PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
May 16, 20260 citationsOpen Access

A Deep Learning Framework for Trial-Level Assessment of Task-Evoked fNIRS Motor Activation Quality

View Full Paper
HNHenry NduANAugusta Nneka Ndu

Key Points

  • The research aims to develop a deep learning framework for evaluating the quality of fNIRS motor activation during tasks.
  • Utilized a trial-level framework with physiologically motivated HbO/HbR pseudo-labeling.
  • Evaluated multiple architectures including CNN-BiLSTM, ResNet1D, and HTCNet.
  • Applied leave-one-subject-out validation on a publicly available hand-gripping fNIRS dataset.
  • Achieved improved assessment accuracy through the proposed deep learning frameworks.
  • Demonstrated effective utilization of temporal deep learning models in fNIRS data analysis.
  • Highlighted the strengths of various architectures in the evaluation of motor activation quality.

Abstract

This manuscript presents a trial-level framework for assessing task-evoked fNIRS motor activation quality using physiologically motivated HbO/HbR pseudo-labelling and temporal deep learning models. The study evaluates CNN-BiLSTM, ResNet1D, and HTCNet architectures under leave-one-subject-out validation using a publicly available hand-gripping fNIRS dataset.

Ask AI
Helpful
Bookmark
Share
View Full Paper

Cite This Study

Ndu et al. (2026) studied this question.

synapsesocial.com/papers/6a080b27a487c87a6a40d390https://doi.org/10.5281/zenodo.20180290
Ask AI
Helpful
Bookmark
Share
View Full Paper

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1Enhancing fNIRS ‐Based Decoding: A Heterogeneous Ensemble Network for Six‐Class Movement Classification2026
  2. 2CNN-based fNIRS signal quality assessment using short-time Fourier transform with short segment signals2024
  3. 3Unleashing the potential of fNIRS with machine learning: classification of fine anatomical movements to empower future brain-computer interface2024 · 22 citations
  4. 4DL-QC-fNIRS: a deep learning tool for automated quality control in functional near-infrared spectroscopy signals2026
  5. 5A deep convolutional neural network for estimating hemodynamic response function with reduction of motion artifacts in fNIRS2022 · 19 citations