Functional near-infrared sensing (fNIR) enables real-time, noninvasive monitoring of cognitive activity by measuring the brain's hemodynamic and metabolic responses. We have demonstrated the ability for non-vocal and non-physical communications through detecting directed changes in cognitive tasks. Building upon past research, this paper reports methods that allow the calibration of the fNIR oxygenation signal to better be used in more complex communicative and selection tasks. This work is then discussed in the context of a faster, continuous fNIR brain-computer interface framework.

No abstract

Controlling computers and other electronic devices using only one's thoughts is an exciting yet unlikely and distant reality for most people. However, for people with locked-in syndrome, their disabilities are so severe that they have no other alternatives. Applications that are consciously controlled using signals from the brain (called brain-computer interfaces, or BCIs) have been shown to restore some communication and environmental control for these individuals. Unfortunately, BCIs can be slow and tedious to learn or operate, reducing their effectiveness. This demonstration presents engaging BCI applications, including a video game and a digital painting program, that enable users to have fun while they improve their control over the brain signals required to use BCIs.

No abstract

BackgroundThe effective detection and comparison of orthologues is crucial for answering many questions in comparative genomics, phylogenetics and evolutionary biology. One of the most common methods for discovering orthologues is widely known as 'Reciprocal Blast'. While this method is simple when comparing only two genomes, performing a large-scale comparison of Multiple Genes across Multiple Taxa becomes a labor-intensive and inefficient task. The low efficiency of this complicated process limits the scope and breadth of questions that would otherwise benefit from this powerful method. FindingsHere we present RecBlast, an intuitive and easy-to-use pipeline that enables fast and easy discovery of orthologues along and across the evolutionary tree. RecBlast is capable of running heavy, large-scale and complex Reciprocal Blast comparisons across multiple genes and multiple taxa, in a completely automatic way. RecBlast is available as a cloud-based web server, which includes an easy-to-use user interface, implemented using cloud computing and an elastic and scalable server architecture. RecBlast is also available as a powerful standalone software supporting multi-processing for large datasets, and a cloud image which can be easily deployed on Amazon Web Services cloud. We also include sample results spanning 448 human genes, which illustrate the potential of RecBlast in detecting orthologues and in highlighting patterns and trends across multiple taxa. ConclusionsRecBlast provides a fast, inexpensive and valuable insight into trends and phenomena across distance phyla, and provides data, visualizations and directions for downstream analysis.RecBlast's fully automatic pipeline provides a new and intuitive discovery platform for researchers from any domain in biology who are interested in evolution, comparative genomics and phylogenetics, regardless of their computational skills.