Towards Effective Non-Invasive Brain-Computer Interfaces Dedicated to Ambulatory Applications

Disabilities affecting mobility, in particular, often lead to exacerbated isolation and thus fewer communication opportunities, resulting in a limited participation in social life. Additionally, as costs for the health-care system can be huge, rehabilitation-related devices and lower-limb prostheses (or orthoses) have been intensively studied so far. However, although many devices are now available, they rarely integrate the direct will of the patient. Indeed, they basically use motion sensors or the residual muscle activities to track the next move. Therefore, to integrate a more direct control from the patient, Brain-Computer Interfaces (BCIs) are here proposed and studied under ambulatory conditions. Basically, a BCI allows you to control any electric device without the need of activating muscles. In this work, the conversion of brain signals into a prosthesis kinematic control is studied following two approaches. First, the subject transmits his desired walking speed to the BCI. Then, this high-level command is converted into a kinematics signal thanks to a Central Pattern Generator (CPG)-based gait model, which is able to produce automatic gait patterns. Our work thus focuses on how BCIs do behave in ambulatory conditions. The second strategy is based on the assumption that the brain is continuously controlling the lower limb. Thus, a direct interpretation, i.e. decoding, from the brain signals is performed. Here, our work consists in determining which part of the brain signals can be used. This manuscript consists of three different parts: ambulatory BCIs, gait modelling and gaitrelated spontaneous brain signals. In the first part, it is shown that for several standard BCIs based on P300 and SSVEP, the effect of gait-related artefacts is not that important. Additionally, it was demonstrated that an online application using SSVEP and P300 are definitely feasible, while the SSVEP interface is much more accepted by subjects based on subjective questionnaires. A comparison between the low-cost Emotiv Epoc Headset and a medical device shows that the former system has the potential to increase the BCI use among end-users. To expand control interfaces to eye movements, electrooculographic signal based recognition methods have been intensively reviewed showing that little difference exists between the standard and modified approaches. In the second part, two gait model approaches are studied. A modified Programmable Central Pattern Generator (PCPG) algorithm is proposed to fit at best the actual patient movement. First, the parameters of the model are continuously adapted depending on the speed thanks to a polynomial interpolation. Then, as gait is not perfectly periodic, a new soft-phase resetting method was proposed and assessed. Second, a Dynamic Recurrent Neural Network combined with sinusoidal inputs is shown to provide a better fit than the PCPG approach.