Alpha-phase synchrony EEG training for multi-resistant chronic low back pain patients: an open-label pilot study.

PURPOSE: Chronic low back pain (cLBP) affects a quarter of a population during its lifetime. The most severe cases include patients not responding to interventions such as 5-week-long in-hospital multi-disciplinary protocols. This document reports on a pilot study offering an alpha-phase synchronization (APS) brain rehabilitation intervention to a population of n = 16 multi-resistant cLBP patients. METHODS: The intervention consists of 20 sessions of highly controlled electroencephalography (EEG) APS operant conditioning (neurofeedback) paradigm delivered in the form of visual feedback. Visual analogue scale for pain, Dallas, Hamilton, and HAD were measured before, after, at 6-month and 12-month follow-up. Full-scalp EEG data were analyzed to study significant changes in the brain's electrical activity. RESULTS: The intervention showed a great and lasting response of most measured clinical scales. The clinical improvement was lasting beyond the 6-month follow-up endpoints. The EEG data confirm that patients did control (intra-session trends) and learned to better control (intersession trends) their APS neuromarker resulting in (nonsignificant) baseline changes in their resting state activity. Last and most significantly, the alpha-phase concentration (APC) neuromarker, specific to phase rather than amplitude, was found to correlate significantly with the reduction in clinical symptoms in a typical dose-response effect. CONCLUSION: This first experiment highlights the role of the APC neuromarker in relation to the nucleus accumbens activity and its role on nociception and the chronicity of pain. This study suggests APC rehabilitation could be used clinically for the most severe cases of cLBP. Its excellent safety profile and availability as a home-use intervention makes it a potentially disruptive tool in the context of nonsteroidal anti-inflammatory drugs and opioid abuses. These slides can be retrieved under Electronic Supplementary Material.

An Unsupervised Multichannel Artifact Detection Method for Sleep EEG Based on Riemannian Geometry.

In biomedical signal processing, we often face the problem of artifacts that distort the original signals. This concerns also sleep recordings, such as EEG. Artifacts may severely affect or even make impossible visual inspection, as well as automatic processing. Many proposed methods concentrate on certain artifact types. Therefore, artifact-free data are often obtained after sequential application of different methods. Moreover, single-channel approaches must be applied to all channels alternately. The aim of this study is to develop a multichannel artifact detection method for multichannel sleep EEG capable of rejecting different artifact types at once. The inspiration for the study is gained from recent advances in the field of Riemannian geometry. The method we propose is tested on real datasets. The performance of the proposed method is measured by comparing detection results with the expert labeling as a reference and evaluated against a simpler method based on Riemannian geometry that has previously been proposed, as well as against the state-of-the-art method FASTER. The obtained results prove the effectiveness of the proposed method.

Pinpointing a highly specific pathological functional connection that turns phantom sound into distress.

It has been suggested that an auditory phantom percept is the result of multiple, parallel but overlapping networks. One of those networks encodes tinnitus loudness and is electrophysiologically separable from a nonspecific distress network. The present study investigates how these networks anatomically overlap, what networks are involved, and how and when these networks interact. Electroencephalography data of 317 tinnitus patients and 256 healthy subjects were analyzed, using independent component analysis. Results demonstrate that tinnitus is characterized by at least 2 major brain networks, each consisting of multiple independent components. One network reflects tinnitus distress, while another network reflects the loudness of the tinnitus. The component coherence analysis shows that the independent components that make up the distress and loudness networks communicate within their respective network at several discrete frequencies in parallel. The distress and loudness networks do not intercommunicate for patients without distress, but do when patients are distressed by their tinnitus. The obtained data demonstrate that the components that build up these 2 separable networks communicate at discrete frequencies within the network, and only between the distress and loudness networks in those patients in whom the symptoms are also clinically linked.

Beta bursts question the ruling power for brain-computer interfaces.

Objective: Current efforts to build reliable brain-computer interfaces (BCI) span multiple axes from hardware, to software, to more sophisticated experimental protocols, and personalized approaches. However, despite these abundant efforts, there is still room for significant improvement. We argue that a rather overlooked direction lies in linking BCI protocols with recent advances in fundamental neuroscience.Approach: In light of these advances, and particularly the characterization of the burst-like nature of beta frequency band activity and the diversity of beta bursts, we revisit the role of beta activity in 'left vs. right hand' motor imagery (MI) tasks. Current decoding approaches for such tasks take advantage of the fact that MI generates time-locked changes in induced power in the sensorimotor cortex and rely on band-passed power changes in single or multiple channels. Although little is known about the dynamics of beta burst activity during MI, we hypothesized that beta bursts should be modulated in a way analogous to their activity during performance of real upper limb movements.Main results and Significance: We show that classification features based on patterns of beta burst modulations yield decoding results that are equivalent to or better than typically used beta power across multiple open electroencephalography datasets, thus providing insights into the specificity of these bio-markers.

Prediction of treatment response and the effect of independent component neurofeedback in obsessive-compulsive disorder: a randomized, sham-controlled, double-blind study.

AIMS: The goal of this study was to assess the effect of independent component neurofeedback (NFB) on EEG and clinical symptoms in patients with obsessive-compulsive disorder (OCD). Subsequently, we explored predictors of treatment response and EEG correlates of clinical symptoms. METHODS: In a randomized, double-blind, parallel design, 20 inpatients with OCD underwent 25 sessions of NFB or sham feedback (SFB). NFB aimed at reducing EEG activity in an independent component previously reported abnormal in this diagnosis. Resting-state EEG recorded before and after the treatment was analyzed to assess its posttreatment changes, relationships with clinical symptoms and treatment response. RESULTS: Overall, clinical improvement in OCD patients was not accompanied by EEG change as assessed by standardized low-resolution electromagnetic tomography and normative independent component analysis. Pre- to posttreatment comparison of the trained component and frequency did not yield significant results; however, in the NFB group, the nominal values at the downtrained frequency were lower after treatment. The NFB group showed significantly higher percentage reduction of compulsions compared to the SFB group (p = 0.015). Pretreatment higher amount of delta (1-6 Hz) and low alpha oscillations as well as a lower amount of high beta activity predicted a worse treatment outcome. Source localization of these delta and high beta oscillations corresponded with previous EEG resting-state findings in OCD patients compared to healthy controls. CONCLUSION: Independent component NFB in OCD proved useful in percentage improvement of compulsions. Based on our correlation analyses, we hypothesize that we targeted a network related to treatment resistance.

Transfer Learning for P300 Brain-Computer Interfaces by Joint Alignment of Feature Vectors.

This article presents a new transfer learning method named group learning, that jointly aligns multiple domains (many-to-many) and an extension named fast alignment that aligns any further domain to previously aligned group of domains (many-to-one). The proposed group alignment algorithm (GALIA) is evaluated on brain-computer interface (BCI) data and optimal hyper-parameter values of the algorithm are studied for classification performance and computational cost. Six publicly available P300 databases comprising 333 sessions from 177 subjects are used. As compared to the conventional subject-specific train/test pipeline, both group learning and fast alignment significantly improve the classification accuracy except for the database with clinical subjects (average improvement: 2.12±1.88%). GALIA utilizes cyclic approximate joint diagonalization (AJD) to find a set of linear transformations, one for each domain, jointly aligning the feature vectors of all domains. Group learning achieves a many-to-many transfer learning without compromising the classification performance on non-clinical BCI data. Fast alignment further extends the group learning for any unseen domains, allowing a many-to-one transfer learning with the same properties. The former method creates a single machine learning model using data from previous subjects and/or sessions, whereas the latter exploits the trained model for an unseen domain requiring no further training of the classifier.

Comparing Clustering Methods Applied to Tinnitus within a Bootstrapped and Diagnostic-Driven Semi-Supervised Framework.

The understanding of tinnitus has always been elusive and is largely prevented by its intrinsic heterogeneity. To address this issue, scientific research has aimed at defining stable and easily identifiable subphenotypes of tinnitus. This would allow better disentangling the multiple underlying pathophysiological mechanisms of tinnitus. In this study, three-dimensionality reduction techniques and two clustering methods were benchmarked on a database of 2772 tinnitus patients in order to obtain a reliable segmentation of subphenotypes. In this database, tinnitus patients' endotypes (i.e., parts of a population with a condition with distinct underlying mechanisms) are reported when diagnosed by an ENT expert in tinnitus management. This partial labeling of the dataset enabled the design of an original semi-supervised framework. The objective was to perform a benchmark of different clustering methods to get as close as possible to the initial ENT expert endotypes. To do so, two metrics were used: a primary one, the quality of the separation of the endotypes already identified in the database, as well as a secondary one, the stability of the obtained clusterings. The relevance of the results was finally reviewed by two ENT experts in tinnitus management. A 20-cluster clustering was selected as the best-performing, the most-clinically relevant, and the most-stable through bootstrapping. This clustering used a T-SNE method as the dimensionality reduction technique and a k-means algorithm as the clustering method. The characteristics of this clustering are presented in this article.

REM Sleep Impairment May Underlie Sleep-Driven Modulations of Tinnitus in Sleep Intermittent Tinnitus Subjects: A Controlled Study.

(1) Background: Poor sleep and fragmented sleep are associated with several chronic conditions. Tinnitus is an auditory symptom that often negatively combines with poor sleep and has been associated with sleep impairment and sleep apnea. The relationship between tinnitus psychoacoustic characteristics and sleep is still poorly explored, notably for a particular subgroup of patients, for whom the perceived loudness of their tinnitus is highly modulated by sleep. (2) Methods: For this observational prospective study, 30 subjects with tinnitus were recruited, including 15 "sleep intermittent tinnitus" subjects, who had reported significant modulations of tinnitus loudness related to night sleep and naps, and a control group of 15 subjects displaying constant non-sleep-modulated tinnitus. The control group had matching age, gender, self-reported hearing loss grade and tinnitus impact on quality of life with the study group. All patients underwent a polysomnography (PSG) assessment for one complete night and then were asked to fill in a case report form, as well as a report of tinnitus loudness before and after the PSG. (3) Results: "Sleep Intermittent tinnitus" subjects had less Stage 3 sleep (p < 0.01), less Rapid-Eye Movement (REM) Sleep (p < 0.05) and more Stage 2 sleep (p < 0.05) in proportion and duration than subjects from the control group. In addition, in the "sleep Intermittent tinnitus" sample, a correlation was found between REM sleep duration and tinnitus overnight modulation (p < 0.05), as well as tinnitus impact on quality of life (p < 0.05). These correlations were not present in the control group. (4) Conclusions: This study suggests that among the tinnitus population, patients displaying sleep-modulated tinnitus have deteriorated sleep quality. Furthermore, REM sleep characteristics may play a role in overnight tinnitus modulation. Potential pathophysiological explanations accounting for this observation are hypothesized and discussed.

Brain oscillatory activity during spatial navigation: theta and gamma activity link medial temporal and parietal regions.

Brain oscillatory correlates of spatial navigation were investigated using blind source separation (BSS) and standardized low resolution electromagnetic tomography (sLORETA) analyses of 62-channel EEG recordings. Twenty-five participants were instructed to navigate to distinct landmark buildings in a previously learned virtual reality town environment. Data from periods of navigation between landmarks were subject to BSS analyses to obtain source components. Two of these cortical sources were found to exhibit significant spectral power differences during navigation with respect to a resting eyes open condition and were subject to source localization using sLORETA. These two sources were localized as a right parietal component with gamma activation and a right medial-temporal-parietal component with activation in theta and gamma bandwidths. The parietal gamma activity was thought to reflect visuospatial processing associated with the task. The medial-temporal-parietal activity was thought to be more specific to the navigational processing, representing the integration of ego- and allo-centric representations of space required for successful navigation, suggesting theta and gamma oscillations may have a role in integrating information from parietal and medial-temporal regions. Theta activity on this medial-temporal-parietal source was positively correlated with more efficient navigation performance. Results are discussed in light of the depth and proposed closed field structure of the hippocampus and potential implications for scalp EEG data. The findings of the present study suggest that appropriate BSS methods are ideally suited to minimizing the effects of volume conduction in noninvasive recordings, allowing more accurate exploration of deep brain processes.

Tangent space alignment: Transfer learning for Brain-Computer Interface.

Statistical variability of electroencephalography (EEG) between subjects and between sessions is a common problem faced in the field of Brain-Computer Interface (BCI). Such variability prevents the usage of pre-trained machine learning models and requires the use of a calibration for every new session. This paper presents a new transfer learning (TL) method that deals with this variability. This method aims to reduce calibration time and even improve accuracy of BCI systems by aligning EEG data from one subject to the other in the tangent space of the positive definite matrices Riemannian manifold. We tested the method on 18 BCI databases comprising a total of 349 subjects pertaining to three BCI paradigms, namely, event related potentials (ERP), motor imagery (MI), and steady state visually evoked potentials (SSVEP). We employ a support vector classifier for feature classification. The results demonstrate a significant improvement of classification accuracy, as compared to a classical training-test pipeline, in the case of the ERP paradigm, whereas for both the MI and SSVEP paradigm no deterioration of performance is observed. A global 2.7% accuracy improvement is obtained compared to a previously published Riemannian method, Riemannian Procrustes Analysis (RPA). Interestingly, tangent space alignment has an intrinsic ability to deal with transfer learning for sets of data that have different number of channels, naturally applying to inter-dataset transfer learning.

Dimensionality Transcending: A Method for Merging BCI Datasets With Different Dimensionalities.

OBJECTIVE: We present a transfer learning method for datasets with different dimensionalities, coming from different experimental setups but representing the same physical phenomena. We focus on the case where the data points are symmetric positive definite (SPD) matrices describing the statistical behavior of EEG-based brain computer interfaces (BCI). METHOD: Our proposal uses a two-step procedure that transforms the data points so that they become matched in terms of dimensionality and statistical distribution. In the dimensionality matching step, we use isometric transformations to map each dataset into a common space without changing their geometric structures. The statistical matching is done using a domain adaptation technique adapted for the intrinsic geometry of the space where the datasets are defined. RESULTS: We illustrate our proposal on time series obtained from BCI systems with different experimental setups (e.g., different number of electrodes, different placement of electrodes). The results show that the proposed method can be used to transfer discriminative information between BCI recordings that, in principle, would be incompatible. CONCLUSION AND SIGNIFICANCE: Such findings pave the way to a new generation of BCI systems capable of reusing information and learning from several sources of data despite differences in their electrodes positioning.

A portable neurofeedback device for treating chronic subjective tinnitus: Feasibility and results of a pilot study.

BACKGROUND: Several clinical studies have shown that neurofeedback (NFB) has the potential to significantly improve the quality of life of patients complaining of chronic subjective tinnitus. Yet the clinical applicability of such a therapeutic approach in the everyday practice has not been tested so far. OBJECTIVE: This study aims at investigating the feasibility and efficacy of a semi-automated NFB intervention by means of a portable device that eventually could be used by the patients at home on an everyday basis. The duration of setup procedures is minimized through the use of a dry electrodes electroencephalography (EEG) headset and an automated user-interface. METHODS: We conducted a pilot clinical study (non-controlled, single arm, NCT03773926). According to a predetermined power calculation, a homogeneous population of 33 subjects with strict inclusion criteria was enrolled. After inclusion, all patients underwent 10 NFB sessions lasting 50min each, over a period of 5 weeks and a 3-month follow-up period. According to previous studies, the NFB training aimed at increasing the alpha-band power (8-12Hz) in the EEG power spectrum on the averaged signal of leads FC1, FC2, F3 and F4. Tinnitus handicap inventory (THI) was used as a primary outcome measure. Secondary outcome measures were the visual analog scales (VAS) and the change of the alpha-band power within sessions and across training. Time points of assessment were before intervention (T1), after intervention (T2) and at the 3-month follow-up (T3). RESULTS: Patient exhibited a clinically significant decrease of the THI score, with a 23% decrease (N=28) on average between T1 and T2 and a 31% decrease (N=25) between T1 and T3. A significant increase of the alpha-band power within sessions was observed. No significant increase of the alpha-band power across sessions was observed. For the 19 subjects where sufficient data were exploitable, a significant correlation was found between the evolution of the alpha-band training across sessions and the evolution of the THI between T1 and T2. The sessions were well tolerated and no adverse effect was reported. CONCLUSION: This study suggests that neurofeedback has potential to suit everyday clinical practice with the goal to significantly reduce tinnitus intrusiveness. The merits and limitations of this NFB procedure are discussed, especially with respect to the choice of EEG electrodes to ensure a good signal quality.

The neural correlates of tinnitus-related distress.

Tinnitus is an auditory phantom percept with a tone, hissing, or buzzing sound in the absence of any objective physical sound source. About 6% to 25% of the affected people report interference with their lives as tinnitus causes a considerable amount of distress. However, the underlying neurophysiological mechanism for the development of tinnitus-related distress remains not well understood. Hence we focus on the cortical and subcortical source differences in resting-state EEG between tinnitus patients with different grades of distress using continuous scalp EEG recordings and Low Resolution Electromagnetic Tomography (LORETA). Results show more synchronized alpha activity in the tinnitus patients with a serious amount of distress with peaks localized to various emotion-related areas. These areas include subcallosal anterior cingulate cortex, the insula, parahippocampal area and amygdala. In addition, less alpha synchronized activity was found in the posterior cingulate cortex, precuneus and DLPFC. A comparison between the tinnitus group with distress and the Nova Tech EEG (NTE) normative database demonstrated increased synchronized alpha and beta activity and less synchronized delta and theta activity in the dorsal anterior cingulate cortex in tinnitus patients with distress. It is interesting that the areas found show some overlap with the emotional component of the pain matrix and the distress related areas in asthmatic dyspnea. Unpleasantness of pain activates the anterior cingulate and prefrontal cortices, amygdala, and insula. As such, it might be that distress is related to alpha and beta activity in the dorsal anterior cingulate cortex, the amount of distress perceived to an alpha network consisting of the amygdala-anterior cingulate cortex-insula-parahippocampal area.

On the "dependence" of "independent" group EEG sources; an EEG study on two large databases.

The aim of this work is to study the coherence profile (dependence) of robust eyes-closed resting EEG sources isolated by group blind source separation (gBSS). We employ a test-retest strategy using two large sample normative databases (N = 57 and 84). Using a BSS method in the complex Fourier domain, we show that we can rigourously study the out-of-phase dependence of the extracted components, albeit they are extracted so as to be in-phase independent (by BSS definition). Our focus on lagged communication between components effectively yields dependence measures unbiased by volume conduction effects, which is a major concern about the validity of any dependence measures issued by EEG measurements. We are able to show the organization of the extracted components in two networks. Within each network components oscillate coherently with multiple-frequency dynamics, whereas between networks they exchange information at non-random multiple time-lag rates.

Differentiating a network of executive attention: LORETA neurofeedback in anterior cingulate and dorsolateral prefrontal cortices.

INTRODUCTION: This study examines the differential effects of space-specific neuro-operant learning, utilizing low-resolution electromagnetic tomographic (LORETA) neurofeedback in three regions of training (ROTs), namely, the anterior cingulate gyrus (AC) and right and left dorsolateral prefrontal cortices (RPFC and LPFC respectively). METHODS: This study was conducted with 14 nonclinical students with a mean age of 22. We utilized electrophysiological measurements and subtests of the WAIS-III for premeasures and postmeasures. RESULTS: The data indicate that the AC shares a significant association with the RPFC and LPFC; however, each of the ROTs exhibits different cortical effects in all frequencies when trained exclusively. DISCUSSION: LORETA neurofeedback (LNFB) appears to enhance the functioning and strengthening of networks of cortical units physiologically related to each ROT; moreover, significant changes are mapped for each frequency domain, showing the associations within this possible attentional network.

The Riemannian Potato Field: A Tool for Online Signal Quality Index of EEG.

Electroencephalographic (EEG) recordings are contaminated by instrumental, environmental, and biological artifacts, resulting in low signal-to-noise ratio. Artifact detection is a critical task for real-time applications where the signal is used to give a continuous feedback to the user. In these applications, it is therefore necessary to estimate online a signal quality index (SQI) in order to stop the feedback when the signal quality is unacceptable. In this paper, we introduce the Riemannian potato field (RPF) algorithm as such SQI. It is a generalization and extensionof theRiemannian potato, a previouslypublished real-time artifact detection algorithm, whose performance is degraded as the number of channels increases. The RPF overcomes this limitation by combining the outputs of several smaller potatoes into a unique SQI resulting in a higher sensitivity and specificity, regardless of the number of electrodes. We demonstrate these results on a clinical dataset totalizing more than 2200 h of EEG recorded at home, that is, in a non-controlled environment.

Riemannian Procrustes Analysis: Transfer Learning for Brain-Computer Interfaces.

OBJECTIVE: This paper presents a Transfer Learning approach for dealing with the statistical variability of electroencephalographic (EEG) signals recorded on different sessions and/or from different subjects. This is a common problem faced by brain-computer interfaces (BCI) and poses a challenge for systems that try to reuse data from previous recordings to avoid a calibration phase for new users or new sessions for the same user. METHOD: We propose a method based on Procrustes analysis for matching the statistical distributions of two datasets using simple geometrical transformations (translation, scaling, and rotation) over the data points. We use symmetric positive definite matrices (SPD) as statistical features for describing the EEG signals, so the geometrical operations on the data points respect the intrinsic geometry of the SPD manifold. Because of its geometry-aware nature, we call our method the Riemannian Procrustes analysis (RPA). We assess the improvement in transfer learning via RPA by performing classification tasks on simulated data and on eight publicly available BCI datasets covering three experimental paradigms (243 subjects in total). RESULTS: Our results show that the classification accuracy with RPA is superior in comparison to other geometry-aware methods proposed in the literature. We also observe improvements in ensemble classification strategies when the statistics of the datasets are matched via RPA. CONCLUSION AND SIGNIFICANCE: We present a simple yet powerful method for matching the statistical distributions of two datasets, thus paving the way to BCI systems capable of reusing data from previous sessions and avoid the need of a calibration procedure.

Clinical and Experimental Factors Influencing the Efficacy of Neurofeedback in ADHD: A Meta-Analysis.

Meta-analyses have been extensively used to evaluate the efficacy of neurofeedback (NFB) treatment for Attention Deficit/Hyperactivity Disorder (ADHD) in children and adolescents. However, each meta-analysis published in the past decade has contradicted the methods and results from the previous one, thus making it difficult to determine a consensus of opinion on the effectiveness of NFB. This works brings continuity to the field by extending and discussing the last and much controversial meta-analysis by Cortese et al. (1). The extension comprises an update of that work including the latest control trials, which have since been published and, most importantly, offers a novel methodology. Specifically, NFB literature is characterized by a high technical and methodological heterogeneity, which partly explains the current lack of consensus on the efficacy of NFB. This work takes advantage of this by performing a Systematic Analysis of Biases (SAOB) in studies included in the previous meta-analysis. Our extended meta-analysis (k = 16 studies) confirmed the previously obtained results of effect sizes in favor of NFB efficacy as being significant when clinical scales of ADHD are rated by parents (non-blind, p-value = 0.0014), but not when they are rated by teachers (probably blind, p-value = 0.27). The effect size is significant according to both raters for the subset of studies meeting the definition of "standard NFB protocols" (parents' p-value = 0.0054; teachers' p-value = 0.043, k = 4). Following this, the SAOB performed on k = 33 trials identified three main factors that have an impact on NFB efficacy: first, a more intensive treatment, but not treatment duration, is associated with higher efficacy; second, teachers report a lower improvement compared to parents; third, using high-quality EEG equipment improves the effectiveness of the NFB treatment. The identification of biases relating to an appropriate technical implementation of NFB certainly supports the efficacy of NFB as an intervention. The data presented also suggest that the probably blind assessment of teachers may not be considered a good proxy for blind assessments, therefore stressing the need for studies with placebo-controlled intervention as well as carefully reported neuromarker changes in relation to clinical response.

Transfer Learning: A Riemannian Geometry Framework With Applications to Brain-Computer Interfaces.

OBJECTIVE: This paper tackles the problem of transfer learning in the context of electroencephalogram (EEG)-based brain-computer interface (BCI) classification. In particular, the problems of cross-session and cross-subject classification are considered. These problems concern the ability to use data from previous sessions or from a database of past users to calibrate and initialize the classifier, allowing a calibration-less BCI mode of operation. METHODS: Data are represented using spatial covariance matrices of the EEG signals, exploiting the recent successful techniques based on the Riemannian geometry of the manifold of symmetric positive definite (SPD) matrices. Cross-session and cross-subject classification can be difficult, due to the many changes intervening between sessions and between subjects, including physiological, environmental, as well as instrumental changes. Here, we propose to affine transform the covariance matrices of every session/subject in order to center them with respect to a reference covariance matrix, making data from different sessions/subjects comparable. Then, classification is performed both using a standard minimum distance to mean classifier, and through a probabilistic classifier recently developed in the literature, based on a density function (mixture of Riemannian Gaussian distributions) defined on the SPD manifold. RESULTS: The improvements in terms of classification performances achieved by introducing the affine transformation are documented with the analysis of two BCI datasets. CONCLUSION AND SIGNIFICANCE: Hence, we make, through the affine transformation proposed, data from different sessions and subject comparable, providing a significant improvement in the BCI transfer learning problem.