Reduction of Onset Delay in Functional Near-Infrared Spectroscopy: Prediction of HbO/HbR Signals

Abstract

An intrinsic problem when using hemodynamic responses for the brain-machine interface is the slow nature of the physiological process. In this paper, a novel method that estimates the oxyhemoglobin changes caused by neuronal activations is proposed and validated. In monitoring the time responses of blood-oxygen-level-dependent signals with functional near-infrared spectroscopy (fNIRS), the early trajectories of both oxy-and deoxy-hemoglobins in their phase space are scrutinized. Furthermore, to reduce the detection time, a prediction method based upon a kernel-based recursive least squares (KRLS) algorithm is implemented. In validating the proposed approach, the fNIRS signals of finger tapping tasks measured from the left motor cortex are examined. The results show that the KRLS algorithm using the Gaussian kernel yields the best fitting for both HbO (i.e., 87.5%) and HbR (i.e., 85.2%) at q = 15 steps ahead (i.e., 1.63 s ahead at a sampling frequency of 9.19 Hz). This concludes that a neuronal activation can be concluded in about 0.1 s with fNIRS using prediction, which enables an almost real-time practice if combined with EEG.

Detection of Mild Cognitive Impairment Using Convolutional Neural Network: Temporal-Feature Maps of Functional Near-Infrared Spectroscopy

“…These features render fMRI inappropriate for certain research and many clinical applications (Santosa et al, 2014). In contrast, fNIRS is a novel neuroimaging modality with the following advantages: it is non-invasive, safe, less costly, portable, and tolerant of motion artifacts (Perrey, 2008); it also has great temporal resolution and moderate spatial resolution (Ghafoor et al, 2017;Zafar and Hong, 2020). In addition, fNIRS is in progress to improve the spatial and temporal resolutions with the development of bundled-optodes configurations , detection of the initial dip (Zafar and Hong, 2017;Hong and Zafar, 2018), and combination of adaptive method (Iqbal et al, 2018;Hong and Pham, 2019;Pamosoaji et al, 2019) to improve information transfer rate.…”

Classification of Individual Finger Movements from Right Hand Using fNIRS Signals

“…Trail-to-trail variability in fNIRS signal for finger-tapping tasks could be reduced using seed correlation methods that can enhance the classification accuracy [ 47 ]. We also envisage to using estimation algorithms such as the q-step-ahead prediction scheme and the kernel-based recursive least squares (KRLS) algorithm to reduce the onset delay of the changes due to finger-tapping for real-time implementation in the BCI system [ 21 , 48 , 49 , 50 ]. In the study, we considered only data.…”

“…In single-trail classification for a motor imaginary with thumb and complex finger-tapping task achieves an average accuracy of 81% by simply changing the combination of a set of channels, time intervals, and features [ 20 ]. In [ 21 ] thumb and little finger were classified with an accuracy of 87.5% for data. Deep learning approaches are also becoming popular for the classification of these complex finger movements.…”

Decoding of Walking Imagery and Idle State Using Sparse Representation Based on fNIRS

“…When brain activity in a certain area of the brain increases, the oxygen demand increases in that area, which will make more HbR combined with oxygen molecules become HbO to increase oxygen transport. Conversely, when brain activity in that area decreases, the amount of HbO decreases [ 33 , 34 ]. That is the HR.…”