based on SAO only. This is confirmed by online experiments, where the online accuracy on 15 subjects was 78.8±13.1%, with 12 subjects >70% and 4 subject >90%. Conclusion: By integrating the stimulus-induced oscillatory dynamics from sensory cortex, covert somatosensory attention can be reliably decoded by a BCI system calibrated with tactile sensation. Significance: Indeed, real tactile sensation is more consistent during calibration than SAO. This brain-computer interfacing approach may find application for stroke and completely locked-in patients with preserved somatic sensation.