Principal component analysis and manifold learning techniques for the design of brain-computer interfaces based on steady-state visually evoked potentials

被引:6
|
作者
Yesilkaya, Bartu [1 ]
Sayilgan, Ebru [2 ]
Yuce, Yilmaz Kemal [3 ]
Perc, Matjaz [4 ,5 ,6 ,7 ,8 ]
Isler, Yalcin [1 ]
机构
[1] Izmir Katip Celebi Univ, Dept Biomed Engn, Balatcik Campus, TR-35620 Izmir, Turkiye
[2] Izmir Univ Econ, Dept Mechatron Engn, TR-35330 Izmir, Turkiye
[3] Alanya Alaaddin Keykubat Univ, Dept Comp Engn, TR-07425 Antalya, Turkiye
[4] Univ Maribor, Fac Nat Sci & Math, Koroska Cesta 160, Maribor 2000, Slovenia
[5] China Med Univ, China Med Univ Hosp, Dept Med Res, Taichung, Taiwan
[6] Alma Mater Europaea, Slovenska Ulica 17, Maribor 2000, Slovenia
[7] Complex Sci Hub Vienna, Josefstadterstr 39, A-1080 Vienna, Austria
[8] Kyung Hee Univ, Dept Phys, 26 Kyungheedae Ro, Seoul, South Korea
关键词
Manifold learning; Brain-computer interface; Steady-state visual evoked potential; Principal component analysis; Feature reduction;
D O I
10.1016/j.jocs.2023.102000
中图分类号
TP39 [计算机的应用];
学科分类号
081203 ; 0835 ;
摘要
Steady-state visually evoked potentials (SSVEP) are stochastic and nonstationary bioelectric signals. Because of these properties, it is difficult to achieve high classification accuracy, especially when many considered features lead to a complex structure. We therefore propose a manifold learning framework to decrease the number of features and to classify SSVEP data by comparing lower dimensional matrices with well-known machine learning algorithms. We use the AVI-SSVEP Dataset, which includes stimuli at seven different frequencies and 15360 samples per person. The SSVEP features are extracted from relevant and distinctive frequency -domain, time-domain, and time-frequency domain properties, creating a total of 55 feature vectors. We then analyze and compare five divergent manifold learning methods with respect to their performance on nine different machine-learning algorithms. Among all considered manifold learning methods, we show that the Principal Component Analysis has the best classifier performance with an average of 22 components. Moreover, the Naive Bayes classifier with the Principal Component Analysis achieves the maximum accuracy of 50.0%-80.95% for a 7-class classification problem. Our research thus shows that the proposed analytical framework can significantly improve the decoding accuracy of 7-class SSVEP problems, and that it exhibits notable robustness and efficiency for small group datasets.
引用
收藏
页数:9
相关论文
共 50 条
  • [1] Use of Phase in Brain-Computer Interfaces based on Steady-State Visual Evoked Potentials
    Lopez-Gordo, M. A.
    Prieto, A.
    Pelayo, F.
    Morillas, C.
    [J]. NEURAL PROCESSING LETTERS, 2010, 32 (01) : 1 - 9
  • [2] Advancing the detection of steady-state visual evoked potentials in brain-computer interfaces
    Abu-Alqumsan, Mohammad
    Peer, Angelika
    [J]. JOURNAL OF NEURAL ENGINEERING, 2016, 13 (03)
  • [3] Processing and Decoding Steady-State Visual Evoked Potentials for Brain-Computer Interfaces
    Chumerin, Nikolay
    Manyakov, Nikolay V.
    van Vliet, Marijn
    Robben, Arne
    Combaz, Adrien
    Van Hulle, Marc M.
    [J]. DIGITAL IMAGE AND SIGNAL PROCESSING FOR MEASUREMENT SYSTEMS, 2012, : 1 - 33
  • [4] Steady-state somatosensory evoked potentials:: Suitable brain signals for brain-computer interfaces?
    Müller-Putz, GR
    Scherer, R
    Neuper, C
    Pfurtscheller, G
    [J]. IEEE TRANSACTIONS ON NEURAL SYSTEMS AND REHABILITATION ENGINEERING, 2006, 14 (01) : 30 - 37
  • [5] Brain-computer Interface Based on Steady-state Visual Evoked Potentials
    Friganovic, K.
    Medved, M.
    Cifrek, M.
    [J]. 2016 39TH INTERNATIONAL CONVENTION ON INFORMATION AND COMMUNICATION TECHNOLOGY, ELECTRONICS AND MICROELECTRONICS (MIPRO), 2016, : 391 - 395
  • [6] Evaluation of mother wavelets on steady-state visually-evoked potentials for triple-command brain-computer interfaces
    Sayilgan, Ebru
    Yuce, Yilmaz Kemal
    Isler, Yalcin
    [J]. TURKISH JOURNAL OF ELECTRICAL ENGINEERING AND COMPUTER SCIENCES, 2021, 29 (05) : 2263 - 2279
  • [7] Statistical Characterization of Steady-State Visual Evoked Potentials and Their Use in Brain-Computer Interfaces
    Lopez, Miguel A.
    Pelayo, Francisco
    Madrid, Eduardo
    Prieto, Alberto
    [J]. NEURAL PROCESSING LETTERS, 2009, 29 (03) : 179 - 187
  • [8] Waveform-Coded Steady-State Visual Evoked Potentials for Brain-Computer Interfaces
    Nakanishi, Masaki
    Tanji, Yutaro
    Tanaka, Toshihisa
    [J]. IEEE ACCESS, 2021, 9 : 144768 - 144775
  • [9] Multiple channel detection of steady-state visual evoked potentials for brain-computer interfaces
    Friman, Ola
    Volosyak, Ivan
    Graeser, Axel
    [J]. IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING, 2007, 54 (04) : 742 - 750
  • [10] Analysis of User Interaction with a Brain-Computer Interface Based on Steady-State Visually Evoked Potentials: Case Study of a Game
    de Arruda Leite, Harlei Miguel
    de Carvalho, Sarah Negreiros
    da Silva Costa, Thiago Bulhoes
    Attux, Romis
    Hornung, Heiko Horst
    Arantes, Dalton Soares
    [J]. COMPUTATIONAL INTELLIGENCE AND NEUROSCIENCE, 2018, 2018