Portable and Personalized Neurofeedback for Stroke Rehabilitation: Adapting Neurofeedback “Output” Using Multisensory Stimulations

Regarding the output of NF training, current systems rely mostly on a single and static sensory stimulation usually providing basic visual cues. Yet, the feedback is known to play a major role in the learning process of NF or brain-computer interfaces (BCI) [1,2]. The feedback informs the user about the quality of his performance in real time to help him control his or her brain activity. Carefully selecting and adapting feedback is expected to reduce the time required to learn to control the system and their brain activity.

In order to develop NF protocols that can target a specific region or one of the networks that will be identified in WP1.a, we intend to propose novel kinds of NF feedbacks that will involve multisensory stimuli, and notably haptic sensory cues. We could notably stress in a recent survey paper [3] that haptic interfaces have the potential to improve NF performance and increase the pertinence of the feedback provided. As a matter of fact, haptic-based BCI/NF particularly seems to be a promising way for post-stroke motor rehabilitation, as this non-invasive technique may contribute to close the sensorimotor loop between brain and effect [4]. Such haptic feedbacks will therefore be designed and adapted in order to activate brain regions specific to the considered pathology according to our rehabilitation scenario.

We already have experience in using a haptic (and MR-compatible) vibrotactile actuator for NF stimulation of stroke patients [5]. As shown in figure 4, our group has also investigated the use of virtual reality visual feedback and haptic tendon vibration of the wrist in healthy participants [3,6]. Our first results are promising and demonstrate the impact that these new multisensory neurofeedback methods could have in improving stroke rehabilitation. However, how to use these visuo-haptic systems, individually or in combination with other modalities, remains an open question that has not been thoroughly addressed by other groups. These haptic systems are therefore available for this project and we foresee to use them in our NF control loop with stroke patients.

Our ambition within this research axis, and in particular through the planned post-doctoral fellowship, will thus be to leverage our preliminary results and to develop a methodological and experimental framework for implementing an “adapted” haptic-based NF, in association with visual feedback of virtual reality type. This framework should be robust enough to be able to demonstrate feasibility with stroke patients. We will focus on two main aspects. First, we will investigate the notion of “adaptation/personalization” of the sensory feedback. We will notably focus on how to adapt the feedback depending on: 1) the personal characteristics of the user (preference, personality traits, disability properties, etc), and 2) the evolution of his results and performance (during one session, or across sessions). Second, we will investigate the mapping of NF targets (among the various potential features or brain patterns involved) with the different available sensory modalities. In other words, we intend to better qualify which NF features should be sent to which sensory modality. In particular, and in relation with WP1.a, it will be very interesting to assess which modality allows to modulate most efficiently the brain connectivity and the networks identified previously. We are planning to conduct a systematic series of pilot experiments on healthy participants to identify the most promising combinations. For this aim, we can rely on the numerous haptic and virtual reality devices available at IRISA laboratory, including haptic actuators (vibrotactile stimulators, skin-stretching wearable setups, force-feedback/kinaesthetic interfaces) and 3D displays (VR headsets, stereoscopic screens). 

[1] A. Lecuyer et al., “Brain-Computer Interfaces, Virtual Reality, and Videogames,” Computer, vol. 41, no. 10, pp. 66–72, 2008.
[2] L. Perronnet et al. , “Brain training with neurofeedback”, book chapter in “Brain-Computer Interfaces”, Wiley-ISTE, 2016.
[3] M. Fleury et al., “A Survey on the Use of Haptic Feedback for Brain-Computer Interfaces and Neurofeedback,” Front. Neurosci., vol. 14, 2020
[4] M. Gomez-Rodriguez et al. “Closing the sensorimotor loop: haptic feedback facilitates decoding of motor imagery,” J. Neural Eng., vol. 8, no. 3, p. 036005, Jun. 2011
[5] S. Leplaideur et al., “Short-term effect of neck muscle vibration on postural disturbances in stroke patients,” Exp. Brain Res., vol. 234, no. 9, p. 2643, 2016
[6] S. L. Franc et al., “Influence of virtual reality visual feedback on the illusion of movement induced by tendon vibration of wrist in healthy participants,” PLOS ONE, vol. 15, no. 11, p. e0242416, Nov. 2020