Search

A Brain Computer Interface (BCI) allows communication for impaired people unable to express their intention with common channels. Electroencephalography (EEG) represents an effective tool to allow the implementation of a BCI. The present paper describes a modular framework for the implementation of the graphic interface for binary BCIs based on the selection of symbols in a table. The proposed system is also designed to reduce the time required for writing text. This is made by including a motivational tool, necessary to improve the quality of the collected signals, and by containing a predictive module based on the frequency of occurrence of letters in a language, and of words in a dictionary. The proposed framework is described in a top-down approach through its modules: signal acquisition, analysis, classification, communication, visualization, and predictive engine. The framework, being modular, can be easily modified to personalize the graphic interface to the needs of the subject who has to use the BCI and it can be integrated with different classification strategies, communication paradigms, and dictionaries/languages. The implementation of a scenario and some experimental results on healthy subjects are also reported and discussed: the modules of the proposed scenario can be used as a starting point for further developments, and application on severely disabled people under the guide of specialized personnel. [ABSTRACT FROM AUTHOR]

The article discusses a study which compared the performance of Information Extraction (IE) algorithms. Topics covered include the advantages and disadvantages of Injury Surveillance Systems based on traditional hospital records or clinical data, the common sources of injury surveillance data and the integration of clinical based registries with unconventional data sources like newspaper stories.

The article discusses the application of Inter-Greedy (IG) approach in the measurement of intima-media thickness (IMT) of carotid arteries in ultrasound images. Topics covered include the deployment of a fusion technique to remove bias in segmentation algorithms, the estimation of IMT through fused boundaries from multiple sources, and the mean error of Inter-Greedy technique. Also mentioned is the ability of the Inter-Greedy technique to enhance segmentation and measurement performance.