Most favorable stimulation duration in the sensorimotor cortex for fNIRS-based BCI

Khan, M. N. Afzal; Hong, Keum‐Shik

doi:10.1364/boe.434936

Cited by 14 publications

(5 citation statements)

References 91 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Investigating the dynamic relationship between the brain regions simultaneously activated during finger-tapping would be an interesting direction for the future study. In recent studies, different stimulation durations were investigated to find the appropriate duration that can shorten the command generation time [ 54 ]. Keeping in mind the findings of these studies, shorter stimulation durations will also be investigated in the future.…”

Section: Results and Discussionmentioning

confidence: 99%

Classification of Individual Finger Movements from Right Hand Using fNIRS Signals

Khan

Noori

Yazidi

et al. 2021

Sensors

View full text Add to dashboard Cite

Functional near-infrared spectroscopy (fNIRS) is a comparatively new noninvasive, portable, and easy-to-use brain imaging modality. However, complicated dexterous tasks such as individual finger-tapping, particularly using one hand, have been not investigated using fNIRS technology. Twenty-four healthy volunteers participated in the individual finger-tapping experiment. Data were acquired from the motor cortex using sixteen sources and sixteen detectors. In this preliminary study, we applied standard fNIRS data processing pipeline, i.e., optical densities conversation, signal processing, feature extraction, and classification algorithm implementation. Physiological and non-physiological noise is removed using 4th order band-pass Butter-worth and 3rd order Savitzky–Golay filters. Eight spatial statistical features were selected: signal-mean, peak, minimum, Skewness, Kurtosis, variance, median, and peak-to-peak form data of oxygenated haemoglobin changes. Sophisticated machine learning algorithms were applied, such as support vector machine (SVM), random forests (RF), decision trees (DT), AdaBoost, quadratic discriminant analysis (QDA), Artificial neural networks (ANN), k-nearest neighbors (kNN), and extreme gradient boosting (XGBoost). The average classification accuracies achieved were 0.75±0.04, 0.75±0.05, and 0.77±0.06 using k-nearest neighbors (kNN), Random forest (RF) and XGBoost, respectively. KNN, RF and XGBoost classifiers performed exceptionally well on such a high-class problem. The results need to be further investigated. In the future, a more in-depth analysis of the signal in both temporal and spatial domains will be conducted to investigate the underlying facts. The accuracies achieved are promising results and could open up a new research direction leading to enrichment of control commands generation for fNIRS-based brain-computer interface applications.

show abstract

Section: Results and Discussionmentioning

confidence: 99%

Classification of Individual Finger Movements from Right Hand Using fNIRS Signals

Khan

Noori

Yazidi

et al. 2021

Sensors

View full text Add to dashboard Cite

show abstract

“…Although the main effect in “Beta ~ duration” was non-significant, this study did not control the block duration, which might directly affect hemodynamic responses (Khan et al, 2020 ). To clear the confounding from block duration, future protocols need to consider the repetition of movements, as well as a potential plateau effect for blocks longer than 20 seconds (Afzal Khan and Hong, 2021 ).…”

Section: Discussionmentioning

confidence: 99%

Hemodynamic activity is not parsimoniously tuned to index-of-difficulty in movement with dual requirements on speed-accuracy

Ji,

Chen,

Qiao

et al. 2024

Front. Hum. Neurosci.

View full text Add to dashboard Cite

BackgroundReaching movements are crucial for daily living and rehabilitation, for which Fitts' Law describes a speed-accuracy trade-off that movement time increases with task difficulty. This study aims to investigate whether cortical activation in motor-related areas is directly linked to task difficulty as defined by Fitts' Law. Understanding this relationship provides a physiological basis for parameter selection in therapeutic exercises.MethodsSixteen healthy subjects performed 2D reaching movements using a rehabilitation robot, with their cortical responses detected using functional near-infrared spectroscopy (fNIRS). Task difficulty was manipulated by varying target size and distance, resulting in 3 levels of index-of-difficulty (ID). Kinematic signals were recorded alongside cortical activity to assess the relationship among movement time, task difficulty, and cortical activation.ResultsOur results showed that movement time increased with ID by 0.2974s/bit across all subjects (conditional r2 = 0.6434, p < 0.0001), and all subjects showed individual trends conforming Fitts' Law (all p < 0.001). Neither activation in BA4 nor in BA6 showed a significant correlation with ID (p > 0.05), while both the target size and distance, as well as the interaction between them, showed a significant relationship with BA4 or BA6 activation (all p < 0.05).ConclusionThis study found that although kinematic measures supported Fitts' Law, cortical activity in motor-related areas during reaching movements did not correlate directly with task difficulty as defined by Fitts' Law. Additional factors such as muscle activation may call for different cortical control even when difficulty was identical.

show abstract

“…It may be possible that the decrease in performance observed with smaller time windows as described previously is compensated by the increase in training examples. Also, even though previous work has shown potential at classifying fNIRS data using short time windows from the task trigger onset, hence focusing on the initial dip part of the hemodynamic response (Zafar and Hong, 2017 ; Khan and Hong, 2021 ), it may be possible that later segments are more useful to discriminate between conditions. Further studies would need to be conducted to compare different centering of shorter time windows as an extension of RQ3b and RQ3c .…”

Section: Discussionmentioning

confidence: 99%

Benchmarking framework for machine learning classification from fNIRS data

Benerradi

Clos²,

Landowska³

et al. 2023

Front. Neuroergonomics

View full text Add to dashboard Cite

BackgroundWhile efforts to establish best practices with functional near infrared spectroscopy (fNIRS) signal processing have been published, there are still no community standards for applying machine learning to fNIRS data. Moreover, the lack of open source benchmarks and standard expectations for reporting means that published works often claim high generalisation capabilities, but with poor practices or missing details in the paper. These issues make it hard to evaluate the performance of models when it comes to choosing them for brain-computer interfaces.MethodsWe present an open-source benchmarking framework, BenchNIRS, to establish a best practice machine learning methodology to evaluate models applied to fNIRS data, using five open access datasets for brain-computer interface (BCI) applications. The BenchNIRS framework, using a robust methodology with nested cross-validation, enables researchers to optimise models and evaluate them without bias. The framework also enables us to produce useful metrics and figures to detail the performance of new models for comparison. To demonstrate the utility of the framework, we present a benchmarking of six baseline models [linear discriminant analysis (LDA), support-vector machine (SVM), k-nearest neighbours (kNN), artificial neural network (ANN), convolutional neural network (CNN), and long short-term memory (LSTM)] on the five datasets and investigate the influence of different factors on the classification performance, including: number of training examples and size of the time window of each fNIRS sample used for classification. We also present results with a sliding window as opposed to simple classification of epochs, and with a personalised approach (within subject data classification) as opposed to a generalised approach (unseen subject data classification).Results and discussionResults show that the performance is typically lower than the scores often reported in literature, and without great differences between models, highlighting that predicting unseen data remains a difficult task. Our benchmarking framework provides future authors, who are achieving significant high classification scores, with a tool to demonstrate the advances in a comparable way. To complement our framework, we contribute a set of recommendations for methodology decisions and writing papers, when applying machine learning to fNIRS data.

show abstract

Most favorable stimulation duration in the sensorimotor cortex for fNIRS-based BCI

Cited by 14 publications

References 91 publications

Classification of Individual Finger Movements from Right Hand Using fNIRS Signals

Classification of Individual Finger Movements from Right Hand Using fNIRS Signals

Hemodynamic activity is not parsimoniously tuned to index-of-difficulty in movement with dual requirements on speed-accuracy

Benchmarking framework for machine learning classification from fNIRS data

Contact Info

Product

Resources

About