ORCID Profile
0000-0002-4746-4633
Current Organisation
École Polytechnique Fédérale de Lausanne
Does something not look right? The information on this page has been harvested from data sources that may not be up to date. We continue to work with information providers to improve coverage and quality. To report an issue, use the Feedback Form.
Publisher: IEEE
Date: 06-2019
Publisher: Springer Science and Business Media LLC
Date: 22-05-2023
DOI: 10.1038/S41598-023-34801-9
Abstract: The analysis of motor evoked potentials (MEPs) generated by transcranial magnetic stimulation (TMS) is crucial in research and clinical medical practice. MEPs are characterized by their latency and the treatment of a single patient may require the characterization of thousands of MEPs. Given the difficulty of developing reliable and accurate algorithms, currently the assessment of MEPs is performed with visual inspection and manual annotation by a medical expert making it a time-consuming, inaccurate, and error-prone process. In this study, we developed DELMEP, a deep learning-based algorithm to automate the estimation of MEP latency. Our algorithm resulted in a mean absolute error of about 0.5 ms and an accuracy that was practically independent of the MEP litude. The low computational cost of the DELMEP algorithm allows employing it in on-the-fly characterization of MEPs for brain-state-dependent and closed-loop brain stimulation protocols. Moreover, its learning ability makes it a particularly promising option for artificial-intelligence-based personalized clinical applications.
Publisher: Cold Spring Harbor Laboratory
Date: 17-11-2020
DOI: 10.1101/2020.11.16.382267
Abstract: Visual motion discrimination involves reciprocal interactions in the alpha band between the primary visual cortex (V1) and the mediotemporal area (V5/MT). We investigated whether modulating alpha phase synchronization using in idualized multisite transcranial alternating current stimulation (tACS) over V5 and V1 regions would improve motion discrimination. We tested 3 groups of healthy subjects: 1) an in idualized In-Phase V1 alpha -V5 alpha tACS (0° lag) group, 2) an in idualized Anti-Phase V1 alpha -V5 alpha tACS (180° lag) group and 3) a sham tACS group. Motion discrimination and EEG activity were compared before, during and after tACS. Performance significantly improved in the Anti-Phase group compared to that in the In-Phase group at 10 and 30 minutes after stimulation. This result could be explained by changes in bottom-up alpha-V1 gamma-V5 phase- litude coupling. Thus, Anti-Phase V1 alpha -V5 alpha tACS might impose an optimal phase lag between stimulation sites due to the inherent speed of wave propagation, hereby supporting optimized neuronal communication. Alpha multisite (V1 and V5) tACS influences global motion discrimination and integration Phase- litude coupling is associated with visual performance Multisite Anti-Phase stimulation of strategic visual areas (V1 and V5) is associated with connectivity changes in the visual cortex and thus, associated with changes in direction acuity
Publisher: Elsevier BV
Date: 11-2022
Publisher: Elsevier BV
Date: 2016
Publisher: Ovid Technologies (Wolters Kluwer Health)
Date: 04-2023
DOI: 10.1161/STROKEAHA.122.040001
Abstract: Most studies on stroke have been designed to examine one deficit in isolation yet, survivors often have multiple deficits in different domains. While the mechanisms underlying multiple-domain deficits remain poorly understood, network-theoretical methods may open new avenues of understanding. Fifty subacute stroke patients (7±3days poststroke) underwent diffusion-weighted magnetic resonance imaging and a battery of clinical tests of motor and cognitive functions. We defined indices of impairment in strength, dexterity, and attention. We also computed imaging-based probabilistic tractography and whole-brain connectomes. To efficiently integrate inputs from different sources, brain networks rely on a rich-club of a few hub nodes. Lesions harm efficiency, particularly when they target the rich-club. Overlaying in idual lesion masks onto the tractograms enabled us to split the connectomes into their affected and unaffected parts and associate them to impairment. We computed efficiency of the unaffected connectome and found it was more strongly correlated to impairment in strength, dexterity, and attention than efficiency of the total connectome. The magnitude of the correlation between efficiency and impairment followed the order attention dexterity ≈ strength (strength: | r |=.03, P =0.02, dexterity: | r |=.30, P =0.05, attention: | r |=.55, P .001). Network weights associated with the rich-club were more strongly correlated to efficiency than non-rich-club weights. Attentional impairment is more sensitive to disruption of coordinated networks between brain regions than motor impairment, which is sensitive to disruption of localized networks. Providing more accurate reflections of actually functioning parts of the network enables the incorporation of information about the impact of brain lesions on connectomics contributing to a better understanding of underlying stroke mechanisms.
Publisher: IEEE
Date: 07-2019
Publisher: Frontiers Media SA
Date: 03-04-2020
Publisher: Cold Spring Harbor Laboratory
Date: 16-05-2022
DOI: 10.1101/2022.05.15.491882
Abstract: Cortico-cortical paired associative stimulation (ccPAS), which repeatedly pairs single pulse TMS over two distant brain regions with a specific time interval, is thought to modulate synaptic plasticity. Applied to the motion cortical pathway, ccPAS has been shown to improve motion discrimination when specifically targeting backward projections, stimulating the medio-temporal area (MT) followed by the primary visual cortex (V1). However, there is no direct neuroimaging evidence of the spatial selectivity of the ccPAS effects (i.e., pathway or direction specificity) or detailing the exact nature of the ccPAS effects (i.e., the oscillatory signature, timing…). In this study, we applied ccPAS along the motion discrimination pathway, in the top-down direction (MT-to-V1: “Backward ccPAS”) and in the bottom-up direction (V1-to-MT: “Forward ccPAS”) in sixteen healthy volunteers and compared changes in visual network activity in response to single pulse TMS over V1 and MT using spectral granger causality (sGC). The sGC results showed common increases in direct V1-to-MT and V1-to-IPS bottom-up inputs in the high Beta/low Gamma band (25-40 Hz) for both ccPAS, probably reflecting task exposure. However, a clear distinction in information transfer occurred in the re-entrant MT-to-V1 signals, which were only modulated by Backward ccPAS. This difference was predictive of the behavioural improvements at the motion discrimination task. Our results support the view of the possibility to specifically enhance re-entrant Alpha oscillatory signals from MT-to-V1 to promote motion discrimination performance through Backward ccPAS. These findings contribute to better understanding visual processing in healthy subjects and how it can be modulated to pave the way to clinical translation in vision handicapped patients. The changes in re-entrant MT-to-V1 inputs could help to provide single-subject prediction scenarios in patients suffering from a visual system stroke, in whom visual recovery might partly rely on the top-down inputs to the spared V1 neurons.
Publisher: Cold Spring Harbor Laboratory
Date: 04-03-2022
DOI: 10.1101/2022.03.03.482512
Abstract: Understanding the causal relationship between a focal lesion and brain network (re)organization is a crucial step in order to accurately predict the resulting symptoms of the lesion and implement personalized rehabilitation strategies. Transcranial magnetic stimulation (TMS) can be used to create local and transient neural perturbations, the so-called “virtual lesion” approach. In this study, we tested how a virtual lesion applied to the Early Visual Areas (EVA) or the extrastriate, motion-sensitive, medio-temporal area (hMT+/V5) in different contexts affects behavior and changes neuronal network activity and organisation, assessed with functional magnetic resonance imaging (fMRI). We applied short trains of 10 Hz TMS to healthy participants over the EVA or hMT+/MT. This was done both at rest and at relevant times during a motion discrimination task, while concurrent fMRI was performed. Regional BOLD activity was analysed using a general linear model (GLM) and functional brain networks were assessed using independent component analysis (ICA). Motion direction discrimination and motion awareness were related to the imaging data. TMS applied over the EVA and hMT+/V5 induced transient modulation of motion perception and discrimination. Comparing resting versus active states, both TMS sites showed a common suppression of local and remote brain activity at rest while an over-activation of the stimulated areas and related networks were found during the task. More subtly, distinct dynamic and topological TMS-induced networks properties could be revealed depending on the exact visual processing stages TMS was applied at. In particular, brain networks associated with EVA stimulation showed a clear context-dependency and were spatially more restricted than for hMT+/V5 stimulation. The present findings highlight the possibility of interfering with distinct visual processing stages and the possibility of imaging the local and remote neural correlates of the behavioral impact. They also confirm the complexity of TMS effects on BOLD activity, switching from signal suppression at rest to irrelevant neural noise addition during active visual processing, even differing throughout the time course of information processing. Moreover, the networks analyses suggest that the EVA might be more resilient to focal perturbations by means of a “virtual lesion” than hMT+/V5. As a critical processing center, the EVA may be important for maintaining stability in the visual network, whereas the perturbation of a more specialized region such as hMT+/V5 has greater impact on local and network activity, as well as on behavior. These findings add to the understanding of visual motion processing and especially to the impact and potential mechanism of focal lesions in this system.
No related grants have been discovered for Friedhelm Hummel.