'Are you even listening?' - EEG-based decoding of absolute auditory attention to natural speech.

Objective.In this study, we use electroencephalography (EEG) recordings to determine whether a subject is actively listening to a presented speech stimulus. More precisely, we aim to discriminate between an active listening condition, and a distractor condition where subjects focus on an unrelated distractor task while being exposed to a speech stimulus. We refer to this task as absolute auditory attention decoding.Approach.We re-use an existing EEG dataset where the subjects watch a silent movie as a distractor condition, and introduce a new dataset with two distractor conditions (silently reading a text and performing arithmetic exercises). We focus on two EEG features, namely neural envelope tracking (NET) and spectral entropy (SE). Additionally, we investigate whether the detection of such an active listening condition can be combined with a selective auditory attention decoding (sAAD) task, where the goal is to decide to which of multiple competing speakers the subject is attending. The latter is a key task in so-called neuro-steered hearing devices that aim to suppress unattended audio, while preserving the attended speaker.Main results.Contrary to a previous hypothesis of higher SE being related with actively listening rather than passively listening (without any distractors), we find significantly lower SE in the active listening condition compared to the distractor conditions. Nevertheless, the NET is consistently significantly higher when actively listening. Similarly, we show that the accuracy of a sAAD task improves when evaluating the accuracy only on the highest NET segments. However, the reverse is observed when evaluating the accuracy only on the lowest SE segments.Significance.We conclude that the NET is more reliable for decoding absolute auditory attention as it is consistently higher when actively listening, whereas the relation of the SE between actively and passively listening seems to depend on the nature of the distractor.

Neuronal tuning and population representations of shape and category in human visual cortex.

Object recognition and categorization are essential cognitive processes which engage considerable neural resources in the human ventral visual stream. However, the tuning properties of human ventral stream neurons for object shape and category are virtually unknown. We performed large-scale recordings of spiking activity in human Lateral Occipital Complex in response to stimuli in which the shape dimension was dissociated from the category dimension. Consistent with studies in nonhuman primates, the neuronal representations were primarily shape-based, although we also observed category-like encoding for images of animals. Surprisingly, linear decoders could reliably classify stimulus category even in data sets that were entirely shape-based. In addition, many recording sites showed an interaction between shape and category tuning. These results represent a detailed study on shape and category coding at the neuronal level in the human ventral visual stream, furnishing essential evidence that reconciles human imaging and macaque single-cell studies.

Avoiding Post-Processing With Event-Based Detection in Biomedical Signals.

OBJECTIVE: Finding events of interest is a common task in biomedical signal processing. The detection of epileptic seizures and signal artefacts are two key examples. Epoch-based classification is the typical machine learning framework to detect such signal events because of the straightforward application of classical machine learning techniques. Usually, post-processing is required to achieve good performance and enforce temporal dependencies. Designing the right post-processing scheme to convert these classification outputs into events is a tedious, and labor-intensive element of this framework. METHODS: We propose an event-based modeling framework that directly works with events as learning targets, stepping away from ad-hoc post-processing schemes to turn model outputs into events. We illustrate the practical power of this framework on simulated data and real-world data, comparing it to epoch-based modeling approaches. RESULTS: We show that event-based modeling (without tailored post-processing) performs on par with or better than epoch-based modeling with extensive post-processing. CONCLUSION: These results show the power of treating events as direct learning targets, instead of using ad-hoc post-processing to obtain them, severely reducing design effort. Significance The event-based modeling framework can easily be applied to other event detection problems in signal processing, removing the need for intensive task-specific post-processing.

What are wereallydecoding? Unveiling biases in EEG-based decoding of the spatial focus of auditory attention.

Objective.Spatial auditory attention decoding (Sp-AAD) refers to the task of identifying the direction of the speaker to which a person is attending in a multi-talker setting, based on the listener's neural recordings, e.g. electroencephalography (EEG). The goal of this study is to thoroughly investigate potential biases when training such Sp-AAD decoders on EEG data, particularly eye-gaze biases and latent trial-dependent confounds, which may result in Sp-AAD models that decode eye-gaze or trial-specific fingerprints rather than spatial auditory attention.Approach.We designed a two-speaker audiovisual Sp-AAD protocol in which the spatial auditory and visual attention were enforced to be either congruent or incongruent, and we recorded EEG data from sixteen participants undergoing several trials recorded at distinct timepoints. We trained a simple linear model for Sp-AAD based on common spatial patterns filters in combination with either linear discriminant analysis (LDA) or k-means clustering, and evaluated them both across- and within-trial.Main results.We found that even a simple linear Sp-AAD model is susceptible to overfitting to confounding signal patterns such as eye-gaze and trial fingerprints (e.g. due to feature shifts across trials), resulting in artificially high decoding accuracies. Furthermore, we found that changes in the EEG signal statistics across trials deteriorate the trial generalization of the classifier, even when the latter is retrained on the test trial with an unsupervised algorithm.Significance.Collectively, our findings confirm that there exist subtle biases and confounds that can strongly interfere with the decoding of spatial auditory attention from EEG. It is expected that more complicated non-linear models based on deep neural networks, which are often used for Sp-AAD, are even more vulnerable to such biases. Future work should perform experiments and model evaluations that avoid and/or control for such biases in Sp-AAD tasks.

Identifying temporal correlations between natural single-shot videos and EEG signals.

Objective.Electroencephalography (EEG) is a widely used technology for recording brain activity in brain-computer interface (BCI) research, where understanding the encoding-decoding relationship between stimuli and neural responses is a fundamental challenge. Recently, there is a growing interest in encoding-decoding natural stimuli in a single-trial setting, as opposed to traditional BCI literature where multi-trial presentations of synthetic stimuli are commonplace. While EEG responses to natural speech have been extensively studied, such stimulus-following EEG responses to natural video footage remain underexplored.Approach.We collect a new EEG dataset with subjects passively viewing a film clip and extract a few video features that have been found to be temporally correlated with EEG signals. However, our analysis reveals that these correlations are mainly driven by shot cuts in the video. To avoid the confounds related to shot cuts, we construct another EEG dataset with natural single-shot videos as stimuli and propose a new set of object-based features.Main results.We demonstrate that previous video features lack robustness in capturing the coupling with EEG signals in the absence of shot cuts, and that the proposed object-based features exhibit significantly higher correlations. Furthermore, we show that the correlations obtained with these proposed features are not dominantly driven by eye movements. Additionally, we quantitatively verify the superiority of the proposed features in a match-mismatch task. Finally, we evaluate to what extent these proposed features explain the variance in coherent stimulus responses across subjects.Significance.This work provides valuable insights into feature design for video-EEG analysis and paves the way for applications such as visual attention decoding.

Deciding individual treatment for primary retroperitoneal sarcoma.

INTRODUCTION: Retroperitoneal sarcomas (RPS) are rare mesenchymal tumors that account for only 0.1-0.2% of all malignancies. Management of this disease is challenging, and resection remains the cornerstone of treatment. Ongoing international collaboration has expanded our knowledge of this disease, allowing for a more personalized approach to RPS patients resulting in improved survival over time. Due to the heterogeneity of RPS, with differing recurrence patterns and sensitivities to neoadjuvant therapies based on histology and grade, management of RPS should be tailored to the individual patient. AREAS COVERED: Our review focuses on a histology-driven approach in the management of primary RPS. We searched relevant articles from 1993 to 2023 that investigated prognostic factors and treatment of patients with RPS and summarized recent advances and future directions in the field. EXPERT OPINION: Deeper understanding of the role of neoadjuvant radiotherapy and ongoing trials investigating the role of neoadjuvant chemotherapy will potentially contribute to the development of individualized treatment pathways.

Sensor selection and miniaturization limits for detection of interictal epileptiform discharges with wearable EEG.

Objective.The goal of this paper is to investigate the limits of electroencephalography (EEG) sensor miniaturization in a set-up consisting of multiple galvanically isolated EEG units to record interictal epileptiform discharges (IEDs), referred to as 'spikes', in people with epilepsy.Approach.A dataset of high-density EEG recordings (257 channels) was used to emulate local EEG sensor units with short inter-electrode distances. A computationally efficient sensor selection and interictal spike detection algorithm was developed and used to assess the influence of the inter-electrode distance and the number of such EEG units on spike detection performance. Signal-to-noise ratio, correlation with a clinical-grade IEDs detector and Cohen's kappa coefficient of agreement were used to quantify performance. Bayesian statistics were used to confirm the statistical significance of the observed results.Main results.We found that EEG recording equipment should be specifically designed to measure the small signal power at short inter-electrode distance by providing an input referred noise<300 nV. We also found that an inter-electrode distance of minimum 5 cm between electrodes in a setup with a minimum of two EEG units is required to obtain near equivalent performance in interictal spike detection to standard EEG.Significance.These findings provide design guidelines for miniaturizing EEG systems for long term ambulatory monitoring of interictal spikes in epilepsy patients.

Bandwidth-Efficient Distributed Neural Network Architectures With Application to Neuro-Sensor Networks.

In this paper, we describe a design methodology to design distributed neural network architectures that can perform efficient inference within sensor networks with communication bandwidth constraints. The different sensor channels are distributed across multiple sensor devices, which have to exchange data over bandwidth-limited communication channels to solve a classification task. Our design methodology starts from a centralized neural network and transforms it into a distributed architecture in which the channels are distributed over different nodes. The distributed network consists of two parallel branches, whose outputs are fused at the fusion center. The first branch collects classification results from local, node-specific classifiers while the second branch compresses each node's signal and then reconstructs the multi-channel time series for classification at the fusion center. We further improve bandwidth gains by dynamically activating the compression path when the local classifications do not suffice. We validate this method on a motor execution task in an emulated EEG sensor network and analyze the resulting bandwidth-accuracy trade-offs. Our experiments show that the proposed framework enables up to a factor 20 in bandwidth reduction and factor 9 in power reduction with minimal loss (up to 2%) in classification accuracy compared to the centralized baseline on the demonstrated task. The proposed method offers a way to transform a centralized architecture to a distributed, bandwidth-efficient network amenable for low-power sensor networks. While the application focus of this paper is on wearable brain-computer interfaces, the proposed methodology can be applied in other sensor network-like applications as well.

Spatially Variant Ultrasound Attenuation Mapping Using a Regularized Linear Least-Squares Approach.

Quantitative ultrasound methods aim to estimate the acoustic properties of the underlying medium, such as the attenuation and backscatter coefficients, and have applications in various areas including tissue characterization. In practice, tissue heterogeneity makes the coefficient estimation challenging. In this work, we propose a computationally efficient algorithm to map spatial variations of the attenuation coefficient. Our proposed approach adopts a fast, linear least-squares strategy to fit the signal model to data from pulse-echo measurements. As opposed to existing approaches, we directly estimate the attenuation map, that is, the local attenuation coefficient at each axial location by solving a joint estimation problem. In particular, we impose a physical model that couples all these local estimates and combine it with a smooth regularization to obtain a smooth map. Compared to the conventional spectral log difference method and the more recent ALGEBRA approach, we demonstrate that the attenuation estimates obtained by our method are more accurate and better correlate with the ground-truth attenuation profiles over a wide range of spatial and contrast resolutions.

Time-Adaptive Unsupervised Auditory Attention Decoding Using EEG-Based Stimulus Reconstruction.

The goal of auditory attention decoding (AAD) is to determine to which speaker out of multiple competing speakers a listener is attending based on the brain signals recorded via, e.g., electroencephalography (EEG). AAD algorithms are a fundamental building block of so-called neuro-steered hearing devices that would allow identifying the speaker that should be amplified based on the brain activity. A common approach is to train a subject-specific stimulus decoder that reconstructs the amplitude envelope of the attended speech signal. However, training this decoder requires a dedicated 'ground-truth' EEG recording of the subject under test, during which the attended speaker is known. Furthermore, this decoder remains fixed during operation and can thus not adapt to changing conditions and situations. Therefore, we propose an online time-adaptive unsupervised stimulus reconstruction method that continuously and automatically adapts over time when new EEG and audio data are streaming in. The adaptive decoder does not require ground-truth attention labels obtained from a training session with the end-user and instead can be initialized with a generic subject-independent decoder or even completely random values. We propose two different implementations: a sliding window and recursive implementation, which we extensively validate on three independent datasets based on multiple performance metrics. We show that the proposed time-adaptive unsupervised decoder outperforms a time-invariant supervised decoder, representing an important step toward practically applicable AAD algorithms for neuro-steered hearing devices.

Objective evaluation of stimulation artefact removal techniques in the context of neural spike sorting.

Objective. We present a framework to objectively test and compare stimulation artefact removal techniques in the context of neural spike sorting.Approach. To this end, we used realistic hybrid ground-truth spiking data, with superimposed artefacts fromin vivorecordings. We used the framework to evaluate and compare several techniques: blanking, template subtraction by averaging, linear regression, and a multi-channel Wiener filter (MWF).Main results. Our study demonstrates that blanking and template subtraction result in a poorer spike sorting performance than linear regression and MWF, while the latter two perform similarly. Finally, to validate the conclusions found from the hybrid evaluation framework, we also performed a qualitative analysis onin vivorecordings without artificial manipulations.Significance. Our framework allows direct quantification of the impact of the residual artefact on the spike sorting accuracy, thereby allowing for a more objective and more relevant comparison compared to indirect signal quality metrics that are estimated from the signal statistics. Furthermore, the availability of a ground truth in the form of single-unit spiking activity also facilitates a better estimation of such signal quality metrics.

Electrocardiogram Quality Assessment Using Unsupervised Deep Learning.

OBJECTIVE: Noise and disturbances hinder effective interpretation of recorded ECG. To identify the clean parts of a recording, free from such disturbances, various quality indicators have been developed. Previous instances of these indicators focus on human-defined desirable properties of a clean signal. The reliance on human-specified properties places an inherent limitation on the potential power of signal quality indicators. To move away from this limitation, we propose a data-driven quality indicator. METHODS: We use an unsupervised deep learning model, the auto-encoder, to derive the quality indicator. For different quality assessment settings we compare the performance of our quality indicator with traditional indicators. RESULTS: The data-driven method performs consistently strong across tasks while performance of the traditional indicators varies strongly from task to task. CONCLUSION: This strong performance indicates the potential of data-driven quality indicators for use in ECG processing, removing the reliance on expert-specified desirable properties. SIGNIFICANCE: The proposed methodology can easily be extended towards learning quality indicators in other data modalities.

EEG miniaturization limits for stimulus decoding with EEG sensor networks.

Objective. Unobtrusive electroencephalography (EEG) monitoring in everyday life requires the availability of highly miniaturized EEG devices (mini-EEGs), which ideally consist of a wireless node with a small scalp area footprint, in which the electrodes, amplifier and wireless radio are embedded. By attaching a multitude of mini-EEGs at relevant positions on the scalp, a wireless 'EEG sensor network' (WESN) can be formed. However, each mini-EEG in the network only has access to its own local electrodes, thereby recording local scalp potentials with short inter-electrode distances. This is unlike using traditional cap-EEG, which by the virtue of re-referencing can measure EEG across arbitrarily large distances on the scalp. We evaluate the implications and limitations of such far-driven miniaturization on neural decoding performance.Approach. We collected 255-channel EEG data in an auditory attention decoding (AAD) task. As opposed to previous studies with a lower channel density, this new high-density dataset allows emulation of mini-EEGs with inter-electrode distances down to 1 cm in order to identify and quantify the lower bound on miniaturization for EEG-based stimulus decoding.Main results. We demonstrate that the performance remains reasonably stable for inter-electrode distances down to 3 cm, but decreases quickly for shorter distances if the mini-EEG nodes can be placed at optimal scalp locations and orientations selected by a data-driven algorithm.Significance. The results indicate the potential for the use of mini-EEGs in a WESN context for AAD applications and provide guidance on inter-electrode distances while designing such devices for neuro-steered hearing devices.

End-to-end learnable EEG channel selection for deep neural networks with Gumbel-softmax.

Objective.To develop an efficient, embedded electroencephalogram (EEG) channel selection approach for deep neural networks, allowing us to match the channel selection to the target model, while avoiding the large computational burdens of wrapper approaches in conjunction with neural networks.Approach.We employ a concrete selector layer to jointly optimize the EEG channel selection and network parameters. This layer uses a Gumbel-softmax trick to build continuous relaxations of the discrete parameters involved in the selection process, allowing them be learned in an end-to-end manner with traditional backpropagation. As the selection layer was often observed to include the same channel twice in a certain selection, we propose a regularization function to mitigate this behavior. We validate this method on two different EEG tasks: motor execution and auditory attention decoding. For each task, we compare the performance of the Gumbel-softmax method with a baseline EEG channel selection approach tailored towards this specific task: mutual information and greedy forward selection with the utility metric respectively.Main results.Our experiments show that the proposed framework is generally applicable, while performing at least as well as (and often better than) these state-of-the-art, task-specific approaches.Significance.The proposed method offers an efficient, task- and model-independent approach to jointly learn the optimal EEG channels along with the neural network weights.

Fast linear least-squares method for ultrasound attenuation and backscatter estimation.

The ultrasonic attenuation and backscatter coefficients of tissues are relevant acoustic parameters due to their wide range of clinical applications. In this paper, a linear least-squares method for the estimation of these coefficients in a homogeneous region of interest based on pulse-echo measurements is proposed. The method efficiently fits an ultrasound backscattered signal model to the measurements in both the frequency and depth dimension simultaneously at a low computational cost. It is demonstrated that the inclusion of depth information has a positive effect particularly on the accuracy of the estimated attenuation. The sensitivity of the attenuation and backscatter coefficients' estimates to several predefined parameters such as the window length, window overlap and usable bandwidth of the spectrum is also studied. Comparison of the proposed method with a benchmark approach based on dynamic programming highlights better performance of our method in estimating these coefficients, both in terms of accuracy and computation time. Further analysis of the computation time as a function of the predefined parameters indicates our method's potential to be used in real-time clinical settings.

Utility metric for unsupervised feature selection.

Feature selection techniques are very useful approaches for dimensionality reduction in data analysis. They provide interpretable results by reducing the dimensions of the data to a subset of the original set of features. When the data lack annotations, unsupervised feature selectors are required for their analysis. Several algorithms for this aim exist in the literature, but despite their large applicability, they can be very inaccessible or cumbersome to use, mainly due to the need for tuning non-intuitive parameters and the high computational demands. In this work, a publicly available ready-to-use unsupervised feature selector is proposed, with comparable results to the state-of-the-art at a much lower computational cost. The suggested approach belongs to the methods known as spectral feature selectors. These methods generally consist of two stages: manifold learning and subset selection. In the first stage, the underlying structures in the high-dimensional data are extracted, while in the second stage a subset of the features is selected to replicate these structures. This paper suggests two contributions to this field, related to each of the stages involved. In the manifold learning stage, the effect of non-linearities in the data is explored, making use of a radial basis function (RBF) kernel, for which an alternative solution for the estimation of the kernel parameter is presented for cases with high-dimensional data. Additionally, the use of a backwards greedy approach based on the least-squares utility metric for the subset selection stage is proposed. The combination of these new ingredients results in the utility metric for unsupervised feature selection U2FS algorithm. The proposed U2FS algorithm succeeds in selecting the correct features in a simulation environment. In addition, the performance of the method on benchmark datasets is comparable to the state-of-the-art, while requiring less computational time. Moreover, unlike the state-of-the-art, U2FS does not require any tuning of parameters.

EEG-based detection of the locus of auditory attention with convolutional neural networks.

In a multi-speaker scenario, the human auditory system is able to attend to one particular speaker of interest and ignore the others. It has been demonstrated that it is possible to use electroencephalography (EEG) signals to infer to which speaker someone is attending by relating the neural activity to the speech signals. However, classifying auditory attention within a short time interval remains the main challenge. We present a convolutional neural network-based approach to extract the locus of auditory attention (left/right) without knowledge of the speech envelopes. Our results show that it is possible to decode the locus of attention within 1-2 s, with a median accuracy of around 81%. These results are promising for neuro-steered noise suppression in hearing aids, in particular in scenarios where per-speaker envelopes are unavailable.

Unsupervised Self-Adaptive Auditory Attention Decoding.

When multiple speakers talk simultaneously, a hearing device cannot identify which of these speakers the listener intends to attend to. Auditory attention decoding (AAD) algorithms can provide this information by, for example, reconstructing the attended speech envelope from electroencephalography (EEG) signals. However, these stimulus reconstruction decoders are traditionally trained in a supervised manner, requiring a dedicated training stage during which the attended speaker is known. Pre-trained subject-independent decoders alleviate the need of having such a per-user training stage but perform substantially worse than supervised subject-specific decoders that are tailored to the user. This motivates the development of a new unsupervised self-adapting training/updating procedure for a subject-specific decoder, which iteratively improves itself on unlabeled EEG data using its own predicted labels. This iterative updating procedure enables a self-leveraging effect, of which we provide a mathematical analysis that reveals the underlying mechanics. The proposed unsupervised algorithm, starting from a random decoder, results in a decoder that outperforms a supervised subject-independent decoder. Starting from a subject-independent decoder, the unsupervised algorithm even closely approximates the performance of a supervised subject-specific decoder. The developed unsupervised AAD algorithm thus combines the two advantages of a supervised subject-specific and subject-independent decoder: it approximates the performance of the former while retaining the 'plug-and-play' character of the latter. As the proposed algorithm can be used to automatically adapt to new users, as well as over time when new EEG data is being recorded, it contributes to more practical neuro-steered hearing devices.

Modulating ultrasound contrast generation from injectable nanodroplets for proton range verification by varying the degree of superheat.

PURPOSE: Despite the physical benefits of protons over conventional photon radiation in cancer treatment, range uncertainties impede the ability to harness the full potential of proton therapy. While monitoring the proton range in vivo could reduce the currently adopted safety margins, a routinely applicable range verification technique is still lacking. Recently, phase-change nanodroplets were proposed for proton range verification, demonstrating a reproducible relationship between the proton range and generated ultrasound contrast after radiation-induced vaporization at 25°C. In this study, previous findings are extended with proton irradiations at different temperatures, including the physiological temperature of 37°C, for a novel nanodroplet formulation. Moreover, the potential to modulate the linear energy transfer (LET) threshold for vaporization by varying the degree of superheat is investigated, where the aim is to demonstrate vaporization of nanodroplets directly by primary protons. METHODS: Perfluorobutane nanodroplets with a shell made of polyvinyl alcohol (PVA-PFB) or 10,12-pentacosadyinoic acid (PCDA-PFB) were dispersed in polyacrylamide hydrogels and irradiated with 62 MeV passively scattered protons at temperatures of 37°C and 50°C. Nanodroplet transition into echogenic microbubbles was assessed using ultrasound imaging (gray value and attenuation analysis) and optical images. The proton range was measured independently and compared to the generated contrast. RESULTS: Nanodroplet design proved crucial to ensure thermal stability, as PVA-shelled nanodroplets dramatically outperformed their PCDA-shelled counterpart. At body temperature, a uniform radiation response proximal to the Bragg peak is attributed to nuclear reaction products interacting with PVA-PFB nanodroplets, with the 50% drop in ultrasound contrast being 0.17 mm ± 0.20 mm (mean ± standard deviation) in front of the proton range. Also at 50°C, highly reproducible ultrasound contrast profiles were obtained with shifts of -0.74 mm ± 0.09 mm (gray value analysis), -0.86 mm ± 0.04 mm (attenuation analysis) and -0.64 mm ± 0.29 mm (optical analysis). Moreover, a strong contrast enhancement was observed near the Bragg peak, suggesting that nanodroplets were sensitive to primary protons. CONCLUSIONS: By varying the degree of superheat of the nanodroplets' core, one can modulate the intensity of the generated ultrasound contrast. Moreover, a submillimeter reproducible relationship between the ultrasound contrast and the proton range was obtained, either indirectly via the visualization of secondary reaction products or directly through the detection of primary protons, depending on the degree of superheat. The potential of PVA-PFB nanodroplets for in vivo proton range verification was confirmed by observing a reproducible radiation response at physiological temperature, and further studies aim to assess the nanodroplets' performance in a physiological environment. Ultimately, cost-effective online or offline ultrasound imaging of radiation-induced nanodroplet vaporization could facilitate the reduction of safety margins in treatment planning and enable adaptive proton therapy.

Effect of number and placement of EEG electrodes on measurement of neural tracking of speech.

Measurement of neural tracking of natural running speech from the electroencephalogram (EEG) is an increasingly popular method in auditory neuroscience and has applications in audiology. The method involves decoding the envelope of the speech signal from the EEG signal, and calculating the correlation with the envelope of the audio stream that was presented to the subject. Typically EEG systems with 64 or more electrodes are used. However, in practical applications, set-ups with fewer electrodes are required. Here, we determine the optimal number of electrodes, and the best position to place a limited number of electrodes on the scalp. We propose a channel selection strategy based on an utility metric, which allows a quick quantitative assessment of the influence of a channel (or a group of channels) on the reconstruction error. We consider two use cases: a subject-specific case, where the optimal number and position of the electrodes is determined for each subject individually, and a subject-independent case, where the electrodes are placed at the same positions (in the 10-20 system) for all the subjects. We evaluated our approach using 64-channel EEG data from 90 subjects. In the subject-specific case we found that the correlation between actual and reconstructed envelope first increased with decreasing number of electrodes, with an optimum at around 20 electrodes, yielding 29% higher correlations using the optimal number of electrodes compared to all electrodes. This means that our strategy of removing electrodes can be used to improve the correlation metric in high-density EEG recordings. In the subject-independent case, we obtained a stable decoding performance when decreasing from 64 to 22 channels. When the number of channels was further decreased, the correlation decreased. For a maximal decrease in correlation of 10%, 32 well-placed electrodes were sufficient in 91% of the subjects.