Time-Variant Linear Discriminant Analysis Improves Hand Gesture and Finger Movement Decoding for Invasive Brain-Computer Interfaces

Abstract

Invasive brain-computer interfaces yield remarkable performance in a multitude of applications. For classification experiments, high-gamma bandpower features and linear discriminant analysis (LDA) are commonly used due to simplicity and robustness. However, LDA is inherently static and not suited to account for transient information that is typically present in high-gamma features. To resolve this issue, we here present an extension of LDA to the time-variant feature space. We call this method time-variant linear discriminant analysis (TVLDA). It intrinsically provides a feature reduction stage, which makes external approaches thereto obsolete, such as feature selection techniques or common spatial patterns (CSPs). As well, we propose a time-domain whitening stage which equalizes the pronounced 1/f-shape of the typical brain-wave spectrum. We evaluated our proposed architecture based on recordings from 15 epilepsy patients with temporarily implanted subdural grids, who participated in additional research experiments besides clinical treatment. The experiments featured two different motor tasks involving three high-level gestures and individual finger movement. We used log-transformed bandpower features from the high-gamma band (50–300 Hz, excluding power-line harmonics) for classification. On average, whitening improved the classification performance by about 11%. On whitened data, TVLDA outperformed LDA with feature selection by 11.8%, LDA with CSPs by 13.9%, and regularized LDA with vectorized features by 16.4%. At the same time, TVLDA only required one or two internal features to achieve this. TVLDA provides stable results even if very few trials are available. It is easy to implement, fully automatic and deterministic. Due to its low complexity, TVLDA is suited for real-time brain-computer interfaces. Training is done in less than a second. TVLDA performed particularly well in experiments with data from high-density electrode arrays. For example, the three high-level gestures were correctly identified at a rate of 99% over all subjects. Similarly, the decoding accuracy of individual fingers was 96% on average over all subjects. To our knowledge, these mean accuracies are the highest ever reported for three-class and five-class motor-control BCIs.

Topological Data Analysis (TDA) Techniques Enhance Hand Pose Classification from ECoG Neural Recordings

“…ECoG signals contain rich information correlated with motor activities, in particular with regards to hand gesture decoding [7]. Such a problem has received a lot of attention recently [8], [9], and can be seen in form of a multiclass classification problem. In this context, each hand gesture corresponds to a class, and the recorded ECoG signals are the objects to be classified.…”

“…In this context, each hand gesture corresponds to a class, and the recorded ECoG signals are the objects to be classified. Previous studies approached this issue as a 3 class classification problem, classifying only the movement trials and ignoring the rest state [8]. In our work we'll deal with 4 classes, considering the rest state as an indipendent class.…”

“…The patient provided written informed consent before participating in the experiments. For additional details (such as detailed electrode placement), we refer to [8], where this patient appears as S2.…”

Investigating Neural Representation of Finger-Movement Directions Using Electroencephalography Independent Components

“…The participants of these studies were epilepsy patients. In [3], three hand gestures and finger tapping movements were classified from electrodes in the somatosensory areas with a classification accuracy of 96.5% with high-density sensorimotor coverage. Another study [4] aimed to classify four different articulators and four different tongue movement directions using a small area in the sensorimotor cortex.…”

Evaluating a Novel P300-Based Real-Time Image Ranking BCI

“…Methods such as linear discriminant analysis (Guger et al, 2009), stepwise linear discriminant analysis (Sellers and Donchin, 2006), support vector machines (Thulasidas et al, 2006), and matched filtering (Serby et al, 2005) have been utilized in P300-based BCIs. The field is maturing with the adoption of new machine learning algorithms that reduce the amount of training data required to achieve sufficient classification accuracy, such as the time-variant Linear discriminant analysis proposed by Gruenwald et al (2019). Complementing these advances, researchers have also focused on improving P300-based BCIs' slow communication rates (Martens et al, 2009;Takano et al, 2009;Mugler et al, 2010).…”