Chapter

Brain-Computer Interface Research

Part of the series SpringerBriefs in Electrical and Computer Engineering pp 23-37

Date:

What’s Your Next Move? Detecting Movement Intention for Stroke Rehabilitation

  • R. ZimmermannAffiliated withRehabilitation Engineering Lab, ETH ZurichBiomedical Optics Research Lab, University Hospital Zurich Email author 
  • , L. Marchal-CrespoAffiliated withSensory-Motor Systems Lab, ETH ZurichBalgrist University Hospital, University of Zurich
  • , O. LambercyAffiliated withRehabilitation Engineering Lab, ETH Zurich
  • , M. -C. FluetAffiliated withRehabilitation Engineering Lab, ETH Zurich
  • , J. -C. MetzgerAffiliated withRehabilitation Engineering Lab, ETH Zurich
  • , J. EdelmannAffiliated withRehabilitation Engineering Lab, ETH Zurich
  • , J. BrandAffiliated withInstitute of Neuroinformatics, University of Zurich
  • , K. EngAffiliated withInstitute of Neuroinformatics, University of Zurich
  • , R. RienerAffiliated withSensory-Motor Systems Lab, ETH ZurichBalgrist University Hospital, University of Zurich
    • , M. WolfAffiliated withBiomedical Optics Research Lab, University Hospital Zurich
    • , R. GassertAffiliated withRehabilitation Engineering Lab, ETH Zurich

* Final gross prices may vary according to local VAT.

Get Access

Abstract

BCIs have recently been identified as a method to promote restorative neuroplastic changes in patients with severe motor impairment, such as after a stroke. In this chapter, we describe a novel therapeutic strategy for hand rehabilitation making use of this method. The approach consists of recording brain activity in cortical motor areas by means of near-infrared spectroscopy, and complementing the cortical signals with physiological data acquired simultaneously. By combining these signals, we aim at detecting the intention to move using a multi-modal classification algorithm. The classifier output then triggers assistance from a robotic device, in order to execute the movement and provide sensory stimulation at the level of the hand as response to the detected motor intention. Furthermore, the cortical data can be used to control audiovisual feedback, which provides a context and a motivating training environment. It is expected that closing the sensorimotor loop with such a brain-body-robot interface will promote neuroplasticity in sensorimotor networks and support the recovery process.