Yunjie Tong

Physiological denoising of BOLD fMRI data using Regressor Interpolation at Progressive Time Delays (RIPTiDe) processing of concurrent fMRI and near-infrared spectroscopy (NIRS)

Confounding noise in BOLD fMRI data arises primarily from fluctuations in blood flow and oxygenation due to cardiac and respiratory effects, spontaneous low frequency oscillations (LFO) in arterial pressure, and non-task related neural activity. Cardiac noise is particularly problematic, as the low sampling frequency of BOLD fMRI ensures that these effects are aliased in recorded data. Various methods have been proposed to estimate the noise signal through measurement and transformation of the cardiac and respiratory waveforms (e.g. RETROICOR and respiration volume per time (RVT)) and model-free estimation of noise variance through examination of spatial and temporal patterns. We have previously demonstrated that by applying a voxel-specific time delay to concurrently acquired near infrared spectroscopy (NIRS) data, we can generate regressors that reflect systemic blood flow and oxygenation fluctuations effects. Here, we apply this method to the task of removing physiological noise from BOLD data. We compare the efficacy of noise removal using various sets of noise regressors generated from NIRS data, and also compare the noise removal to RETROICOR+RVT. We compare the results of resting state analyses using the original and noise filtered data, and we evaluate the bias for the different noise filtration methods by computing null distributions from the resting data and comparing them with the expected theoretical distributions. Using the best set of processing choices, six NIRS-generated regressors with voxel-specific time delays explain a median of 10.5% of the variance throughout the brain, with the highest reductions being seen in gray matter. By comparison, the nine RETROICOR+RVT regressors together explain a median of 6.8% of the variance in the BOLD data. Detection of resting state networks was enhanced with NIRS denoising, and there were no appreciable differences in the bias of the different techniques. Physiological noise regressors generated using Regressor Interpolation at Progressive Time Delays (RIPTiDe) offer an effective method for efficiently removing hemodynamic noise from BOLD data.

Spatially weighted BOLD signal for comparison of functional magnetic resonance imaging and near-infrared imaging of the brain

We introduce a weighted spatial average of the functional magnetic resonance imaging (fMRI) BOLD signal (blood oxygen level-dependent) that is appropriate for comparison with the changes in oxy- and deoxy-hemoglobin concentrations measured with near-infrared spectroscopy (NIRS) during brain activation. Because the BOLD signal shows a spatial dependence (both in shape and amplitude) within the region of activation, the location of the optical probe with respect to the region of BOLD activation should be taken into account for comparison of the BOLD and NIRS signals. Our new method is based on combining weighted contributions of the BOLD signal from each activated voxel, with a weight given by a hitting density function for photons migrating between a given pair of illumination and collection points. We present a case study where we have found that the new spatially weighted BOLD signal shows a high spatial and temporal correlation with the oxy- and deoxy-hemoglobin concentration changes measured with NIRS during a hand-tapping protocol. These findings reinforce the idea that fMRI and NIRS are sensitive to similar underlying hemodynamic changes, and indicate that the proposed weighted BOLD signal is needed for a quantitative comparison of BOLD and NIRS signals.

Concurrent fNIRS and fMRI processing allows independent visualization of the propagation of pressure waves and bulk blood flow in the cerebral vasculature.

Blood Oxygen Level Dependent (BOLD) functional magnetic resonance imaging (fMRI) measures changes in blood oxygenation, which is affected by physiological processes, including cardiac pulsation, breathing, and low frequency oscillations (LFO). It is challenging to identify spatial and temporal effects of these processes on the BOLD signal because the low sampling rate of BOLD leads to aliasing of higher frequency physiological signal components. In this study, we used concurrent functional near infrared spectroscopy (fNIRS) and fMRI on 6 subjects during a resting state scan. To reduce aliasing, the BOLD fMRI acquisition was repeatedly performed on a set of sequentially acquired slice stacks to lower the TR to 0.5s while retaining high spatial resolution. Regressor interpolation at progressive time delays (RIPTiDe) method was used, in which physiological signal acquired by fNIRS (without aliasing) and its temporal shifts were used as regressors in the fMRI analysis to determine the magnitude and timing of the effects of various physiological processes on the BOLD signal. The details of the timing of the passage of the cardiac pulsation wave and of the cerebral blood itself were mapped. The result suggests that the cardiac signal affects the voxels near large blood vessels (arteries and veins) most strongly, while LFO mostly affected the drainage veins. We hypothesize that this could be the result of differences in the cerebral blood path lengths, and differences in the dynamics of the propagation of the signals. Together these results validate and extend a novel imaging technique to dynamically track the pulse-wave and bulk blood flow with concurrent fMRI and fNIRS.

Data analysis and statistical tests for near-infrared functional studies of the brain

We show some limitations of the standard t test when used together with typical data processing methods in functional Near Infrared Spectroscopy of the brain to assess the significance of multiple correlated points. We studied the occurrence of errors type I (that is the occurrence of false positive points) when typical processing methods are applied to time series of normal random numbers and to time series of simulated baseline systemic fluctuations. Since the results of the two studies are very similar we concluded that normal random numbers can be used to assess the occurrence of error type I due to certain algorithms of data processing. In order to decrease the occurrence of false positive points we propose to use some modified stepwise Bonferroni procedures, among which we studied the performance of Dubey/Armitage-Parmar algorithm. The results of the algorithm are shown for both simulated and experimental data.

Fast optical signals in the peripheral nervous system

We present a study of the near-infrared optical response to electrical stimulation of peripheral nerves. The sural nerve of six healthy subjects between the ages of 22 and 41 was stimulated with transcutaneous electrical pulses in a region located approximately 10 cm above the ankle. A two-wavelength (690 and 830 nm) tissue spectrometer was used to probe the same sural nerve below the ankle. We measured optical changes that peaked 60 to 160 ms after the electrical stimulus. On the basis of the strong wavelength dependence of these fast optical signals, we argue that their origin is mostly from absorption rather than scattering. From these absorption changes, we obtain oxy- and deoxy-hemoglobin concentration changes that describe a rapid hemodynamic response to electrical nerve activation. In five out of six subjects, this hemodynamic response is an increase in total (oxy+deoxy) hemoglobin concentration, consistent with a fast vasodilation. Our findings support the hypothesis that the peripheral nervous system undergoes neurovascular coupling, even though more data is needed to prove such hypothesis.

Spectral and spatial dependence of diffuse optical signals in response to peripheral nerve stimulation

Using non-invasive, near-infrared spectroscopy we have previously reported optical signals measured at or around peripheral nerves in response to their stimulation. Such optical signals featured amplitudes on the order of 0.1% and peaked about 100 ms after peripheral nerve stimulation in human subjects. Here, we report a study of the spatial and spectral dependence of the optical signals induced by stimulation of the human median and sural nerves, and observe that these optical signals are: (1) unlikely due to either dilation or constriction of blood vessels, (2) not associated with capillary bed hemoglobin, (3) likely due to blood vessel(s) displacement, and (4) unlikely due to fiber-skin optical coupling effects. We conclude that the most probable origin of the optical response to peripheral nerve stimulation is from displacement of blood vessels within the optically probed volume, as a result of muscle twitch in adjacent areas.

Studying brain function with near-infrared spectroscopy concurrently with electroencephalography

Near-infrared spectroscopy (NIRS) has been used for functional brain imaging by employing properly designed source-detector matrices. We demonstrate that by embedding a NIRS source-detector matrix within an electroencephalography (EEG) standard multi-channel cap, we can perform functional brain mapping of hemodynamic response and neuronal response simultaneously. In this study, the P300 endogenous evoked response was generated in human subjects using an auditory odd-ball paradigm while concurrently monitoring the hemodynamic response both spatially and temporally with NIRS. The electrical measurements showed the localization of evoked potential P300, which appeared around 320 ms after the odd-ball stimulus. The NIRS measurements demonstrate a hemodynamic change in the fronto-temporal cortex a few seconds after the appearance of P300.

Studying brain function with concurrent near-infrared spectroscopy (NIRS) and functional magnetic resonance imaging (fMRI)

We present concurrent NIRS-fMRI measurements on a human subject during a finger tapping test. The optical data were collected with a frequency domain experimental apparatus (ISS, Inc., Champaign IL) comprising sixteen laser sources at 690 nm, sixteen laser sources at 830 nm and four photomultiplier tube detectors. The lasers were coupled to optical fibers that led the light onto the subjects head. A special optical helmet (fMRI-compatible) with a retractable and resilient set of optical fibers was devised to improve the coupling between the fibers and the scalp. The fMRI data were collected with a 3 Tesla Siemens Trio magnetic resonance scanner and a quadrature birdcage radiofrequency coil. The spatial and temporal comparison of the fMRI and NIRS signals associated with brain activation showed a very good agreement, confirming the role of NIRS as a reliable brain monitor for functional studies.

Functional mapping of the human brain with near-infrared spectroscopy in the frequency-domain

Comparison of the spatial and temporal information retrieved from near-infrared phase and average intensity (DC) data reveals that these data types can play a complementary role in the study of the temporal and spatial features of the optical response associated with brain activation during a finger-tapping test. The optical data have been collected with a frequency-domain tissue imager at two wavelengths (690 and 830 nm) and have been analyzed using standard filtering and folding-average procedures. DC and phase data show different temporal and spatial features. A plausible explanation of the different behavior of DC and phase data has been attempted by using Monte Carlo simulations.

Fast optical response to electrical activation in peripheral nerves

Complex neuronal structures and interactions make studying fast optical signals associated with brain activation difficult, especially in non-invasive measurements that are further complicated by the filtering effect of the scalp and skull. We have chosen to study fast optical signals in the peripheral nervous system to look at a more simplified biological neuronal structure and a system that is more accessible to non-invasive optical studies. In this study, we recorded spatially resolved electrical and optical responses of the human sural nerve to electrical stimulation. A 0.1 ms electrical stimulation was used to activate the sural nerve. Electrical signals were collected by an electromyogram machine and results showed an electrical response spanning a distance of 8 mm across the nerve. Optical signals were collected by a two-wavelength (690 and 830 nm) near-infrared spectrometer and displayed a characteristic decrease in intensity at both wavelengths. Data were taken at multiple positions and then reproduced five times. The average optical data over the five trials showed an optical signal that was spatially consistent with the electrical response to sural nerve stimulation.