Reducing the number of features for seizure prediction of spectral power in intracranial EEG

Keyphrases

• Seizure Prediction 100%
• Number of Features 100%
• Spectral Power 100%
• Prediction Rate 100%
• Intracranial EEG (iEEG) 100%
• Implantable Devices 66%
• False Positive Rate 66%
• Time Differential 66%
• Nonlinear Support Vector Machine 66%
• Reduced Complexity Algorithm 66%
• High Sensitivity 33%
• Low Power 33%
• Reduced Complexity 33%
• Quality of Life Improvement 33%
• People with Epilepsy 33%
• Algorithm Complexity 33%
• Prediction Algorithms 33%
• Moving Average 33%
• Linear Features 33%
• Differential Features 33%
• Interictal 33%
• Support Vector Machine Classification 33%
• Freiburg 33%
• Recursive Feature Elimination 33%
• High Complexity 33%
• Linear Support Vector Machine 33%
• Preictal 33%
• Powered Support 33%
• Post-processing Filter 33%
• Pre-seizure State 33%
• Biomedical Engineering Applications 33%

Neuroscience

• Electroencephalography 100%
• Support Vector Machine 100%
• Seizure Prediction 100%

Pharmacology, Toxicology and Pharmaceutical Science

• Elimination 100%