EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Improved performance of fNIRS-BCI by stacking of deep learning-derived frequency domain features

Jamila Akhter iD, Hammad Nazeer iD, Noman Naseer iD, Rehan Naeem iD, Karam Dad Kallu, Jiye Lee, Seong Young Ko iD

PLOS ONE · 17.04.2025 · 20 (4) · e0314447

DOI: 10.1371/journal.pone.0314447 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PLOS ONE, online erschienen am 17.04.2025. Verfasst von Jamila Akhter und 6 weiteren Autor:innen. Beteiligte Einrichtung: Department of Mechatronics and Biomedical Engineering, Air University, Islamabad, Pakistan. Hauptthemen laut MEDLINE: Deep Learning, Brain-Computer Interfaces. Die Arbeit wurde bisher 7-mal zitiert (Crossref) und nennt 53 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Korea Institute for Advancement of Technology.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The functional near-infrared spectroscopy-based brain-computer interface (fNIRS-BCI) systems recognize patterns in brain signals and generate control commands, thereby enabling individuals with motor disabilities to regain autonomy. In this study hand gripping data is acquired using fNIRS neuroimaging system, preprocessing is performed using nirsLAB and features extraction is performed using deep learning (DL) Algorithms. For feature extraction and classification stack and fft methods are proposed. Convolutional neural networks (CNN), long short-term memory (LSTM), and bidirectional long-short-term memory (Bi-LSTM) are employed to extract features. The stack method classifies these features using a stack model and the fft method enhances features by applying fast Fourier transformation which is followed by classification using a stack model. The proposed methods are applied to fNIRS signals from twenty participants engaged in a two-class hand-gripping motor activity. The classification performance of the proposed methods is compared with conventional CNN, LSTM, and Bi-LSTM algorithms and one another. The proposed fft and stack methods yield 90.11% and 87.00% classification accuracies respectively, which are significantly higher than those achieved by CNN (85.16%), LSTM (79.46%), and Bi-LSTM (81.88%) conventional algorithms. The results show that the proposed stack and fft methods can be effectively used for the classification of the two and three-class problems in fNIRS-BCI applications.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Deep LearningBrain-Computer InterfacesHumansSpectroscopy, Near-InfraredMaleAdultAlgorithmsFemaleNeural Networks, ComputerYoung AdultHand Strength

Autor:innen und Institutionen

  • Department of Mechatronics and Biomedical Engineering, Air University, Islamabad, Pakistan.
  • MeRIC-Lab (Medical Robotics & Intelligent Control Laboratory), School of Mechanical Engineering, Chonnam National University, Gwangju, South Korea.

ORCID-Profile: Jamila Akhter, Hammad Nazeer, Noman Naseer, Rehan Naeem, Seong Young Ko

Förderung

  • Korea Institute for Advancement of Technology

Zitieren

APA

Akhter, J., Nazeer, H., Naseer, N., Naeem, R., Kallu, K. D., Lee, J., & Ko, S. Y. (2025). Improved performance of fNIRS-BCI by stacking of deep learning-derived frequency domain features. PLOS ONE, 20(4), e0314447. https://doi.org/10.1371/journal.pone.0314447

Vancouver

Akhter J, Nazeer H, Naseer N, Naeem R, Kallu KD, Lee J, et al. Improved performance of fNIRS-BCI by stacking of deep learning-derived frequency domain features. PLOS ONE. 2025;20(4):e0314447. doi:10.1371/journal.pone.0314447

BibTeX
@article{Akhter2025,
  author = {Akhter, Jamila and Nazeer, Hammad and Naseer, Noman and Naeem, Rehan and Kallu, Karam Dad and Lee, Jiye and Ko, Seong Young},
  title = {Improved performance of fNIRS-BCI by stacking of deep learning-derived frequency domain features},
  journal = {PLOS ONE},
  year = {2025},
  volume = {20},
  number = {4},
  pages = {e0314447},
  doi = {10.1371/journal.pone.0314447}
}
Kostenlos starten →