Assessing consciousness in patients with disorders of consciousness using soft-clustering.

Consciousness is something we experience in our everyday life, more especially between the time we wake up in the morning and go to sleep at night, but also during the rapid eye movement (REM) sleep stage. Disorders of consciousness (DoC) are states in which a person's consciousness is damaged, possibly after a traumatic brain injury. Completely locked-in syndrome (CLIS) patients, on the other hand, display covert states of consciousness. Although they appear unconscious, their cognitive functions are mostly intact. Only, they cannot externally display it due to their quadriplegia and inability to speak. Determining these patients' states constitutes a challenging task. The ultimate goal of the approach presented in this paper is to assess these CLIS patients consciousness states. EEG data from DoC patients are used here first, under the assumption that if the proposed approach is able to accurately assess their consciousness states, it will assuredly do so on CLIS patients too. This method combines different sets of features consisting of spectral, complexity and connectivity measures in order to increase the probability of correctly estimating their consciousness levels. The obtained results showed that the proposed approach was able to correctly estimate several DoC patients' consciousness levels. This estimation is intended as a step prior attempting to communicate with them, in order to maximise the efficiency of brain-computer interfaces (BCI)-based communication systems.

Application of Soft-Clustering to Assess Consciousness in a CLIS Patient.

Completely locked-in (CLIS) patients are characterized by sufficiently intact cognitive functions, but a complete paralysis that prevents them to interact with their surroundings. On one hand, studies have shown that the ability to communicate plays an important part in these patients' quality of life and prognosis. On the other hand, brain-computer interfaces (BCIs) provide a means for them to communicate using their brain signals. However, one major problem for such patients is the difficulty to determine if they are conscious or not at a specific time. This work aims to combine different sets of features consisting of spectral, complexity and connectivity measures, to increase the probability of correctly estimating CLIS patients' consciousness levels. The proposed approach was tested on data from one CLIS patient, which is particular in the sense that the experimenter was able to point out one time frame Δt during which he was undoubtedly conscious. Results showed that the method presented in this paper was able to detect increases and decreases of the patient's consciousness levels. More specifically, increases were observed during this Δt, corroborating the assertion of the experimenter reporting that the patient was definitely conscious then. Assessing the patients' consciousness is intended as a step prior attempting to communicate with them, in order to maximize the efficiency of BCI-based communication systems.

Yes/No Classification of EEG data from CLIS patients.

The goal of this research is to evaluate the usability of new features to classify EEG data from several completely locked-in patients (CLIS), and eventually build a more reliable communication system for them. Patients in such state are completely paralyzed, preventing them to be able to talk, but they retain their cognitive abilities.The data were obtained from four CLIS patients and recorded during an auditory paradigm task during which they were asked yes/no questions. Spectral measures such as the relative power of δ, θ, α, ß and γ frequency bands, spectral edge frequencies (SEF50 and SEF95), complexity measure obtained from Poincaré plots and connectivity measures such as the imaginary part of coherency and the weighted Symbolic Mutual Information (wSMI) were used as features. The data was classified using Random Forest and Support Vector Machine, two methods successfully used to classify mental states in both healthy subjects and patients. Additionally, two cases were studied. The first case uses data recorded when the patient is answering questions, while in the second case it also includes data recorded when the experimenter is asking the questions.The classification accuracy during training varies between 51.73 to 67.72% in the first case, and from 50.41 to 67.94% for the second case. Overall, wSMI with a time lag of 64 ms gave the best classification accuracy and in general, Random Forest appears to be the best classification method.Clinical relevance This case study investigates the usability of new features based on EEG complexity and connectivity to classify CLIS patients brain signal, what results in a further step toward the demand of more effective EEG-based Brain-Computer Interface communication systems for CLIS patients.

Support Vector Machine Classifiers Show High Generalizability in Automatic Fall Detection in Older Adults.

Falls are a major cause of morbidity and mortality in neurological disorders. Technical means of detecting falls are of high interest as they enable rapid notification of caregivers and emergency services. Such approaches must reliably differentiate between normal daily activities and fall events. A promising technique might be based on the classification of movements based on accelerometer signals by machine-learning algorithms, but the generalizability of classifiers trained on laboratory data to real-world datasets is a common issue. Here, three machine-learning algorithms including Support Vector Machine (SVM), k-Nearest Neighbors (kNN), and Random Forest (RF) were trained to detect fall events. We used a dataset containing intentional falls (SisFall) to train the classifier and validated the approach on a different dataset which included real-world accidental fall events of elderly people (FARSEEING). The results suggested that the linear SVM was the most suitable classifier in this cross-dataset validation approach and reliably distinguished a fall event from normal everyday activity at an accuracy of 93% and similarly high sensitivity and specificity. Thus, classifiers based on linear SVM might be useful for automatic fall detection in real-world applications.

Evaluation of Three Machine Learning Algorithms for the Automatic Classification of EMG Patterns in Gait Disorders.

Gait disorders are common in neurodegenerative diseases and distinguishing between seemingly similar kinematic patterns associated with different pathological entities is a challenge even for the experienced clinician. Ultimately, muscle activity underlies the generation of kinematic patterns. Therefore, one possible way to address this problem may be to differentiate gait disorders by analyzing intrinsic features of muscle activations patterns. Here, we examined whether it is possible to differentiate electromyography (EMG) gait patterns of healthy subjects and patients with different gait disorders using machine learning techniques. Nineteen healthy volunteers (9 male, 10 female, age 28.2 ± 6.2 years) and 18 patients with gait disorders (10 male, 8 female, age 66.2 ± 14.7 years) resulting from different neurological diseases walked down a hallway 10 times at a convenient pace while their muscle activity was recorded via surface EMG electrodes attached to 5 muscles of each leg (10 channels in total). Gait disorders were classified as predominantly hypokinetic (n = 12) or ataxic (n = 6) gait by two experienced raters based on video recordings. Three different classification methods (Convolutional Neural Network-CNN, Support Vector Machine-SVM, K-Nearest Neighbors-KNN) were used to automatically classify EMG patterns according to the underlying gait disorder and differentiate patients and healthy participants. Using a leave-one-out approach for training and evaluating the classifiers, the automatic classification of normal and abnormal EMG patterns during gait (2 classes: "healthy" and "patient") was possible with a high degree of accuracy using CNN (accuracy 91.9%), but not SVM (accuracy 67.6%) or KNN (accuracy 48.7%). For classification of hypokinetic vs. ataxic vs. normal gait (3 classes) best results were again obtained for CNN (accuracy 83.8%) while SVM and KNN performed worse (accuracy SVM 51.4%, KNN 32.4%). These results suggest that machine learning methods are useful for distinguishing individuals with gait disorders from healthy controls and may help classification with respect to the underlying disorder even when classifiers are trained on comparably small cohorts. In our study, CNN achieved higher accuracy than SVM and KNN and may constitute a promising method for further investigation.

Leveraging data and AI to deliver on the promise of digital health.

Rising rates of NCDs threaten fragile healthcare systems in low- and middle-income countries. Fortunately, new digital technology provides tools to more effectively address the growing dual burden of disease. Two-thirds of the world's population subscribed to mobile services by the end of 2018, while the falling price of connectivity and the 5G networks rollout promise to accelerate the use of digital technology. Properly leveraged, we can employ digital solutions and applications to transform health systems from reactive to proactive and even preventive, helping people stay healthy. With artificial intelligence (AI), health systems can be made more predictive by detecting risk factors and helping health professionals respond faster to prevent disease. Yet this rapid pace of growth has also complicated the digital health landscape. Myriad digital health apps compete and overlap in the public and private sectors, and significant gaps in the collection and analysis of digital data threaten to leave some behind. Established in 2010, the Broadband Commission for Sustainable Development is led by ITU and UNESCO and advocates for the transformational impact of broadband technologies for development. Its working group on digital and AI in health, co-chaired by the Novartis Foundation and at different times Nokia, Intel and Microsoft, identifies best practices for countries to realize the potential of digital technology in health and care. Interviewing more than 100 key stakeholders and reviewing over 200 documents, the Working Group set out to identify common challenges that countries face in implementing digital health solutions, and to develop a framework that countries can use to build systems for supporting digital health solutions. Common challenges include a lack of coordination leading to fragmented digital health solutions; lack of systems and workforce capacity to manage data and digital technology, and inadequate financing to support digital health. The working group proposes six building blocks for digital health systems: formulate and execute a national digital health strategy; create policy and regulatory frameworks that support innovation while protecting security and privacy; ensure access to digital infrastructure; ensure interoperability of digital health system components; establish effective partnerships; and sustain adequate financing.

Consciousness Detection in a Complete Locked-in Syndrome Patient through Multiscale Approach Analysis.

Completely locked-in state (CLIS) patients are unable to speak and have lost all muscle movement. From the external view, the internal brain activity of such patients cannot be easily perceived, but CLIS patients are considered to still be conscious and cognitively active. Detecting the current state of consciousness of CLIS patients is non-trivial, and it is difficult to ascertain whether CLIS patients are conscious or not. Thus, it is important to find alternative ways to re-establish communication with these patients during periods of awareness, and one such alternative is through a brain-computer interface (BCI). In this study, multiscale-based methods (multiscale sample entropy, multiscale permutation entropy and multiscale Poincaré plots) were applied to analyze electrocorticogram signals from a CLIS patient to detect the underlying consciousness level. Results from these different methods converge to a specific period of awareness of the CLIS patient in question, coinciding with the period during which the CLIS patient is recorded to have communicated with an experimenter. The aim of the investigation is to propose a methodology that could be used to create reliable communication with CLIS patients.

Transient Amplitude Modulation of Alpha-Band Oscillations by Short-Time Intermittent Closed-Loop tACS.

Non-invasive brain stimulation (NIBS) techniques such as transcranial alternating current stimulation (tACS) have recently become extensively utilized due to their potential to modulate ongoing neuronal oscillatory activity and consequently to induce cortical plasticity relevant for various cognitive functions. However, the neurophysiological basis for stimulation effects as well as their inter-individual differences is not yet understood. In the present study, we used a closed-loop electroencephalography-tACS(EEG-tACS) protocol to examine the modulation of alpha oscillations generated in occipito-parietal areas. In particular, we investigated the effects of a repeated short-time intermittent stimulation protocol (1 s in every trial) applied over the visual cortex (Cz and Oz) and adjusted according to the phase and frequency of visual alpha oscillations on the amplitude of these oscillations. Based on previous findings, we expected higher increases in alpha amplitudes for tACS applied in-phase with ongoing oscillations as compared to an application in anti-phase and this modulation to be present in low-alpha amplitude states of the visual system (eyes opened, EO) but not high (eyes closed, EC). Contrary to our expectations, we found a transient suppression of alpha power in inter-individually derived spatially specific parieto-occipital components obtained via the estimation of spatial filters by using the common spatial patterns approach. The amplitude modulation was independent of the phase relationship between the tACS signal and alpha oscillations, and the state of the visual system manipulated via closed- and open-eye conditions. It was also absent in conventionally analyzed single-channel and multi-channel data from an average parieto-occipital region. The fact that the tACS modulation of oscillations was phase-independent suggests that mechanisms driving the effects of tACS may not be explained by entrainment alone, but rather require neuroplastic changes or transient disruption of neural oscillations. Our study also supports the notion that the response to tACS is subject-specific, where the modulatory effects are shaped by the interplay between the stimulation and different alpha generators. This favors stimulation protocols as well as analysis regimes exploiting inter-individual differences, such as spatial filters to reveal otherwise hidden stimulation effects and, thereby, comprehensively induce and study the effects and underlying mechanisms of tACS.

Online EEG-Based Workload Adaptation of an Arithmetic Learning Environment.

In this paper, we demonstrate a closed-loop EEG-based learning environment, that adapts instructional learning material online, to improve learning success in students during arithmetic learning. The amount of cognitive workload during learning is crucial for successful learning and should be held in the optimal range for each learner. Based on EEG data from 10 subjects, we created a prediction model that estimates the learner's workload to obtain an unobtrusive workload measure. Furthermore, we developed an interactive learning environment that uses the prediction model to estimate the learner's workload online based on the EEG data and adapt the difficulty of the learning material to keep the learner's workload in an optimal range. The EEG-based learning environment was used by 13 subjects to learn arithmetic addition in the octal number system, leading to a significant learning effect. The results suggest that it is feasible to use EEG as an unobtrusive measure of cognitive workload to adapt the learning content. Further it demonstrates that a promptly workload prediction is possible using a generalized prediction model without the need for a user-specific calibration.

Decoding upper limb residual muscle activity in severe chronic stroke.

OBJECTIVE: Stroke is a leading cause of long-term motor disability. Stroke patients with severe hand weakness do not profit from rehabilitative treatments. Recently, brain-controlled robotics and sequential functional electrical stimulation allowed some improvement. However, for such therapies to succeed, it is required to decode patients' intentions for different arm movements. Here, we evaluated whether residual muscle activity could be used to predict movements from paralyzed joints in severely impaired chronic stroke patients. METHODS: Muscle activity was recorded with surface-electromyography (EMG) in 41 patients, with severe hand weakness (Fugl-Meyer Assessment [FMA] hand subscores of 2.93 ± 2.7), in order to decode their intention to perform six different motions of the affected arm, required for voluntary muscle activity and to control neuroprostheses. Decoding of paretic and nonparetic muscle activity was performed using a feed-forward neural network classifier. The contribution of each muscle to the intended movement was determined. RESULTS: Decoding of up to six arm movements was accurate (>65%) in more than 97% of nonparetic and 46% of paretic muscles. INTERPRETATION: These results demonstrate that some level of neuronal innervation to the paretic muscle remains preserved and can be used to implement neurorehabilitative treatments in 46% of patients with severe paralysis and extensive cortical and/or subcortical lesions. Such decoding may allow these patients for the first time after stroke to control different motions of arm prostheses through muscle-triggered rehabilitative treatments.

Cognitive state monitoring and the design of adaptive instruction in digital environments: lessons learned from cognitive workload assessment using a passive brain-computer interface approach.

According to Cognitive Load Theory (CLT), one of the crucial factors for successful learning is the type and amount of working-memory load (WML) learners experience while studying instructional materials. Optimal learning conditions are characterized by providing challenges for learners without inducing cognitive over- or underload. Thus, presenting instruction in a way that WML is constantly held within an optimal range with regard to learners' working-memory capacity might be a good method to provide these optimal conditions. The current paper elaborates how digital learning environments, which achieve this goal can be developed by combining approaches from Cognitive Psychology, Neuroscience, and Computer Science. One of the biggest obstacles that needs to be overcome is the lack of an unobtrusive method of continuously assessing learners' WML in real-time. We propose to solve this problem by applying passive Brain-Computer Interface (BCI) approaches to realistic learning scenarios in digital environments. In this paper we discuss the methodological and theoretical prospects and pitfalls of this approach based on results from the literature and from our own research. We present a strategy on how several inherent challenges of applying BCIs to WML and learning can be met by refining the psychological constructs behind WML, by exploring their neural signatures, by using these insights for sophisticated task designs, and by optimizing algorithms for analyzing electroencephalography (EEG) data. Based on this strategy we applied machine-learning algorithms for cross-task classifications of different levels of WML to tasks that involve studying realistic instructional materials. We obtained very promising results that yield several recommendations for future work.

Learned self-regulation of the lesioned brain with epidural electrocorticography.

INTRODUCTION: Different techniques for neurofeedback of voluntary brain activations are currently being explored for clinical application in brain disorders. One of the most frequently used approaches is the self-regulation of oscillatory signals recorded with electroencephalography (EEG). Many patients are, however, unable to achieve sufficient voluntary control of brain activity. This could be due to the specific anatomical and physiological changes of the patient's brain after the lesion, as well as to methodological issues related to the technique chosen for recording brain signals. METHODS: A patient with an extended ischemic lesion of the cortex did not gain volitional control of sensorimotor oscillations when using a standard EEG-based approach. We provided him with neurofeedback of his brain activity from the epidural space by electrocorticography (ECoG). RESULTS: Ipsilesional epidural recordings of field potentials facilitated self-regulation of brain oscillations in an online closed-loop paradigm and allowed reliable neurofeedback training for a period of 4 weeks. CONCLUSION: Epidural implants may decode and train brain activity even when the cortical physiology is distorted following severe brain injury. Such practice would allow for reinforcement learning of preserved neural networks and may well provide restorative tools for those patients who are severely afflicted.

From assistance towards restoration with epidural brain-computer interfacing.

PURPOSE: Today's implanted brain-computer interfaces make direct contact with the brain or even penetrate the tissue, bearing additional risks with regard to safety and stability. What is more, these approaches aim to control prosthetic devices as assistive tools and do not yet strive to become rehabilitative tools for restoring lost motor function. METHODS: We introduced a less invasive, implantable interface by applying epidural electrocorticography in a chronic stroke survivor with a persistent motor deficit. He was trained to modulate his natural motor-related oscillatory brain activity by receiving online feedback. RESULTS: Epidural recordings of field potentials in the beta-frequency band projecting onto the anatomical hand knob proved most successful in discriminating between the attempt to move the paralyzed hand and to rest. These spectral features allowed for fast and reliable control of the feedback device in an online closed-loop paradigm. Only seven training sessions were required to significantly improve maximum wrist extension. CONCLUSIONS: For patients suffering from severe motor deficits, epidural implants may decode and train the brain activity generated during attempts to move with high spatial resolution, thus facilitating specific and high-intensity practice even in the absence of motor control. This would thus transform them from pure assistive devices to restorative tools in the context of reinforcement learning and neurorehabilitation.

Estimation of diffusion coefficients from voltammetric signals by support vector and gaussian process regression.

BACKGROUND: Support vector regression (SVR) and Gaussian process regression (GPR) were used for the analysis of electroanalytical experimental data to estimate diffusion coefficients. RESULTS: For simulated cyclic voltammograms based on the EC, Eqr, and EqrC mechanisms these regression algorithms in combination with nonlinear kernel/covariance functions yielded diffusion coefficients with higher accuracy as compared to the standard approach of calculating diffusion coefficients relying on the Nicholson-Shain equation. The level of accuracy achieved by SVR and GPR is virtually independent of the rate constants governing the respective reaction steps. Further, the reduction of high-dimensional voltammetric signals by manual selection of typical voltammetric peak features decreased the performance of both regression algorithms compared to a reduction by downsampling or principal component analysis. After training on simulated data sets, diffusion coefficients were estimated by the regression algorithms for experimental data comprising voltammetric signals for three organometallic complexes. CONCLUSIONS: Estimated diffusion coefficients closely matched the values determined by the parameter fitting method, but reduced the required computational time considerably for one of the reaction mechanisms. The automated processing of voltammograms according to the regression algorithms yields better results than the conventional analysis of peak-related data.

Epidural electrocorticography of phantom hand movement following long-term upper-limb amputation.

INTRODUCTION: Prostheses for upper-limb amputees are currently controlled by either myoelectric or peripheral neural signals. Performance and dexterity of these devices is still limited, particularly when it comes to controlling hand function. Movement-related brain activity might serve as a complementary bio-signal for motor control of hand prosthesis. METHODS: We introduced a methodology to implant a cortical interface without direct exposure of the brain surface in an upper-limb amputee. This bi-directional interface enabled us to explore the cortical physiology following long-term transhumeral amputation. In addition, we investigated neurofeedback of electrocorticographic brain activity related to the patient's motor imagery to open his missing hand, i.e., phantom hand movement, for real-time control of a virtual hand prosthesis. RESULTS: Both event-related brain activity and cortical stimulation revealed mutually overlapping cortical representations of the phantom hand. Phantom hand movements could be robustly classified and the patient required only three training sessions to gain reliable control of the virtual hand prosthesis in an online closed-loop paradigm that discriminated between hand opening and rest. CONCLUSION: Epidural implants may constitute a powerful and safe alternative communication pathway between the brain and external devices for upper-limb amputees, thereby facilitating the integrated use of different signal sources for more intuitive and specific control of multi-functional devices in clinical use.

Spatial filtering based on canonical correlation analysis for classification of evoked or event-related potentials in EEG data.

Classification of evoked or event-related potentials is an important prerequisite for many types of brain-computer interfaces (BCIs). To increase classification accuracy, spatial filters are used to improve the signal-to-noise ratio of the brain signals and thereby facilitate the detection and classification of evoked or event-related potentials. While canonical correlation analysis (CCA) has previously been used to construct spatial filters that increase classification accuracy for BCIs based on visual evoked potentials, we show in this paper, how CCA can also be used for spatial filtering of event-related potentials like P300. We also evaluate the use of CCA for spatial filtering on other data with evoked and event-related potentials and show that CCA performs consistently better than other standard spatial filtering methods.

Coupling brain-machine interfaces with cortical stimulation for brain-state dependent stimulation: enhancing motor cortex excitability for neurorehabilitation.

Motor recovery after stroke is an unsolved challenge despite intensive rehabilitation training programs. Brain stimulation techniques have been explored in addition to traditional rehabilitation training to increase the excitability of the stimulated motor cortex. This modulation of cortical excitability augments the response to afferent input during motor exercises, thereby enhancing skilled motor learning by long-term potentiation-like plasticity. Recent approaches examined brain stimulation applied concurrently with voluntary movements to induce more specific use-dependent neural plasticity during motor training for neurorehabilitation. Unfortunately, such approaches are not applicable for the many severely affected stroke patients lacking residual hand function. These patients require novel activity-dependent stimulation paradigms based on intrinsic brain activity. Here, we report on such brain state-dependent stimulation (BSDS) combined with haptic feedback provided by a robotic hand orthosis. Transcranial magnetic stimulation (TMS) of the motor cortex and haptic feedback to the hand were controlled by sensorimotor desynchronization during motor-imagery and applied within a brain-machine interface (BMI) environment in one healthy subject and one patient with severe hand paresis in the chronic phase after stroke. BSDS significantly increased the excitability of the stimulated motor cortex in both healthy and post-stroke conditions, an effect not observed in non-BSDS protocols. This feasibility study suggests that closing the loop between intrinsic brain state, cortical stimulation and haptic feedback provides a novel neurorehabilitation strategy for stroke patients lacking residual hand function, a proposal that warrants further investigation in a larger cohort of stroke patients.

Assessing attention and cognitive function in completely locked-in state with event-related brain potentials and epidural electrocorticography.

OBJECTIVE: Patients in the completely locked-in state (CLIS), due to, for example, amyotrophic lateral sclerosis (ALS), no longer possess voluntary muscle control. Assessing attention and cognitive function in these patients during the course of the disease is a challenging but essential task for both nursing staff and physicians. APPROACH: An electrophysiological cognition test battery, including auditory and semantic stimuli, was applied in a late-stage ALS patient at four different time points during a six-month epidural electrocorticography (ECoG) recording period. Event-related cortical potentials (ERP), together with changes in the ECoG signal spectrum, were recorded via 128 channels that partially covered the left frontal, temporal and parietal cortex. MAIN RESULTS: Auditory but not semantic stimuli induced significant and reproducible ERP projecting to specific temporal and parietal cortical areas. N1/P2 responses could be detected throughout the whole study period. The highest P3 ERP was measured immediately after the patient's last communication through voluntary muscle control, which was paralleled by low theta and high gamma spectral power. Three months after the patient's last communication, i.e., in the CLIS, P3 responses could no longer be detected. At the same time, increased activity in low-frequency bands and a sharp drop of gamma spectral power were recorded. SIGNIFICANCE: Cortical electrophysiological measures indicate at least partially intact attention and cognitive function during sparse volitional motor control for communication. Although the P3 ERP and frequency-specific changes in the ECoG spectrum may serve as indicators for CLIS, a close-meshed monitoring will be required to define the exact time point of the transition.

Synaptic energy drives the information processing mechanisms in spiking neural networks.

Flow of energy and free energy minimization underpins almost every aspect of naturally occurring physical mechanisms. Inspired by this fact this work establishes an energy-based framework that spans the multi-scale range of biological neural systems and integrates synaptic dynamic, synchronous spiking activity and neural states into one consistent working paradigm. Following a bottom-up approach, a hypothetical energy function is proposed for dynamic synaptic models based on the theoretical thermodynamic principles and the Hopfield networks. We show that a synapse exposes stable operating points in terms of its excitatory postsynaptic potential as a function of its synaptic strength. We postulate that synapses in a network operating at these stable points can drive this network to an internal state of synchronous firing. The presented analysis is related to the widely investigated temporal coherent activities (cell assemblies) over a certain range of time scales (binding-by-synchrony). This introduces a novel explanation of the observed (poly)synchronous activities within networks regarding the synaptic (coupling) functionality. On a network level the transitions from one firing scheme to the other express discrete sets of neural states. The neural states exist as long as the network sustains the internal synaptic energy.

Prediction of P300 BCI aptitude in severe motor impairment.

Brain-computer interfaces (BCIs) provide a non-muscular communication channel for persons with severe motor impairments. Previous studies have shown that the aptitude with which a BCI can be controlled varies from person to person. A reliable predictor of performance could facilitate selection of a suitable BCI paradigm. Eleven severely motor impaired participants performed three sessions of a P300 BCI web browsing task. Before each session auditory oddball data were collected to predict the BCI aptitude of the participants exhibited in the current session. We found a strong relationship of early positive and negative potentials around 200 ms (elicited with the auditory oddball task) with performance. The amplitude of the P2 (r  =  -0.77) and of the N2 (r  =  -0.86) had the strongest correlations. Aptitude prediction using an auditory oddball was successful. The finding that the N2 amplitude is a stronger predictor of performance than P3 amplitude was reproduced after initially showing this effect with a healthy sample of BCI users. This will reduce strain on the end-users by minimizing the time needed to find suitable paradigms and inspire new approaches to improve performance.