Font Size: a A A

Icon-based Communication Through a Brain Computer Interface

Posted on:2017-01-28Degree:Ph.DType:Dissertation
University:Northeastern UniversityCandidate:Ahani, AsiehFull Text:PDF
GTID:1448390005965021Subject:Biomedical engineering
Abstract/Summary:
Alternative and Augmentative Communication (AAC) is typically used by people with Severe Speech and Physical Impairment (SSPI) and is one of the main application areas for Brain Computer Interface (BCI) technology. The target population includes people with Cerebral Palsy (CP), Amyotrophic Lateral Sclerosis (ALS) and Locked-In Syndrome (LIS). Word-based AAC systems are mainly faster than letter-based counterparts and are usually supplemented by icons to aid the users. Those icon-based AAC systems that use binary signaling methods such as single click can convert into a single input BCI systems. The common way of displaying icons is combining hierarchical layouts with some form of scanning such as matrix speller paradigm. The scanning method help users identify their target icon on the screen, however it ties screen space to vocabulary size and navigation complexity, which may require users to make repetitive head, neck, or eye movements to visually locate their intended targets on the screen. Rapid Serial Visual Presentation (RSVP) is an alternative interface that minimizes required movement by displaying all icons at a fix location, one at a time. We have designed a system called "RSVP IconMessenger" that combines P300 signal detection with the icon-based semantic message construction system of RSVP-iconCHAT. The results of this study conducted with 10 healthy participants suggest that the system has potential as an AAC system in real-time typing applications. Furthermore we designed and implemented "IconMessenger" as an icon-based BCI-AAC system that combines ERP signal detection with a unified framework for different presentation paradigms including RSVP, matrix Row&Column Presentation (RCP) and matrix Single Character Presentation (SCP). IconMessenger also takes advantage of a unique semantic gram language model, incorporated tightly in the inference engine. In this study, we assess the ERP shape, classification accuracy and typing performance of different presentation paradigms on 10 healthy participants.
Keywords/Search Tags:AAC, Icon-based, Presentation
Related items