A5047784390- EEG and Brain-Computer Interfaces
- Gaze Tracking and Assistive Technology
- Semantic Web and Ontologies
- Data Mining Algorithms and Applications
- Advanced Text Analysis Techniques
- Data Quality and Management
- Neonatal and fetal brain pathology
- Topic Modeling
- Tensor decomposition and applications
- Epilepsy research and treatment
- Muscle activation and electromyography studies
- Neural dynamics and brain function
- Cognitive Computing and Networks
- Neurological disorders and treatments
- ECG Monitoring and Analysis
- Neuroscience and Neuropharmacology Research
- Obstructive Sleep Apnea Research
- Neuroscience and Neural Engineering
- Functional Brain Connectivity Studies
- Blind Source Separation Techniques
- Spinal Cord Injury Research
- Advanced Graph Neural Networks
- Neural Networks and Applications
CEA LETI
2020-2025
Clinatec
2020-2025
CEA Grenoble
2020-2025
Commissariat à l'Énergie Atomique et aux Énergies Alternatives
2020-2025
Université Grenoble Alpes
2020-2025
Polytechnique Montréal
2020-2021
A spinal cord injury interrupts the communication between brain and region of that produces walking, leading to paralysis1,2. Here, we restored this with a digital bridge enabled an individual chronic tetraplegia stand walk naturally in community settings. This brain-spine interface (BSI) consists fully implanted recording stimulation systems establish direct link cortical signals3 analogue modulation epidural electrical targeting regions involved production walking4-6. highly reliable BSI...
Abstract Objective. The article aims at addressing 2 challenges to step motor brain-computer interface (BCI) out of laboratories: asynchronous control complex bimanual effectors with large numbers degrees freedom, using chronic and safe recorders, the decoding performance stability over time without frequent decoder recalibration. Approach. Closed-loop adaptive/incremental training is one strategy create a model stable time. Adaptive decoders update their parameters new incoming data,...
Introduction Phase-amplitude coupling (PAC), the modulation of high-frequency neural oscillations by phase slower oscillations, is increasingly recognized as a marker goal-directed motor behavior. Despite this interest, its specific role and potential value in decoding attempted movements remain unclear. Methods This study investigates whether PAC-derived features can be leveraged to classify different behaviors from ECoG signals within Brain-Computer Interface (BCI) systems. data were...
Large-scale knowledge graphs have become prevalent on the Web and demonstrated their usefulness for several tasks. One challenge associated to is necessity keep a graph schema (which generally manually defined) that accurately reflects content. In this paper, we present an approach extracts expressive taxonomy based embeddings, linked data statistics clustering. Our results show learned not only able retain original classes but also identifies new classes, thus giving up-to-date view of graph.
Focal cooling is emerging as a relevant therapy for drug-resistant epilepsy (DRE). However, we lack data on its effectiveness in controlling seizures that originate deep-seated areas like the hippocampus. We present thermoelectric solution focal brain specifically targets these structures.
While high-quality taxonomies are essential to the Semantic Web, building them for large knowledge graphs is an expensive process. Likewise, creating that accurately reflect content of dynamic another challenge. In this paper, we propose a method automatically extract taxonomy from graph embeddings, and evaluate it on DBpedia. Our approach produces by leveraging type information contained in tree-like structure unsupervised hierarchical clustering performed over entity embeddings. We then...
Motor Brain-Computer Interfaces (BCIs) create new communication pathways between the brain and external effectors for patients with severe motor impairments. Control of complex such as robotic arms or exoskeletons is generally based on real-time decoding high-resolution neural signals. However, high-dimensional noisy signals pose challenges, limitations in generalization ability model increased computational demands.The use sparse decoders may offer a way to address these challenges. A...
Quantifying and evaluating properly the performances is a critical issue in BCI experiments. The choice of most adapted metrics can be difficult because they are specific to experimental paradigm, task control, data. In current study evaluation criteria for closed-loop adaptive dynamic hierarchical discrete-continuous brain-computer interfaces examined. challenges such as imbalanced multi-class bias discrete decoding (classification), online computing cost analysis results time considered....
Brain-computer interfaces (BCIs) still face many challenges to step out of laboratories be used in real-life applications. A key one persists the high performance control diverse effectors for complex tasks, using chronic and safe recorders. This must robust over time decoding without continuous recalibration decoders. In article, asynchronous an exoskeleton by a tetraplegic patient chronically implanted epidural electrocorticography (EpiCoG) implant is demonstrated. For this purpose,...