A Comparison of Classification Techniques to Predict Brain-Computer Interfaces Accuracy Using Classifier-Based Latency Estimation

P300-based Brain-Computer Interface (BCI) performance is vulnerable to latency jitter. To investigate the role of latency jitter on BCI system performance, we proposed the classifier-based latency estimation (CBLE) method. In our previous study, CBLE was based on least-squares (LS) and stepwise line...

Full description

Bibliographic Details
Main Authors: Md Rakibul Mowla, Jesus D. Gonzalez-Morales, Jacob Rico-Martinez, Daniel A. Ulichnie, David E. Thompson
Format: Article
Language:English
Published: MDPI AG 2020-10-01
Series:Brain Sciences
Subjects:
Online Access:https://www.mdpi.com/2076-3425/10/10/734
id doaj-9f47bf74613d4712987b4ec470535890
record_format Article
spelling doaj-9f47bf74613d4712987b4ec4705358902020-11-25T03:55:07ZengMDPI AGBrain Sciences2076-34252020-10-011073473410.3390/brainsci10100734A Comparison of Classification Techniques to Predict Brain-Computer Interfaces Accuracy Using Classifier-Based Latency EstimationMd Rakibul Mowla0Jesus D. Gonzalez-Morales1Jacob Rico-Martinez2Daniel A. Ulichnie3David E. Thompson4Mike Wiegers Department of Electrical & Computer Engineering, Kansas State University, Manhattan, KS 66506, USAMike Wiegers Department of Electrical & Computer Engineering, Kansas State University, Manhattan, KS 66506, USAMike Wiegers Department of Electrical & Computer Engineering, Kansas State University, Manhattan, KS 66506, USADepartment of Biomedical Engineering, Wichita State University, Wichita, KS 67260, USAMike Wiegers Department of Electrical & Computer Engineering, Kansas State University, Manhattan, KS 66506, USAP300-based Brain-Computer Interface (BCI) performance is vulnerable to latency jitter. To investigate the role of latency jitter on BCI system performance, we proposed the classifier-based latency estimation (CBLE) method. In our previous study, CBLE was based on least-squares (LS) and stepwise linear discriminant analysis (SWLDA) classifiers. Here, we aim to extend the CBLE method using sparse autoencoders (SAE) to compare the SAE-based CBLE method with LS- and SWLDA-based CBLE. The newly-developed SAE-based CBLE and previously used methods are also applied to a newly-collected dataset to reduce the possibility of spurious correlations. Our results showed a significant (<inline-formula><math display="inline"><semantics><mrow><mi>p</mi><mo><</mo><mn>0.001</mn></mrow></semantics></math></inline-formula>) negative correlation between BCI accuracy and estimated latency jitter. Furthermore, we also examined the effect of the number of electrodes on each classification technique. Our results showed that on the whole, CBLE worked regardless of the classification method and electrode count; by contrast the effect of the number of electrodes on BCI performance was classifier dependent.https://www.mdpi.com/2076-3425/10/10/734brain-computer interfaces (BCI)classification methodsP300 spellerP3 latency estimationsparse autoencoders (SAE)
collection DOAJ
language English
format Article
sources DOAJ
author Md Rakibul Mowla
Jesus D. Gonzalez-Morales
Jacob Rico-Martinez
Daniel A. Ulichnie
David E. Thompson
spellingShingle Md Rakibul Mowla
Jesus D. Gonzalez-Morales
Jacob Rico-Martinez
Daniel A. Ulichnie
David E. Thompson
A Comparison of Classification Techniques to Predict Brain-Computer Interfaces Accuracy Using Classifier-Based Latency Estimation
Brain Sciences
brain-computer interfaces (BCI)
classification methods
P300 speller
P3 latency estimation
sparse autoencoders (SAE)
author_facet Md Rakibul Mowla
Jesus D. Gonzalez-Morales
Jacob Rico-Martinez
Daniel A. Ulichnie
David E. Thompson
author_sort Md Rakibul Mowla
title A Comparison of Classification Techniques to Predict Brain-Computer Interfaces Accuracy Using Classifier-Based Latency Estimation
title_short A Comparison of Classification Techniques to Predict Brain-Computer Interfaces Accuracy Using Classifier-Based Latency Estimation
title_full A Comparison of Classification Techniques to Predict Brain-Computer Interfaces Accuracy Using Classifier-Based Latency Estimation
title_fullStr A Comparison of Classification Techniques to Predict Brain-Computer Interfaces Accuracy Using Classifier-Based Latency Estimation
title_full_unstemmed A Comparison of Classification Techniques to Predict Brain-Computer Interfaces Accuracy Using Classifier-Based Latency Estimation
title_sort comparison of classification techniques to predict brain-computer interfaces accuracy using classifier-based latency estimation
publisher MDPI AG
series Brain Sciences
issn 2076-3425
publishDate 2020-10-01
description P300-based Brain-Computer Interface (BCI) performance is vulnerable to latency jitter. To investigate the role of latency jitter on BCI system performance, we proposed the classifier-based latency estimation (CBLE) method. In our previous study, CBLE was based on least-squares (LS) and stepwise linear discriminant analysis (SWLDA) classifiers. Here, we aim to extend the CBLE method using sparse autoencoders (SAE) to compare the SAE-based CBLE method with LS- and SWLDA-based CBLE. The newly-developed SAE-based CBLE and previously used methods are also applied to a newly-collected dataset to reduce the possibility of spurious correlations. Our results showed a significant (<inline-formula><math display="inline"><semantics><mrow><mi>p</mi><mo><</mo><mn>0.001</mn></mrow></semantics></math></inline-formula>) negative correlation between BCI accuracy and estimated latency jitter. Furthermore, we also examined the effect of the number of electrodes on each classification technique. Our results showed that on the whole, CBLE worked regardless of the classification method and electrode count; by contrast the effect of the number of electrodes on BCI performance was classifier dependent.
topic brain-computer interfaces (BCI)
classification methods
P300 speller
P3 latency estimation
sparse autoencoders (SAE)
url https://www.mdpi.com/2076-3425/10/10/734
work_keys_str_mv AT mdrakibulmowla acomparisonofclassificationtechniquestopredictbraincomputerinterfacesaccuracyusingclassifierbasedlatencyestimation
AT jesusdgonzalezmorales acomparisonofclassificationtechniquestopredictbraincomputerinterfacesaccuracyusingclassifierbasedlatencyestimation
AT jacobricomartinez acomparisonofclassificationtechniquestopredictbraincomputerinterfacesaccuracyusingclassifierbasedlatencyestimation
AT danielaulichnie acomparisonofclassificationtechniquestopredictbraincomputerinterfacesaccuracyusingclassifierbasedlatencyestimation
AT davidethompson acomparisonofclassificationtechniquestopredictbraincomputerinterfacesaccuracyusingclassifierbasedlatencyestimation
AT mdrakibulmowla comparisonofclassificationtechniquestopredictbraincomputerinterfacesaccuracyusingclassifierbasedlatencyestimation
AT jesusdgonzalezmorales comparisonofclassificationtechniquestopredictbraincomputerinterfacesaccuracyusingclassifierbasedlatencyestimation
AT jacobricomartinez comparisonofclassificationtechniquestopredictbraincomputerinterfacesaccuracyusingclassifierbasedlatencyestimation
AT danielaulichnie comparisonofclassificationtechniquestopredictbraincomputerinterfacesaccuracyusingclassifierbasedlatencyestimation
AT davidethompson comparisonofclassificationtechniquestopredictbraincomputerinterfacesaccuracyusingclassifierbasedlatencyestimation
_version_ 1724470652542386176