2013
DOI: 10.1088/1741-2560/10/3/031001
|View full text |Cite
|
Sign up to set email alerts
|

Abstract: In recent years, numerous brain-computer interfaces (BCIs) based on motor-imagery have been proposed which incorporate features such as adaptive classification, error detection and correction, fusion with auxiliary signals and shared control capabilities. Due to the added complexity of such algorithms, the evaluation strategy and metrics used for analysis must be carefully chosen to accurately represent the performance of the BCI. In this article, metrics are reviewed and contrasted using both simulated exampl… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
2
1

Citation Types

0
71
0
1

Year Published

2015
2015
2022
2022

Publication Types

Select...
4
2
2

Relationship

2
6

Authors

Journals

citations
Cited by 88 publications
(72 citation statements)
references
References 43 publications
0
71
0
1
Order By: Relevance
“…Detection of ErrP has been the one of the representative error prevention techniques, and many previous literatures have demonstrated that the ErrP detection can improve the overall performance of various BCI systems [38][39][40][41]57,58]. In particular, Combaz et al [38] and Schmidt et al [40] developed visual spellers with an automatic error-correction function based on ErrP detection.…”
Section: Discussionmentioning
confidence: 99%
See 1 more Smart Citation
“…Detection of ErrP has been the one of the representative error prevention techniques, and many previous literatures have demonstrated that the ErrP detection can improve the overall performance of various BCI systems [38][39][40][41]57,58]. In particular, Combaz et al [38] and Schmidt et al [40] developed visual spellers with an automatic error-correction function based on ErrP detection.…”
Section: Discussionmentioning
confidence: 99%
“…Recently, Spüler et al [41] confirmed that ErrP could be used to increase the overall bit rate of a P300-based mental speller. The ErrP-based error correction method has an advantage over our eye-gaze-based error prevention approach in that it can be applied to various speller types regardless of the BCI paradigms such as motor imagery, P300, and SSVEP [38][39][40][41]57,58]. However, in contrast to our eye gazebased error prevention method, the ErrP-based error correction method can detect typing errors only after presenting the mistyped characters on the screen, which spends unnecessary time.…”
Section: Discussionmentioning
confidence: 99%
“…In addition, the efficiency of a BCI system is often measured using the Information Transfer Rate (ITR), which measures how much information can be conveyed by the system in a given time (e.g., in Bits per second) (Wolpaw, 2002). Thomas et al (2013) listed several metrics used to quantify the performance of a BCI and evoked the evaluation strategies employed to compare the performance of two or more BCI. They suggested the following metrics to measure the performance of BCI based on synchronised control (i.e., the system is periodically available to the user when it is on, but does not support the non-control): accuracy (i.e., classification error), kappa, bit−rate (Wolpaw or Nykopp), confusion matrix and task specificity.…”
Section: Usabilitymentioning
confidence: 99%
“…Many potential BCI applications exist [1]; one of them is the P300 speller, an application that uses the P300 Event-Related Potential and enables the user to spell words by observing a screen of flashing letters.…”
Section: Introductionmentioning
confidence: 99%