Seizure prediction using adaptive neuro-fuzzy inference system.

In this study, we present a neuro-fuzzy approach of seizure prediction from invasive Electroencephalogram (EEG) by applying adaptive neuro-fuzzy inference system (ANFIS). Three nonlinear seizure predictive features were extracted from a patient's data obtained from the European Epilepsy Database, one of the most comprehensive EEG database for epilepsy research. A total of 36 hours of recordings including 7 seizures was used for analysis. The nonlinear features used in this study were similarity index, phase synchronization, and nonlinear interdependence. We designed an ANFIS classifier constructed based on these features as input. Fuzzy if-then rules were generated by the ANFIS classifier using the complex relationship of feature space provided during training. The membership function optimization was conducted based on a hybrid learning algorithm. The proposed method achieved highest sensitivity of 80% with false prediction rate as low as 0.46 per hour.

Preliminary results of mental workload and task engagement assessment using electroencephalogram in a space suit.

In this paper, we present preliminary results of subject's mental workload and task engagement assessment in an experimental space suit. We have quantified the mental workload and task engagement based on changes in electroencephalogram (EEG). EEG signals were collected from subjects scalp using a commercial wireless EEG device in two experimental conditions - when subjects did not wear space suit (control condition) and when subjects wore space suit. Brain state changes were estimated and compared with the direct responses for different tasks and different conditions. We found that the spacesuit experiment introduced a greater mental workload where subject's stress levels were higher than control experiment.

Changes in dynamical characteristics of epileptic EEG in rats using recurrence quantification analysis.

In this paper, we used Recurrence Quantification Analysis (RQA) in order to study pre-epileptic characteristics in rat's EEG recordings. Four adult rats were used to collect epileptic EEG data in an experiment of animal model of epilepsy. Three RQA measures, recurrence rate, determinism, and entropy were calculated from EEG recordings from rats. A moving average filter was used to identify the decreasing trend in pre-epileptic dynamics which will be useful early detection of seizures.

Fuzzy rule-based seizure prediction based on correlation dimension changes in intracranial EEG.

In this paper, we present a method for epileptic seizure prediction from intracranial EEG recordings. We applied correlation dimension, a nonlinear dynamics based univariate characteristic measure for extracting features from EEG segments. Finally, we designed a fuzzy rule-based system for seizure prediction. The system is primarily designed based on expert's knowledge and reasoning. A spatial-temporal filtering method was used in accordance with the fuzzy rule-based inference system for issuing forecasting alarms. The system was evaluated on EEG data from 10 patients having 15 seizures.

Human performance evaluation based on EEG signal analysis: a prospective review.

Electroencephalogram (EEG) signal, the signature of brain activity, can be used to quantify for human performance evaluation. There are ongoing efforts by scientists and researchers in this area. Different traditional and novel signal processing and analysis methods have been applied to evaluate performance, mental workload, and task engagement based on EEG signals. Linear change in the indices with the increase in task difficulty was reported. In addition, EEG index has been used as parameter for performance optimization. In this review article, we will discuss briefly the literature on human performance estimation based on some physiological parameters, EEG in particular. In this paper, the current state of the research field is presented and possible future research options are discussed.