Self-paced Movement Intention Detection from Human Brain Signals: Invasive and Non-invasive EEG

Neural signatures of humans’ movement intention can be exploited by future neuroprosthesis. We propose a method for detecting self-paced upper limb movement intention from brain signals acquired with both invasive and noninvasive methods. In the first study with scalp electroencephalograph (EEG) signals from healthy controls, we report single trial detection of movement intention using movement related potentials (MRPs) in a frequency range between 0.1 to 1 Hz. Movement intention can be detected above chance level (p

Non-Invasive Estimates of Local Field Potentials for Brain-Computer Interfaces

Pierre Ferrez

Recent experiments have shown the possibility to use the brain electrical activity to directly control the movement of robots or prosthetic devices in real time. Such neuroprostheses can be invasive or non-invasive, depending on how the brain signals are recorded. In principle, invasive approaches will provide a more natural and flexible control of neuroprostheses, but their use in humans is debatable given the inherent medical risks. Non-invasive approaches mainly use scalp electroencephalogram (EEG) signals and their main disadvantage is that these signals represent the noisy spatiotemporal overlapping of activity arising from very diverse brain regions; i.e., a single scalp electrode picks up and mixes the temporal activity of myriads of neurons at very different brain areas. In order to combine the benefits of both approaches, we propose to rely on the non-invasive estimation of local field potentials (eLFP) in the whole human brain from the scalp measured EEG data using a recently developed inverse solution (ELECTRA) to the EEG inverse problem. The goal of a linear inverse procedure is to deconvolve or unmix the scalp signals attributing to each brain area its own temporal activity. To illustrate the advantage of this approach we compare, using identical set of spectral features, classification of rapid voluntary finger self-tapping with left and right hands based on scalp EEG and eLFP on three subjects using different number of electrodes. It is shown that the eLFP-based Gaussian classifier outperforms the EEG-based Gaussian classifier for the three subjects.

The MIT Press2007

Motor state transitions and Breathing in Brain Machine Interfaces

Bastien Orset

Brain-Machine interfaces aim to create a direct neural link between user's brain and machines. This goal has pushed scientists to investigate a large spectrum of applications in the realm of assistive and rehabilitation technologies. However, despite great progress, the possibility of being in full control of a device with our brain is still far to be reached. One of the primary limitations of BMIs is the notion of self-paced control, which states that a user should be able to activate or deactivate such interface by mere modulation of brain activity. Many studies have investigated this question, focusing notably on the use of endogenous signals (such as Event-Related Desynchronization/Synchronization, ERD/S) to control the activation of these interfaces in natural interaction scenarios. This question is, however, largely ignored when looking at their interruption. Hence, in this first part of the thesis, my work was devoted to investigate how a BMI user would be able to control the interruption of non-invasive BMIs based on Motor Imagery (MI), a paradigm promoting natural and endogenous signals that can be used for BMI control. I investigated the decoding of motor termination as well as its correlates characterized by the post-movement ERS phenomenon. Specifically, this part of the thesis aimed to study (i) the feasibility to decode motor termination from a specific neural correlate as well as the adaptation from the user in closed-loop scenarios, (ii) the benefits of using motor termination correlates to detect the stopping process, and (iii) the effect of BMI effectors such as an exoskeleton on the detection of motor termination correlates. The obtained results provide new insights into closed-loop decoding of motor termination. In particular, I show that BMI users exhibit an adaptation of their EEG correlates enabling them to have a reliable control when switching off a BMI in closed-loop scenarios. Second, I show that the decoding of motor termination is a particular process different from a resting state and, hence, should be decoded independently so as to achieve a faster and more robust detection. Finally, I investigate the use of BMI effectors with respect to the decoding of motor termination showing an effect of the effectors on the correlates of motor termination. However, due to the nature of these correlates, motor termination can still be reliably decoded with a similar latency. In the second part of my thesis, I also investigate how the interoceptive system affects BMI and particularly how breathing signals affect BMI based on MI. Breathing has been shown to have a key effect on numerous human functions, including, perceptual, cognitive and motor functions. However, currently not much is known about the role of respiratory signals in BMI and such signals have rather been considered as physiological noise. In my thesis, I evaluated how the breathing process affects the correlates of MI (ERD in mu and beta bands) as well as actual BMI performance. The results provide an extensive analysis regarding the effect of the breathing cycle specifically on mu-ERDs, showing stronger ERD during the late expiration phase. Moreover, I identified a link between breathing and BMI performance and propose that breathing signals are a valuable predictor for BMI performance highlighting the importance of monitoring such signals and, more generally, present the interoceptive system as a key component of motor preparation and motor imagery

EPFL2021

Motor state transitions and Breathing in Brain Machine Interfaces

Bastien Orset

EPFL2021