Probability Distributions.- Linear Models for Regression.- Linear Models for Classification.- Neural Networks.- Kernel Methods.- Sparse Kernel Machines.- Graphical Models.- Mixture Models and EM.- Approximate Inference.- Sampling Methods.- Continuous Latent Variables.- Sequential Data.- Combining Models.

Pattern Recognition and Machine Learning

Steady-state visually evoked potentials: focus on essential paradigms and future perspectives.

Brain-computer interface (BCI) systems based on the steady-state visual evoked potential (SSVEP) provide higher information throughput and require shorter training than BCI systems using other brain signals. To elicit an SSVEP, a repetitive visual stimulus (RVS) has to be presented to the user. The RVS can be rendered on a computer screen by alternating graphical patterns, or with external light sources able to emit modulated light. The properties of an RVS (e.g., frequency, color) depend on the rendering device and influence the SSVEP characteristics. This affects the BCI information throughput and the levels of user safety and comfort. Literature on SSVEP-based BCIs does not generally provide reasons for the selection of the used rendering devices or RVS properties. In this paper, we review the literature on SSVEP-based BCIs and comprehensively report on the different RVS choices in terms of rendering devices, properties, and their potential influence on BCI performance, user safety and comfort.

/pdf/a-survey-of-stimulation-methods-used-in-ssvep-based-bcis-38omyw8mqs.pdf

A survey of stimulation methods used in SSVEP-based BCIs

In this paper, novel methods for detecting steady-state visual evoked potentials using multiple electroencephalogram (EEG) signals are presented. The methods are tailored for brain-computer interfacing, where fast and accurate detection is of vital importance for achieving high information transfer rates. High detection accuracy using short time segments is obtained by finding combinations of electrode signals that cancel strong interference signals in the EEG data. Data from a test group consisting of 10 subjects are used to evaluate the new methods and to compare them to standard techniques. Using 1-s signal segments, six different visual stimulation frequencies could be discriminated with an average classification accuracy of 84%. An additional advantage of the presented methodology is that it is fully online, i.e., no calibration data for noise estimation, feature extraction, or electrode selection is needed

Multiple Channel Detection of Steady-State Visual Evoked Potentials for Brain-Computer Interfaces

A textbook of physiology

This study proposes and validates a novel steady-state visual evoked potential (SSVEP) detection approach, multiview MAX-VAR canonical correlation, that finds a common unique subspace that encompasses all the SSVEP responses pertaining to a specific subject. The method employs a generalized canonical correlation framework that efficiently computes a projection matrix that optimizes test data to achieve higher SSVEP identification performance. We used a SSVEP benchmark dataset using a 40 target BCI experiment to evaluate the proposed method. The results demonstrate that the multiview MAX-VAR canonical correlation approach outperforms the compared methods with respect to both accuracy and information transfer rates (ITRs). From the statistical significance tests, it is observed that the proposed approach effectively achieves superior performance at short window lengths making it a propitious algorithm for real time brain computer interfaces (BCI).

Multiview MAX-VAR canonical correlation approach for enhancing SSVEP based BCIs

In recent decades, motor imagery (MI) based brain-computer interface (BCI) is acting as a rehabilitation tool for motor disabled people. But it has limited applications due to its lower classification performance. To improve it, this paper introduces band power (BP) and second order difference plot (SODP) for the detection of various motor imagery (MI) activities. First, filter bank technique was implemented to the signals and sets of sub-bands were generated. The BP was evaluated for all sub-bands. To study MI activities more effectively, SODP was applied to each sub-band and area of SODP was calculated. The features (i.e. band power and area of SODP) of all sub-bands were combined and the significant features $(\mathrm{p}\lt 0.05)$ were extracted from one-way analysis of variance (ANOVA). The significant features were fed into multi-class support vector machine (SVM) for the decoding of MI activities. BCI competition 2008 benchmark MI dataset-II-a was used to validate the proposed technique. The performance of the proposed technique was evaluated in terms of classification accuracy (%CA), precision (P), sensitivity (S) and F1-score. The results show that the present technique improved the performance of MI based BCI system and superior to the existing methods reported in the literature.

Classification of Motor Imagery Activities Using Band Power and Second Order Difference Plot

Near infrared (NIR) light in the wavelength range from 700 to 900 nm can pass through skin, bone and other tissues relatively easily. As a result, NIR techniques allow a noninvasive assessment of hemoglobin saturation for a wide range of applications, such as in the study of muscle metabolism, the diagnosis of vascular disorders, brain imaging, and breast cancer detection. Near infrared Spectroscopy (NIRS) is an effective tool to measure the hemoglobin concentration in the tissues, which can discriminate optically the oxy- and deoxy- hemoglobin species because of their different near-infrared absorption spectra. 
We have developed an NIRS probe consisting of a laser diode of 830 nm wavelength and a PIN photodiode in reflectance mode. We have selected a set of healthy volunteers (mean age 30, range 26-40 years) for the study. The probe is placed on forearm of each subject and the backscattered light intensity is measured by occluding the blood flow at 210, 110 and 85 mmHg pressures. Recovery time, peak time and time after 50% release of the cuff pressure are determined from the optical densities during the post occlusive state of forearm. 
These parameters are useful for determining the transient increase in blood flow after the release of blood occlusion. Clinically, the functional aspects of blood flow in the limbs could be evaluated noninvasively by NIRS.

M. Ramasubba Reddy

Papers

Multiview MAX-VAR canonical correlation approach for enhancing SSVEP based BCIs

Classification of Motor Imagery Activities Using Band Power and Second Order Difference Plot

Noninvasive measurement of postocclusive parameters in human forearm blood by near infrared spectroscopy