A reusable benchmark of brain-age prediction from M/EEG resting-state signals.

Population-level modeling can define quantitative measures of individual aging by applying machine learning to large volumes of brain images. These measures of brain age, obtained from the general population, helped characterize disease severity in neurological populations, improving estimates of diagnosis or prognosis. Magnetoencephalography (MEG) and Electroencephalography (EEG) have the potential to further generalize this approach towards prevention and public health by enabling assessments of brain health at large scales in socioeconomically diverse environments. However, more research is needed to define methods that can handle the complexity and diversity of M/EEG signals across diverse real-world contexts. To catalyse this effort, here we propose reusable benchmarks of competing machine learning approaches for brain age modeling. We benchmarked popular classical machine learning pipelines and deep learning architectures previously used for pathology decoding or brain age estimation in 4 international M/EEG cohorts from diverse countries and cultural contexts, including recordings from more than 2500 participants. Our benchmarks were built on top of the M/EEG adaptations of the BIDS standard, providing tools that can be applied with minimal modification on any M/EEG dataset provided in the BIDS format. Our results suggest that, regardless of whether classical machine learning or deep learning was used, the highest performance was reached by pipelines and architectures involving spatially aware representations of the M/EEG signals, leading to R2 scores between 0.60-0.74. Hand-crafted features paired with random forest regression provided robust benchmarks even in situations in which other approaches failed. Taken together, this set of benchmarks, accompanied by open-source software and high-level Python scripts, can serve as a starting point and quantitative reference for future efforts at developing M/EEG-based measures of brain aging. The generality of the approach renders this benchmark reusable for other related objectives such as modeling specific cognitive variables or clinical endpoints.

Blink- and saccade-related suppression effects in early visual areas of the human brain: Intracranial EEG investigations during natural viewing conditions.

Blinks and saccades, both ubiquitous in natural viewing conditions, cause rapid changes of visual inputs that are hardly consciously perceived. The neural dynamics in early visual areas of the human brain underlying this remarkable visual stability are still incompletely understood. We used electrocorticography (ECoG) from electrodes directly implanted on the human early visual areas V1, V2, V3d/v, V4d/v and the fusiform gyrus to investigate blink- and saccade-related neuronal suppression effects during non-experimental, free viewing conditions. We found a characteristic, biphasic, broadband gamma power decrease-increase pattern in all investigated visual areas. During saccades, a decrease in gamma power clearly preceded eye movement onset, at least in V1. This may indicate that cortical information processing is actively suppressed in human early visual areas before and during saccades, which then possibly mediates perceptual visual suppression. The following eye movement offset-related increase in gamma power may indicate the recovery of visual perception and the resumption of visual processing.

The CLAIRE COVID-19 initiative: approach, experiences and recommendations.

A volunteer effort by Artificial Intelligence (AI) researchers has shown it can deliver significant research outcomes rapidly to help tackle COVID-19. Within two months, CLAIRE's self-organising volunteers delivered the World's first comprehensive curated repository of COVID-19-related datasets useful for drug-repurposing, drafted review papers on the role CT/X-ray scan analysis and robotics could play, and progressed research in other areas. Given the pace required and nature of voluntary efforts, the teams faced a number of challenges. These offer insights in how better to prepare for future volunteer scientific efforts and large scale, data-dependent AI collaborations in general. We offer seven recommendations on how to best leverage such efforts and collaborations in the context of managing future crises.

Machine-learning-based diagnostics of EEG pathology.

Machine learning (ML) methods have the potential to automate clinical EEG analysis. They can be categorized into feature-based (with handcrafted features), and end-to-end approaches (with learned features). Previous studies on EEG pathology decoding have typically analyzed a limited number of features, decoders, or both. For a I) more elaborate feature-based EEG analysis, and II) in-depth comparisons of both approaches, here we first develop a comprehensive feature-based framework, and then compare this framework to state-of-the-art end-to-end methods. To this aim, we apply the proposed feature-based framework and deep neural networks including an EEG-optimized temporal convolutional network (TCN) to the task of pathological versus non-pathological EEG classification. For a robust comparison, we chose the Temple University Hospital (TUH) Abnormal EEG Corpus (v2.0.0), which contains approximately 3000 EEG recordings. The results demonstrate that the proposed feature-based decoding framework can achieve accuracies on the same level as state-of-the-art deep neural networks. We find accuracies across both approaches in an astonishingly narrow range from 81 to 86%. Moreover, visualizations and analyses indicated that both approaches used similar aspects of the data, e.g., delta and theta band power at temporal electrode locations. We argue that the accuracies of current binary EEG pathology decoders could saturate near 90% due to the imperfect inter-rater agreement of the clinical labels, and that such decoders are already clinically useful, such as in areas where clinical EEG experts are rare. We make the proposed feature-based framework available open source and thus offer a new tool for EEG machine learning research.

Large-scale cortical travelling waves predict localized future cortical signals.

Predicting future brain signal is highly sought-after, yet difficult to achieve. To predict the future phase of cortical activity at localized ECoG and MEG recording sites, we exploit its predominant, large-scale, spatiotemporal dynamics. The dynamics are extracted from the brain signal through Fourier analysis and principal components analysis (PCA) only, and cast in a data model that predicts future signal at each site and frequency of interest. The dominant eigenvectors of the PCA that map the large-scale patterns of past cortical phase to future ones take the form of smoothly propagating waves over the entire measurement array. In ECoG data from 3 subjects and MEG data from 20 subjects collected during a self-initiated motor task, mean phase prediction errors were as low as 0.5 radians at local sites, surpassing state-of-the-art methods of within-time-series or event-related models. Prediction accuracy was highest in delta to beta bands, depending on the subject, was more accurate during episodes of high global power, but was not strongly dependent on the time-course of the task. Prediction results did not require past data from the to-be-predicted site. Rather, best accuracy depended on the availability in the model of long wavelength information. The utility of large-scale, low spatial frequency traveling waves in predicting future phase activity at local sites allows estimation of the error introduced by failing to account for irreducible trajectories in the activity dynamics.

A transcallosal fibre system between homotopic inferior frontal regions supports complex linguistic processing.

Inferior frontal regions in the left and right hemisphere support different aspects of language processing. In the canonical model, left inferior frontal regions are mostly involved in processing based on phonological, syntactic and semantic features of language, whereas the right inferior frontal regions process paralinguistic aspects like affective prosody. Using diffusion tensor imaging (DTI)-based probabilistic fibre tracking in 20 healthy volunteers, we identify a callosal fibre system connecting left and right inferior frontal regions that are involved in linguistic processing of varying complexity. Anatomically, we show that the interhemispheric fibres are highly aligned and distributed along a rostral to caudal gradient in the body and genu of the corpus callosum to connect homotopic inferior frontal regions. In the light of converging data, taking previous DTI-based tracking studies and clinical case studies into account, our findings suggest that the right inferior frontal cortex not only processes paralinguistic aspects of language (such as affective prosody), as purported by the canonical model, but also supports the computation of linguistic aspects of varying complexity in the human brain. Our model may explain patterns of right-hemispheric contribution to stroke recovery as well as disorders of prosodic processing. Beyond language-related brain function, we discuss how inter-species differences in interhemispheric connectivity and fibre density, including the system we described here may also explain differences in transcallosal information transfer and cognitive abilities across different mammalian species.

Computationally optimized ECoG stimulation with local safety constraints.

Direct stimulation of the cortical surface is used clinically for cortical mapping and modulation of local activity. Future applications of cortical modulation and brain-computer interfaces may also use cortical stimulation methods. One common method to deliver current is through electrocorticography (ECoG) stimulation in which a dense array of electrodes are placed subdurally or epidurally to stimulate the cortex. However, proximity to cortical tissue limits the amount of current that can be delivered safely. It may be desirable to deliver higher current to a specific local region of interest (ROI) while limiting current to other local areas more stringently than is guaranteed by global safety limits. Two commonly used global safety constraints bound the total injected current and individual electrode currents. However, these two sets of constraints may not be sufficient to prevent high current density locally (hot-spots). In this work, we propose an efficient approach that prevents current density hot-spots in the entire brain while optimizing ECoG stimulus patterns for targeted stimulation. Specifically, we maximize the current along a particular desired directional field in the ROI while respecting three safety constraints: one on the total injected current, one on individual electrode currents, and the third on the local current density magnitude in the brain. This third set of constraints creates a computational barrier due to the huge number of constraints needed to bound the current density at every point in the entire brain. We overcome this barrier by adopting an efficient two-step approach. In the first step, the proposed method identifies the safe brain region, which cannot contain any hot-spots solely based on the global bounds on total injected current and individual electrode currents. In the second step, the proposed algorithm iteratively adjusts the stimulus pattern to arrive at a solution that exhibits no hot-spots in the remaining brain. We report on simulations on a realistic finite element (FE) head model with five anatomical ROIs and two desired directional fields. We also report on the effect of ROI depth and desired directional field on the focality of the stimulation. Finally, we provide an analysis of optimization runtime as a function of different safety and modeling parameters. Our results suggest that optimized stimulus patterns tend to differ from those used in clinical practice.

The dynamics of error processing in the human brain as reflected by high-gamma activity in noninvasive and intracranial EEG.

Error detection in motor behavior is a fundamental cognitive function heavily relying on local cortical information processing. Neural activity in the high-gamma frequency band (HGB) closely reflects such local cortical processing, but little is known about its role in error processing, particularly in the healthy human brain. Here we characterize the error-related response of the human brain based on data obtained with noninvasive EEG optimized for HGB mapping in 31 healthy subjects (15 females, 16 males), and additional intracranial EEG data from 9 epilepsy patients (4 females, 5 males). Our findings reveal a multiscale picture of the global and local dynamics of error-related HGB activity in the human brain. On the global level as reflected in the noninvasive EEG, the error-related response started with an early component dominated by anterior brain regions, followed by a shift to parietal regions, and a subsequent phase characterized by sustained parietal HGB activity. This phase lasted for more than 1 s after the error onset. On the local level reflected in the intracranial EEG, a cascade of both transient and sustained error-related responses involved an even more extended network, spanning beyond frontal and parietal regions to the insula and the hippocampus. HGB mapping appeared especially well suited to investigate late, sustained components of the error response, possibly linked to downstream functional stages such as error-related learning and behavioral adaptation. Our findings establish the basic spatio-temporal properties of HGB activity as a neural correlate of error processing, complementing traditional error-related potential studies.

Deep learning with convolutional neural networks for EEG decoding and visualization.

Deep learning with convolutional neural networks (deep ConvNets) has revolutionized computer vision through end-to-end learning, that is, learning from the raw data. There is increasing interest in using deep ConvNets for end-to-end EEG analysis, but a better understanding of how to design and train ConvNets for end-to-end EEG decoding and how to visualize the informative EEG features the ConvNets learn is still needed. Here, we studied deep ConvNets with a range of different architectures, designed for decoding imagined or executed tasks from raw EEG. Our results show that recent advances from the machine learning field, including batch normalization and exponential linear units, together with a cropped training strategy, boosted the deep ConvNets decoding performance, reaching at least as good performance as the widely used filter bank common spatial patterns (FBCSP) algorithm (mean decoding accuracies 82.1% FBCSP, 84.0% deep ConvNets). While FBCSP is designed to use spectral power modulations, the features used by ConvNets are not fixed a priori. Our novel methods for visualizing the learned features demonstrated that ConvNets indeed learned to use spectral power modulations in the alpha, beta, and high gamma frequencies, and proved useful for spatially mapping the learned features by revealing the topography of the causal contributions of features in different frequency bands to the decoding decision. Our study thus shows how to design and train ConvNets to decode task-related information from the raw EEG without handcrafted features and highlights the potential of deep ConvNets combined with advanced visualization techniques for EEG-based brain mapping. Hum Brain Mapp 38:5391-5420, 2017. © 2017 Wiley Periodicals, Inc.

Predominance of Movement Speed Over Direction in Neuronal Population Signals of Motor Cortex: Intracranial EEG Data and A Simple Explanatory Model.

How neuronal activity of motor cortex is related to movement is a central topic in motor neuroscience. Motor-cortical single neurons are more closely related to hand movement velocity than speed, that is, the magnitude of the (directional) velocity vector. Recently, there is also increasing interest in the representation of movement parameters in neuronal population activity, such as reflected in the intracranial EEG (iEEG). We show that in iEEG, contrasting to what has been previously found on the single neuron level, speed predominates over velocity. The predominant speed representation was present in nearly all iEEG signal features, up to the 600-1000 Hz range. Using a model of motor-cortical signals arising from neuronal populations with realistic single neuron tuning properties, we show how this reversal can be understood as a consequence of increasing population size. Our findings demonstrate that the information profile in large population signals may systematically differ from the single neuron level, a principle that may be helpful in the interpretation of neuronal population signals in general, including, for example, EEG and functional magnetic resonance imaging. Taking advantage of the robust speed population signal may help in developing brain-machine interfaces exploiting population signals.

The Effects of Closed-Loop Medical Devices on the Autonomy and Accountability of Persons and Systems.

Closed-loop medical devices such as brain-computer interfaces are an emerging and rapidly advancing neurotechnology. The target patients for brain-computer interfaces (BCIs) are often severely paralyzed, and thus particularly vulnerable in terms of personal autonomy, decisionmaking capacity, and agency. Here we analyze the effects of closed-loop medical devices on the autonomy and accountability of both persons (as patients or research participants) and neurotechnological closed-loop medical systems. We show that although BCIs can strengthen patient autonomy by preserving or restoring communicative abilities and/or motor control, closed-loop devices may also create challenges for moral and legal accountability. We advocate the development of a comprehensive ethical and legal framework to address the challenges of emerging closed-loop neurotechnologies like BCIs and stress the centrality of informed consent and refusal as a means to foster accountability. We propose the creation of an international neuroethics task force with members from medical neuroscience, neuroengineering, computer science, medical law, and medical ethics, as well as representatives of patient advocacy groups and the public.

Causal interpretation rules for encoding and decoding models in neuroimaging.

Causal terminology is often introduced in the interpretation of encoding and decoding models trained on neuroimaging data. In this article, we investigate which causal statements are warranted and which ones are not supported by empirical evidence. We argue that the distinction between encoding and decoding models is not sufficient for this purpose: relevant features in encoding and decoding models carry a different meaning in stimulus- and in response-based experimental paradigms.We show that only encoding models in the stimulus-based setting support unambiguous causal interpretations. By combining encoding and decoding models trained on the same data, however, we obtain insights into causal relations beyond those that are implied by each individual model type. We illustrate the empirical relevance of our theoretical findings on EEG data recorded during a visuo-motor learning task.

Visualization of the amygdalo-hippocampal border and its structural variability by 7T and 3T magnetic resonance imaging.

The amygdala and the hippocampus are two adjacent structures in the medial temporal lobe that have been broadly investigated in functional and structural neuroimaging due to their central importance in sensory perception, emotion, and memory. Exact demarcation of the amygdalo-hippocampal border (AHB) is, however, difficult in conventional structural imaging. Recent evidence suggests that, due to this difficulty, functional activation sites with high probability of being located in the hippocampus may erroneously be assigned to the amygdala, and vice versa. In the present study, we investigated the potential of ultra-high-field magnetic resonance imaging (MRI) in single sessions for detecting the AHB in humans. We show for the first time the detailed structure of the AHB as it can be visualized in T1-weighted 7T in vivo images at 0.5-mm(3) isotropic resolution. Compared to data acquired at 3T, 7T images revealed considerably more structural detail in the AHB region. Thus, we observed a striking inter-hemispheric and interindividual variability of the exact anatomical configuration of the AHB that points to the necessity of individual imaging of the AHB as a prerequisite for accurate anatomical assignment in this region. The findings of the present study demonstrate the usefulness of ultra-high-field structural MRI to resolve anatomical ambiguities of the human AHB. Highly accurate morphometric and functional investigations in this region at 7T may allow addressing such hitherto unexplored issues as whether the structural configuration of the AHB is related to functional differences in amygdalo-hippocampal interaction.

Reaching the ceiling? Empirical scaling behaviour for deep EEG pathology classification.

Machine learning techniques, particularly deep convolutional neural networks (ConvNets), are increasingly being used to automate clinical EEG analysis, with the potential to reduce the clinical burden and improve patient care. However, further research is required before they can be used in clinical settings, particularly regarding the impact of the number of training samples and model parameters on their testing error. To address this, we present a comprehensive study of the empirical scaling behaviour of ConvNets for EEG pathology classification. We analysed the testing error with increasing the training samples and model size for four different ConvNet architectures. The focus of our experiments is width scaling, and we have increased the number of parameters to up to 1.8 million. Our evaluation was based on two publicly available datasets: the Temple University Hospital (TUH) Abnormal EEG Corpus and the TUH Abnormal Expansion Balanced EEG Corpus, which together contain 10,707 training samples. The results show that the testing error follows a saturating power-law with both model and dataset size. This pattern is consistent across different datasets and ConvNet architectures. Furthermore, empirically observed accuracies saturate at 85%-87%, which may be due to an imperfect inter-rater agreement on the clinical labels. The empirical scaling behaviour of the test performance with dataset and model size has significant implications for deep EEG pathology classification research and practice. Our findings highlight the potential of deep ConvNets for high-performance EEG pathology classification, and the identified scaling relationships provide valuable recommendations for the advancement of automated EEG diagnostics.

Heart cycle-related effects on event-related potentials, spectral power changes, and connectivity patterns in the human ECoG.

The perception of one's own heartbeat is a fundamental interoceptive process that involves cortical and subcortical structures. Yet, the precise spatiotemporal neuronal activity patterns underlying the cortical information processing have remained largely elusive. Although the high temporal and spatial resolution of electrocorticographic (ECoG) recordings is increasingly being exploited in functional neuroimaging, it has not been used to study heart cycle-related effects. Here, we addressed the capacity of ECoG to characterize neuronal signals within the cardiac cycle, as well as to disentangle them from heart cycle-related artifacts. Based on topographical distribution and latency, we identified a biphasic potential within the primary somatosensory cortex, which likely constitutes a heartbeat-evoked potential (HEP) of neuronal origin. We also found two different types of artifacts: i) oscillatory potential changes with a frequency identical to the heart pulse rate, which probably represent pulsatility artifacts and ii) sharp potentials synchronized to the R-peak, corresponding to the onset of ventricular contraction and the cardiac field artifact (CFA) in EEG. Finally, we show that heart cycle-related effects induce pronounced phase-synchrony patterns in the ECoG and that this kind of correlation patterns, which may confound ECoG connectivity studies, can be reduced by a suitable correction algorithm. The present study is, to our knowledge, the first one to show a focally localized cortical HEP that could be clearly and consistently observed over subjects, suggesting a basic role of primary sensory cortex in processing of heart-related sensory inputs. We also conclude that taking into account and reducing heart cycle-related effects may be advantageous for many ECoG studies, and are of crucial importance, particularly for ECoG-based connectivity studies. Thus, in summary, although ECoG poses new challenges, it opens up new possibilities for the investigation of heartbeat-related viscerosensory processing in the human brain.

Somatotopic mapping of natural upper- and lower-extremity movements and speech production with high gamma electrocorticography.

Precise delineation of pathological and eloquent cortices is essential in pre-neurosurgical diagnostics of epilepsy. A limitation of existing experimental procedures, however, is that they critically require active cooperation of the patient, which is not always achievable, particularly in infants and in patients with insufficient cognitive abilities. In the present study, we evaluated the potential of electrocorticographic recordings of high gamma activity during natural, non-experimental behavior of epilepsy patients to localize upper- and lower-extremity motor and language functions, and compared the results with those obtained using electrocortical stimulation. The observed effects were highly significant and functionally specific, and agreed well with the somatotopic organization of the motor cortex, both on the lateral convexity and in the supplementary motor area. Our approach showed a similar specificity and sensitivity for extremity movements as previously obtained from experimental data. We were able to quantify, for the first time, sensitivity and specificity of high gamma underlying non-experimental lower-extremity movements in four patients, and observed values in the same range as for upper extremities (analyzed in six patients). Speech-related responses in the three investigated patients, however, exhibited only a very low sensitivity. The present findings indicate that localization of not only upper- but also lower-extremity movements congruent with electrocortical stimulation mapping is possible based on event-related high gamma responses that can be observed during natural behavior. Thus, non-experimental mapping may be usefully applied as adjunct to established clinical procedures for identification of both upper- and lower-extremity motor functions.

Co-localizing linguistic and musical syntax with intracranial EEG.

Despite general agreement on shared syntactic resources in music and language, the neuroanatomical underpinnings of this overlap remain largely unexplored. While previous studies mainly considered frontal areas as supramodal grammar processors, the domain-general syntactic role of temporal areas has been so far neglected. Here we capitalized on the excellent spatial and temporal resolution of subdural EEG recordings to co-localize low-level syntactic processes in music and language in the temporal lobe in a within-subject design. We used Brain Surface Current Density mapping to localize and compare neural generators of the early negativities evoked by violations of phrase structure grammar in both music and spoken language. The results show that the processing of syntactic violations relies in both domains on bilateral temporo-fronto-parietal neural networks. We found considerable overlap of these networks in the superior temporal lobe, but also differences in the hemispheric timing and relative weighting of their fronto-temporal constituents. While alluding to the dissimilarity in how shared neural resources may be configured depending on the musical or linguistic nature of the perceived stimulus, the combined data lend support for a co-localization of early musical and linguistic syntax processing in the temporal lobe.

Traveling waves and trial averaging: the nature of single-trial and averaged brain responses in large-scale cortical signals.

Analyzing single trial brain activity remains a challenging problem in the neurosciences. We gain purchase on this problem by focusing on globally synchronous fields in within-trial evoked brain activity, rather than on localized peaks in the trial-averaged evoked response (ER). We analyzed data from three measurement modalities, each with different spatial resolutions: magnetoencephalogram (MEG), electroencephalogram (EEG) and electrocorticogram (ECoG). We first characterized the ER in terms of summation of phase and amplitude components over trials. Both contributed to the ER, as expected, but the ER topography was dominated by the phase component. This means the observed topography of cross-trial phase will not necessarily reflect the phase topography within trials. To assess the organization of within-trial phase, traveling wave (TW) components were quantified by computing the phase gradient. TWs were intermittent but ubiquitous in the within-trial evoked brain activity. At most task-relevant times and frequencies, the within-trial phase topography was described better by a TW than by the trial-average of phase. The trial-average of the TW components also reproduced the topography of the ER; we suggest that the ER topography arises, in large part, as an average over TW behaviors. These findings were consistent across the three measurement modalities. We conclude that, while phase is critical to understanding the topography of event-related activity, the preliminary step of collating cortical signals across trials can obscure the TW components in brain activity and lead to an underestimation of the coherent motion of cortical fields.

An extended clinical EEG dataset with 15,300 automatically labelled recordings for pathology decoding.

Automated clinical EEG analysis using machine learning (ML) methods is a growing EEG research area. Previous studies on binary EEG pathology decoding have mainly used the Temple University Hospital (TUH) Abnormal EEG Corpus (TUAB) which contains approximately 3,000 manually labelled EEG recordings. To evaluate and eventually even improve the generalisation performance of machine learning methods for EEG pathology, decoding larger, publicly available datasets is required. A number of studies addressed the automatic labelling of large open-source datasets as an approach to create new datasets for EEG pathology decoding, but little is known about the extent to which training on larger, automatically labelled dataset affects decoding performances of established deep neural networks. In this study, we automatically created additional pathology labels for the Temple University Hospital (TUH) EEG Corpus (TUEG) based on the medical reports using a rule-based text classifier. We generated a dataset of 15,300 newly labelled recordings, which we call the TUH Abnormal Expansion EEG Corpus (TUABEX), and which is five times larger than the TUAB. Since the TUABEX contains more pathological (75%) than non-pathological (25%) recordings, we then selected a balanced subset of 8,879 recordings, the TUH Abnormal Expansion Balanced EEG Corpus (TUABEXB). To investigate how training on a larger, automatically labelled dataset affects the decoding performance of deep neural networks, we applied four established deep convolutional neural networks (ConvNets) to the task of pathological versus non-pathological classification and compared the performance of each architecture after training on different datasets. The results show that training on the automatically labelled TUABEXB dataset rather than training on the manually labelled TUAB dataset increases accuracies on TUABEXB and even for TUAB itself for some architectures. We argue that automatically labelling of large open-source datasets can be used to efficiently utilise the massive amount of EEG data stored in clinical archives. We make the proposed TUABEXB available open source and thus offer a new dataset for EEG machine learning research.

Decoding natural grasp types from human ECoG.

Electrocorticographic (ECoG) signals have been successfully used to provide information about arm movement direction, individual finger movements and even continuous arm movement trajectories. Thus, ECoG has been proposed as a potential control signal for implantable brain-machine interfaces (BMIs) in paralyzed patients. For the neuronal control of a prosthesis with versatile hand/arm functions, it is also necessary to successfully decode different types of grasping movements, such as precision grip and whole-hand grip. Although grasping is one of the most frequent and important hand movements performed in everyday life, until now, the decoding of ECoG activity related to different grasp types has not been systematically investigated. Here, we show that two different grasp types (precision vs. whole-hand grip) can be reliably distinguished in natural reach-to-grasp movements in single-trial ECoG recordings from the human motor cortex. Self-paced movement execution in a paradigm accounting for variability in grasped object position and weight was chosen to create a situation similar to everyday settings. We identified three informative signal components (low-pass-filtered component, low-frequency and high-frequency amplitude modulations), which allowed for accurate decoding of precision and whole-hand grips. Importantly, grasp type decoding generalized over different object positions and weights. Within the frontal lobe, informative signals predominated in the precentral motor cortex and could also be found in the right hemisphere's homologue of Broca's area. We conclude that ECoG signals are promising candidates for BMIs that include the restoration of grasping movements.