Reliability of brain computer interface language sample transcription procedures

Woodworth

2021

PM&R

Self Cite

Background: Brain-computer interface (BCI) technology is an emerging access method to augmentative and alternative communication (AAC) devices. Objectives: To identify, in the early stages of research and development, the perceptions and considerations of interprofessional practice (IPP) team members regarding features and functions for an AAC-BCI device. Design: Qualitative research methodology applying a grounded theory approach using focus groups with a follow-up survey of participants using NVivo analysis software supporting inductive coding of transcription data. Setting: Focus groups held at university, clinic, and industry conference rooms. Discussion was stimulated by a 14-minute video on an AAC-BCI device prototype. The prototype hardware and electroencephalography (EEG) gel and dry electrode headgear were on display. Participants: Convenience sample of practitioners providing rehabilitation or clinical services to individuals with severe communication disorders and movement impairments who use AAC and/or other assistive technology. Interventions: Not applicable. Main Outcome Measures: Descriptive statistics using thematic analysis of participants' opinions, input, and feedback on the ideal design for a noninvasive, EEG-based P300 AAC-BCI device. Results: Interrater and interjudge reliability were at 98% and 100%, respectively, for transcription and researcher coding. Triangulation of multiple data sources supported theme and subtheme identification that included design features, set-up and calibration, services, and effectiveness. An AAC device with BCI access was unanimously confirmed (100%) as a desirable commercial product. Participants felt that the AAC-BCI prototype appeared effective for meeting daily communication needs (75%). Results showed that participants' preference on headgear types would change based on accuracy (91%) and rate (83%) of performance. A data-logging feature was considered beneficial by 100% of participants. Conclusions: IPP teams provided critical impressions on design, services, and features for a commercial AAC-BCI device. Expressed feature and function preferences showed dependence on communication accuracy, rate, and effectiveness. This provides vital guidance for successful clinical deployment.

Section: Discussionmentioning

confidence: 99%

Interprofessional Practitioners' Opinions on Features and Services for an Augmentative and Alternative Communication Brain‐Computer Interface Device

Hill

Woodworth

2021

PM&R

Self Cite

“…The BCI provided access to communication using WordPad and email programs along with access to the internet. Analysis of daily novel text generated data found a wide range for the percent of actual BCI time used for communication purposes [3]. However, use of communication tools (i.e.…”

Section: Discussionmentioning

confidence: 99%

“…[1, 10, 11]), perhaps because it has minimal cognitive load and allows comparison of subtle effects of BCI design. Occasionally, studies permit subjects to use the BCI to generate their own messages [10], but message content and task-dependent performance is only starting to be analyzed [3]. However, copy-spelling models only a small portion of a functional communication task.…”

Section: Introductionmentioning

confidence: 99%

“…Several language sampling procedures have been developed and standardized that allow researchers to collect, analyze, compare and contrast the spoken, written and/or augmentative and alternative communication (AAC) language samples of adults. Common, routine tasks include picture description [18], interviewing [19], story retelling [20], and natural daily conversation [3]. Use of a picture description task, e.g.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Effects of text generation on P300 brain-computer interface performance

Brain-Computer Interfaces

Alcaide-Aguirre

Hill

2016

Self Cite

Brain-computer interfaces (BCIs) are intended to provide independent communication for those with the most severe physical impairments. However, development and testing of BCIs is typically conducted with copy-spelling of provided text, which models only a small portion of a functional communication task. This study was designed to determine how BCI performance is affected by novel text generation. We used a within-subject single-session study design in which subjects used a BCI to perform copy-spelling of provided text and to generate self-composed text to describe a picture. Additional off-line analysis was performed to identify changes in the event-related potentials that the BCI detects and to examine the effects of training the BCI classifier on task-specific data. Accuracy was reduced during the picture description task; (t(8)=2.59 p=0.0321). Creating the classifier using self-generated text data significantly improved accuracy on these data; (t(7)=−2.68, p=0.0317), but did not bring performance up to the level achieved during copy-spelling. Thus, this study shows that the task for which the BCI is used makes a difference in BCI accuracy. Task-specific BCI classifiers are a first step to counteract this effect, but additional study is needed.

“…52, 53 Overall, BCIs for communication have been successful in the laboratory, and for long-term functional use. 19, 54 …”

Section: Communicationmentioning

confidence: 99%

Brain-Computer Interface: Current and Emerging Rehabilitation Applications

Daly

Archives of Physical Medicine and Rehabilitation

2015