Iñaki Iturrate

Latency correction of error potentials between different experiments reduces calibration time for single-trial classification

One fundamental limitation of EEG-based brain-computer interfaces is the time needed to calibrate the system prior to the detection of signals, due to the wide variety of issues affecting the EEG measurements. For event-related potentials (ERP), one of these sources of variability is the application performed: Protocols with different cognitive workloads might yield to different latencies of the ERPs. In this sense, it is still not clear the effect that these latency variations have on the single-trial classification. This work studies the differences in the latencies of error potentials across three experiments with increasing cognitive workloads. A delay-correction algorithm based on the cross-correlation of the averaged signals is presented, and tested with a single-trial classification of the signals. The results showed that latency variations exist between different protocols, and that it is feasible to re-use data from previous experiments to calibrate a classifier able to detect the signals of a new experiment, thus reducing the calibration time.

Teaching brain-machine interfaces as an alternative paradigm to neuroprosthetics control

Brain-machine interfaces (BMI) usually decode movement parameters from cortical activity to control neuroprostheses. This requires subjects to learn to modulate their brain activity to convey all necessary information, thus imposing natural limits on the complexity of tasks that can be performed. Here we demonstrate an alternative and complementary BMI paradigm that overcomes that limitation by decoding cognitive brain signals associated with monitoring processes relevant for achieving goals. In our approach the neuroprosthesis executes actions that the subject evaluates as erroneous or correct, and exploits the brain correlates of this assessment to learn suitable motor behaviours. Results show that, after a short user’s training period, this teaching BMI paradigm operated three different neuroprostheses and generalized across several targets. Our results further support that these error-related signals reflect a task-independent monitoring mechanism in the brain, making this teaching paradigm scalable. We anticipate this BMI approach to become a key component of any neuroprosthesis that mimics natural motor control as it enables continuous adaptation in the absence of explicit information about goals. Furthermore, our paradigm can seamlessly incorporate other cognitive signals and conventional neuroprosthetic approaches, invasive or non-invasive, to enlarge the range and complexity of tasks that can be accomplished.

Word pair classification during imagined speech using direct brain recordings.

People that cannot communicate due to neurological disorders would benefit from an internal speech decoder. Here, we showed the ability to classify individual words during imagined speech from electrocorticographic signals. In a word imagery task, we used high gamma (70–150 Hz) time features with a support vector machine model to classify individual words from a pair of words. To account for temporal irregularities during speech production, we introduced a non-linear time alignment into the SVM kernel. Classification accuracy reached 88% in a two-class classification framework (50% chance level), and average classification accuracy across fifteen word-pairs was significant across five subjects (mean = 58%; p < 0.05). We also compared classification accuracy between imagined speech, overt speech and listening. As predicted, higher classification accuracy was obtained in the listening and overt speech conditions (mean = 89% and 86%, respectively; p < 0.0001), where speech stimuli were directly presented. The results provide evidence for a neural representation for imagined words in the temporal lobe, frontal lobe and sensorimotor cortex, consistent with previous findings in speech perception and production. These data represent a proof of concept study for basic decoding of speech imagery, and delineate a number of key challenges to usage of speech imagery neural representations for clinical applications.

Shared-control brain-computer interface for a two dimensional reaching task using EEG error-related potentials

One of the main problems of EEG-based brain computer interfaces (BCIs) is their low information rate, thus for complex tasks the user needs large amounts of time to solve the task. In an attempt to reduce this time and improve the application robustness, recent works have explored shared-control strategies where the device does not only execute the decoded commands, but it is also involved in executing the task. This work proposes a shared-control BCI using error potentials for a 2D reaching task with discrete actions and states. The proposed system has several interesting properties: the system is scalable without increasing the complexity of the users mental task; the interaction is natural for the user, as the mental task is to monitor the device performance to promote its task learning (in this context the reaching task); and the system has the potential to be combined with additional brain signals to recover or learn from interaction errors. Online control experiments were performed with four subjects, showing that it was possible to reach a goal location from any starting point within a 5×5 grid in around 23 actions (about 19 seconds of EEG signal), both with fixed goals and goals freely chosen by the users.

Using frequency-domain features for the generalization of EEG error-related potentials among different tasks

EEG brain-computer interfaces (BCI) require a calibration phase prior to the on-line control of the device, which is a difficulty for the practical development of this technology as it is user-, session- and task-specific. The large body of research in BCIs based on event-related potentials (ERP) use temporal features, which have demonstrated to be stable for each user along time, but do not generalize well among tasks different from the calibration task. This paper explores the use of low frequency features to improve the generalization capabilities of the BCIs using error-potentials. The results show that there exists a stable pattern in the frequency domain that allows a classifier to generalize among the tasks. Furthermore, the study also shows that it is possible to combine temporal and frequency features to obtain the best of both domains.

EEG-based decoding of error-related brain activity in a real-world driving task.

OBJECTIVES Recent studies have started to explore the implementation of brain-computer interfaces (BCI) as part of driving assistant systems. The current study presents an EEG-based BCI that decodes error-related brain activity. Such information can be used, e.g., to predict drivers intended turning direction before reaching road intersections. APPROACH We executed experiments in a car simulator (N = 22) and a real car (N = 8). While subject was driving, a directional cue was shown before reaching an intersection, and we classified the presence or not of an error-related potentials from EEG to infer whether the cued direction coincided with the subjects intention. In this protocol, the directional cue can correspond to an estimation of the driving direction provided by a driving assistance system. We analyzed ERPs elicited during normal driving and evaluated the classification performance in both offline and online tests. RESULTS An average classification accuracy of 0.698 ± 0.065 was obtained in offline experiments in the car simulator, while tests in the real car yielded a performance of 0.682 ± 0.059. The results were significantly higher than chance level for all cases. Online experiments led to equivalent performances in both simulated and real car driving experiments. These results support the feasibility of decoding these signals to help estimating whether the drivers intention coincides with the advice provided by the driving assistant in a real car. SIGNIFICANCE The study demonstrates a BCI system in real-world driving, extending the work from previous simulated studies. As far as we know, this is the first online study in real car decoding drivers error-related brain activity. Given the encouraging results, the paradigm could be further improved by using more sophisticated machine learning approaches and possibly be combined with applications in intelligent vehicles.

mano: A Wearable Hand Exoskeleton for Activities of Daily Living and Neurorehabilitation

Hand sensorimotor impairments are among the most common consequences of injuries affecting the central and peripheral nervous systems, leading to a drastic reduction in the quality of life for affected individuals. Combining wearable robotic exoskeletons and human–machine interfaces is a promising avenue for the restoration and substitution of lost and impaired functions for these users. In this study, we present a novel hand exoskeleton, mano, designed to assist and restore hand functions of people with motor disabilities during activities of daily living (ADL) and in neurorehabilitative scenarios. Compared to state-of-the-art devices, our system is fully wearable, portable, and minimally obtrusive on the hand. The exoskeleton can actively control flexion and extension of all fingers, while allowing natural somatosensorial interactions with the environment surrounding the users. We evaluated the device from four different perspectives. A mechanical characterization, showing that the exoskeleton can cover more than 70% of healthy hand workspace and it can achieve forces at the fingertips sufficient for ADL. A functional characterization, where we showed how two users who suffered from spinal cord injuries were able to perform several ADL for the first time since their accidents. Third, we evaluated the system from a neuroimaging perspective, showing that the device can elicit EEG brain patterns typical of natural hand motions. We finally exemplified the control of the hand exoskeleton within an exemplar framework, a brain–machine interface scenario, showing that motor intention can be decoded for a continuous control of the device. Overall, our results showed that the device represents an ecological solution for use both in ADL and in scenarios aimed at promoting sensorimotor recovery.

Detecting intention to grasp during reaching movements from EEG

Brain-computer interfaces (BCI) have been shown to be a promising tool in rehabilitation and assistive scenarios. Within these contexts, brain signals can be decoded and used as commands for a robotic device, allowing to translate users intentions into motor actions in order to support the users impaired neuro-muscular system. Recently, it has been suggested that slow cortical potentials (SCPs), negative deflections in the electroencephalographic (EEG) signals peaking around one second before the initiation of movements, might be of interest because they offer an accurate time resolution for the provided feedback. Many state-of-the-art studies exploiting SCPs have focused on decoding intention of movements related to walking and arm reaching, but up to now few studies have focused on decoding the intention to grasp, which is of fundamental importance in upper-limb tasks. In this work, we present a technique that exploits EEG to decode grasping correlates during reaching movements. Results obtained with four subjects show the existence of SCPs prior to the execution of grasping movements and how they can be used to classify, with accuracy rates greater than 70% across all subjects, the intention to grasp. Using a sliding window approach, we have also demonstrated how this intention can be decoded on average around 400 ms before the grasp movements for two out of four subjects, and after the onset of grasp itself for the two other subjects.

Decoding of two hand grasping types from EEG

Introduction: Arm and hand movements are essential for performing activities of daily living (ADL). As a result, people with severe motor disabilities would greatly benefit from hand neuroprostheses for restoring grasping capabilities. Non-invasive prostheses mostly rely on EEG correlates of reaching, such as anticipatory potentials for movement initiation [1] or sensorimotor rhythms for movement execution [2]. In this work, we report EEG correlates for two different grasping types and the feasibility of performing reliable detection in single trials.

Exploiting Task Constraints for Self-Calibrated Brain-Machine Interface Control Using Error-Related Potentials

This paper presents a new approach for self-calibration BCI for reaching tasks using error-related potentials. The proposed method exploits task constraints to simultaneously calibrate the decoder and control the device, by using a robust likelihood function and an ad-hoc planner to cope with the large uncertainty resulting from the unknown task and decoder. The method has been evaluated in closed-loop online experiments with 8 users using a previously proposed BCI protocol for reaching tasks over a grid. The results show that it is possible to have a usable BCI control from the beginning of the experiment without any prior calibration. Furthermore, comparisons with simulations and previous results obtained using standard calibration hint that both the quality of recorded signals and the performance of the system were comparable to those obtained with a standard calibration approach.