A comparison of feature extraction strategies using wavelet dictionaries and feature selection methods for single trial P300-based BCI.

The P300 component of event-related potentials (ERPs) is widely used in the implementation of brain computer interfaces (BCI). In this context, one of the main issues to solve is the binary classification problem that entails differentiating between electroencephalographic (EEG) signals with and without P300. Given the particularly unfavorable signal-to-noise ratio (SNR) in the single-trial detection scenario, this is a challenging problem in the pattern recognition field. To the best of our knowledge, there are no previous experimental studies comparing feature extraction and selection methods for single trial P300-based BCIs using unified criteria and data. In order to improve the performance and robustness of single-trial classifiers, we analyzed and compared different alternatives for the feature generation and feature selection blocks. We evaluated different orthogonal decompositions based on the wavelet transform for feature extraction, as well as different filter, wrapper, and embedded alternatives for feature selection. Accuracies over 75% were obtained for most of the analyzed strategies with a relatively low computational cost, making them attractive for a practical BCI implementation using inexpensive hardware. Graphical Abstract Experiments performed for P300 detection.

Effect of static and dynamic exercise on heart rate and blood pressure variabilities.

PURPOSE: This study examines the effect of static and dynamic leg exercises on heart rate variability (HRV) and blood pressure variability (BPV) in humans. METHODS: 10 healthy male subjects were studied at rest, during static exercise performed at 30% of maximal voluntary contraction (SX30), and during dynamic cycling exercises done at 30% of VO2max (DX30) and at 60% of VO2max (DX60). Respiration, heart rate, and blood pressure signals were digitized to analyze temporal and spectral parameters involving short and overall indexes (SD, deltaRANGE, RMSSD, Total power), power of the low (LF), middle (MF), and high (HF) frequency components, and the baroreceptor sensitivity by the alphaMF index. RESULTS: During SX30, indexes of HRV as SD, deltaRANGE, Total power, and MF in absolute units increased in relation with rest values and were significantly higher (P < 0.001) than during DX30 and DX60; HF during SX30, in normalized and absolute units, was not different of the rest condition but was higher (P < 0.001) than HF during DX30 and DX60. Parameters of BPV as SD and deltaRANGE increased (P < 0.001) during both type of exercises, and significant (P < 0.01) increments were observed on MF during SX30 and DX30; systolic HF was attenuated during DX30 (P < 0.05), whereas diastolic HF was augmented during DX60 (P < 0.001). Compared with rest condition, the alphaMF index decreased (P < 0.01) only during dynamic exercises. CONCLUSION: Because HRV and BPV response is different when induced by static or dynamic exercise, differences in the autonomic activity can be advised. Instead of the vagal withdrawal and sympathetic augmentation observed during dynamic exercise, the increase in the overall HRV and the MF component during static exercise suggest an increased activity of both autonomic branches.

The effects of body position, controlled breathing and exercise on the heart rate variability parameters in healthy subjects.

The behavior of temporal and spectral parameters of the heart rate variability was determined during 5 different maneuvers in order to characterize the level of sympathetic-vagal activity. The discriminating capacity of two spectral parameter estimation schemes were compared, and the respiratory influence was evaluated. One hundred and ten records of instantaneous heart rate and respiratory amplitude were analyzed, both in temporal and spectral perspectives. The records were obtained from 22 healthy subjects, under a five-stage protocol: supine, controlled breathing, standing, exercise, and recovery. A discriminating and characteristic behavior was found among the maneuvers, specifically for the dispersion parameters, low and intermediate partial components. The two integration-normalization procedures used in the estimation of the spectral components showed similarities, producing a functional interpretation independent of the selected procedure. The division of the low frequency component in two separate bands, allowed an improved discriminating capacity. The respiration had an important influence in the controlled breathing maneuver while being lower in the rest of the stages. In conclusion, the maneuvers determined a typical behavior of the dispersion and spectral parameters (low and intermediate partial components) of the heart rate variability, showing adequate distinctive levels in the sympathetic-vagal activity for each maneuver.