Refine
Year of publication
- 2024 (54) (remove)
Institute
- Fachbereich Medizintechnik und Technomathematik (23)
- IfB - Institut für Bioengineering (10)
- Fachbereich Elektrotechnik und Informationstechnik (8)
- Fachbereich Luft- und Raumfahrttechnik (7)
- INB - Institut für Nano- und Biotechnologien (7)
- Fachbereich Chemie und Biotechnologie (5)
- Fachbereich Wirtschaftswissenschaften (5)
- ECSM European Center for Sustainable Mobility (4)
- Nowum-Energy (2)
- Arbeitsstelle fuer Hochschuldidaktik und Studienberatung (1)
Language
- English (54) (remove)
Document Type
- Article (34)
- Conference Proceeding (12)
- Part of a Book (3)
- Book (2)
- Preprint (2)
- Working Paper (1)
Keywords
- Engineering education (2)
- Hot S-parameter (2)
- 1P hub loads (1)
- 3-D printing (1)
- 316L (1)
- ABE (1)
- Accuracy (1)
- Acid crash (1)
- Active humidity control (1)
- Acyl-amino acids (1)
- Acylation (1)
- Additive manufacturing (1)
- Aeroelasticity (1)
- Ageing (1)
- Agent-based modeling (1)
- Aminoacylase (1)
- Analytical models (1)
- Anatomy (1)
- Bio-inspired systems (1)
- Biocatalysis (1)
- Biological hydrogen (1)
- Biosurfactants (1)
- Boundary integral equations, (1)
- Butanol (1)
- C. acetobutylicum (1)
- Capacitive model (1)
- Carrier solvents (1)
- Centrifugal twisting moment (1)
- Chaperone (1)
- Circuit simulation (1)
- Coal (1)
- Collagen fibrils (1)
- Comparative simulation (1)
- Complex-valued eigenvalues (1)
- Connective tissues (1)
- Constitutive model (1)
- Control engineering (1)
- DAC (1)
- DC machines (1)
- Damage mechanics theory (1)
- Dark fermentation (1)
- Data analysis (1)
- Data visualization (1)
- Database (1)
- Discontinuous fractures (1)
- Dynamic modeling (1)
- E. coli detection (1)
- ESATAN-TMS (1)
- Eigenvalue trajectories (1)
- Electrochemical impedance spectroscopy (1)
- Electronic cigarettes (1)
- Electronic learning (1)
- Energy dispatch (1)
- Energy market (1)
- Enzyme coverage (1)
- Exponential Euler scheme, (1)
- Extension–twist coupling (1)
- Extracellular matrix (ECM) (1)
- Fault detection (1)
- Field-effect biosensor (1)
- Flutter (1)
- Focusing (1)
- Free-base nicotine (1)
- Fuel cell (1)
- Furnace (1)
- Fusion (1)
- German (1)
- Germany (1)
- H2 (1)
- Health management system (1)
- Hydrogenotrophic methanogens (1)
- Hydrolysis (1)
- Impedance analysis (1)
- Instruments (1)
- Interior transmission problem (1)
- LPBF (1)
- LQR (1)
- Li7La3Zr2O12 (1)
- LiGaO2 (1)
- Lifting propeller (1)
- Low field NMR (1)
- Low voltage (1)
- Lunar Surface (1)
- MOS (1)
- MPC (1)
- Manifolds (1)
- Master stamp (1)
- Matlab (1)
- Measurement (1)
- Mechanical stability (1)
- Metabolic shift (1)
- Methane (1)
- Methanogenesis (1)
- Mode converter (1)
- Modeling (1)
- Monitoring (1)
- Multianalyte detection (1)
- Muscle (1)
- Musculoskeletal system (1)
- Non-model-based Evaluation (1)
- Non-parallel fissures (1)
- Nonlinear eigenvalue problems (1)
- Obstacle avoidance (1)
- Online diagnostic (1)
- Online services (1)
- Open Data (1)
- Open source (1)
- Organic waste (1)
- P2G (1)
- PEM fuel cell (1)
- PPO (1)
- Parabolic SPDEs (1)
- Path planning (1)
- Penicillin (1)
- Photolithographic mimics (1)
- Physiology (1)
- Plasma (1)
- Plasma diagnostics (1)
- Power dissipation (1)
- Pre-culture (1)
- Pretreatment (1)
- Propeller whirl flutter (1)
- Quartz crystal microbalance (1)
- Raman spectroscopy (1)
- Refining (1)
- Reinforcement Learning (1)
- Relative air humidity (1)
- Renewable energy sources (1)
- Rotatory Inverted Pendulum (1)
- SLM (1)
- Simulation (1)
- Software packages (1)
- Surface imprinted polymer (1)
- Synchronous machines (1)
- Tapered ends (1)
- Thermal Model (1)
- Thermal analysis (1)
- Thermal conductivity (1)
- Throughput (1)
- Time-series (1)
- Training (1)
- Trapeze effect (1)
- Uniaxial compression test (1)
- Unmanned aerial vehicles (1)
- Unsteady aerodynamics (1)
- Weak organic acids (1)
- aquaculture (1)
- artificial olfactory image (1)
- biodegradable polymers (1)
- biomethane (1)
- bubble column (1)
- carbon dioxide removal (1)
- catalytic metal (1)
- chip-based sensor setup (1)
- climate neutrality (1)
- coupled Néel–Brownian relaxation dynamics (1)
- direct air capture (1)
- economics (1)
- electrical conductivity of liquids (1)
- electronic nose (1)
- field-effect structure (1)
- frequency mixing magnetic detection (1)
- ga-doping (1)
- garnet solid electrolyte (1)
- gas sensor (1)
- glycine (1)
- hydrogels (1)
- hydrogen peroxide (1)
- impedance spectroscopy (1)
- key performance indicators (1)
- large language models (1)
- light-addressing technologies (1)
- magnetic biosensing (1)
- magnetic nanoparticles (1)
- magnetic relaxation (1)
- metal-oxide-semiconductor structure (1)
- methanation (1)
- microfluidics (1)
- micromagnetic simulation (1)
- micronutrients (1)
- negative emissions (1)
- optical spore trapping (1)
- plug flow reactor (1)
- polyaspartic acid (1)
- rollout (1)
- scanned light pulse technique (1)
- semantic role labeling (1)
- solid-state battery (1)
- speaker attribution (1)
- sterilization conditions (1)
- superabsorbent polymers (1)
- swelling properties (1)
- temperature (1)
- thermometry (1)
- visualization (1)
Analyzing electroencephalographic (EEG) time series can be challenging, especially with deep neural networks, due to the large variability among human subjects and often small datasets. To address these challenges, various strategies, such as self-supervised learning, have been suggested, but they typically rely on extensive empirical datasets. Inspired by recent advances in computer vision, we propose a pretraining task termed "frequency pretraining" to pretrain a neural network for sleep staging by predicting the frequency content of randomly generated synthetic time series. Our experiments demonstrate that our method surpasses fully supervised learning in scenarios with limited data and few subjects, and matches its performance in regimes with many subjects. Furthermore, our results underline the relevance of frequency information for sleep stage scoring, while also demonstrating that deep neural networks utilize information beyond frequencies to enhance sleep staging performance, which is consistent with previous research. We anticipate that our approach will be advantageous across a broad spectrum of applications where EEG data is limited or derived from a small number of subjects, including the domain of brain-computer interfaces.