Low intensity transcranial electrical stimulation (TES) in humans, encompassing transcranial direct current (tDCS), transcutaneous spinal Direct Current Stimulation (tsDCS), transcranial alternating current (tACS), and transcranial random noise (tRNS) stimulation or their combinations, appears to be safe. No serious adverse events (SAEs) have been reported so far in over 18,000 sessions administered to healthy subjects, neurological and psychiatric patients, as summarized here. Moderate adverse events (AEs), as defined by the necessity to intervene, are rare, and include skin burns with tDCS due to suboptimal electrode-skin contact. Very rarely mania or hypomania was induced in patients with depression (11 documented cases), yet a causal relationship is difficult to prove because of the low incidence rate and limited numbers of subjects in controlled trials. Mild AEs (MAEs) include headache and fatigue following stimulation as well as prickling and burning sensations occurring during tDCS at peak-to-baseline intensities of 1–2 mA and during tACS at higher peak-to-peak intensities above 2 mA. The prevalence of published AEs is different in studies specifically assessing AEs vs. those not assessing them, being higher in the former. AEs are frequently reported by individuals receiving placebo stimulation. The profile of AEs in terms of frequency, magnitude and type is comparable in healthy and clinical populations, and this is also the case for more vulnerable populations, such as children, elderly persons, or pregnant women. Combined interventions (e.g., co-application of drugs, electrophysiological measurements, neuroimaging) were not associated with further safety issues. Safety is established for low-intensity ‘conventional’ TES defined as <4 mA, up to 60 min duration per day. Animal studies and modeling evidence indicate that brain injury could occur at predicted current densities in the brain of 6.3–13 A/m2 that are over an order of magnitude above those produced by tDCS in humans. Using AC stimulation fewer AEs were reported compared to DC. In specific paradigms with amplitudes of up to 10 mA, frequencies in the kHz range appear to be safe. In this paper we provide structured interviews and recommend their use in future controlled studies, in particular when trying to extend the parameters applied. We also discuss recent regulatory issues, reporting practices and ethical issues. These recommendations achieved consensus in a meeting, which took place in Göttingen, Germany, on September 6–7, 2016 and were refined thereafter by email correspondence.

No abstract

Deep generative architectures provide a way to model not only images but also complex, 3-dimensional objects, such as point clouds. In this work, we present a novel method to obtain meaningful representations of 3D shapes that can be used for challenging tasks including 3D points generation, reconstruction, compression, and clustering. Contrary to existing methods for 3D point cloud generation that train separate decoupled models for representation learning and generation, our approach is the first endto-end solution that allows to simultaneously learn a latent space of representation and generate 3D shape out of it. Moreover, our model is capable of learning meaningful compact binary descriptors with adversarial training conducted on a latent space. To achieve this goal, we extend a deep Adversarial Autoencoder model (AAE) to accept 3D input and create 3D output. Thanks to our end-to-end training regime, the resulting method called 3D Adversarial Autoencoder (3dAAE) obtains either binary or continuous latent space that covers a much wider portion of training data distribution. Finally, our quantitative evaluation shows that 3dAAE provides state-of-the-art results for 3D points clustering and 3D object retrieval.

The purpose of this study was to describe the methodology necessary for simultaneous recording of intracranial EEG (ICEEG) and magnetoencephalography (MEG) and to assess the sensitivity of whole-head MEG versus depth electrode EEG in the detection and localization of epileptic spikes. Interictal MEG and depth electrode activities from the temporal mesial and occipital lobes were simultaneously recorded from four candidates for epilepsy surgery. Implanted depth electrodes identified neocortical and mesial structures of ictal onset. Interictal spikes detected by these same depth electrodes were compared with simultaneous MEG events. MEG detections of ICEEG spikes, ICEEG versus MEG spike amplitudes, number of ICEEG contacts involved in the spike, and anatomic locations of MEG equivalent current dipoles were analyzed. MEG detected and localized 95% of the neocortical spikes, but only 25% to 60% of spikes from mesial structures. Mesial temporal spikes resulted in lower MEG spike amplitudes, when compared with neocortical spikes. Equivalent current dipoles of MEG spikes localized to the ictal onset zones in all four patients. MEG can detect and localize interictal epileptiform spikes that are recorded from depth electrodes in both neocortical and mesial structures, despite the lesser amplitude of spikes of mesial origin.