Spatial variation of changes in test-retest reliability of functional connectivity after global signal regression: The effect of considering hemodynamic delay.

Global signal regression (GSR) is a controversial analysis method, since its removal of signal has been observed to reduce the reliability of functional connectivity estimates. Here, we used test-retest reliability to characterize potential differences in spatial patterns between conventional, static GSR (sGSR) and a novel dynamic form of GSR (dGSR). In contrast with sGSR, dGSR models the global signal at a time delay to correct for blood arrival time. Thus, dGSR accounts for greater variation in global signal, removes blood-flow-related nuisance signal, and leaves higher quality neuronal signal remaining. We used intraclass correlation coefficients (ICCs) to estimate the reliability of functional connectivity in 462 healthy controls from the Human Connectome Project. We tested across two factors: denoising method used (control, sGSR, and dGSR), and interacquisition interval (between days, or within session while varying phase encoding direction). Reliability was estimated regionally to identify topographic patterns for each condition. sGSR and dGSR provided global reductions in reliability compared with the non-GSR control. Test-retest reliability was highest in the frontoparietal and default mode regions, and lowest in sensorimotor cortex for all conditions. dGSR provides more effective denoising in regions where both strategies greatly reduce reliability. Both GSR methods substantially reduced test-retest reliability, which was most evident in brain regions that had low reliability prior to denoising. These findings suggest that reliability of interregional correlation is likely inflated by the global signal, which is thought to primarily reflect dynamic blood flow.

An efficient functional magnetic resonance imaging data reduction strategy using neighborhood preserving embedding algorithm.

High dimensionality data have become common in neuroimaging fields, especially group-level functional magnetic resonance imaging (fMRI) datasets. fMRI connectivity analysis is a widely used, powerful technique for studying functional brain networks to probe underlying mechanisms of brain function and neuropsychological disorders. However, data-driven technique like independent components analysis (ICA), can yield unstable and inconsistent results, confounding the true effects of interest and hindering the understanding of brain functionality and connectivity. A key contributing factor to this instability is the information loss that occurs during fMRI data reduction. Data reduction of high dimensionality fMRI data in the temporal domain to identify the important information within group datasets is necessary for such analyses and is crucial to ensure the accuracy and stability of the outputs. In this study, we describe an fMRI data reduction strategy based on an adapted neighborhood preserving embedding (NPE) algorithm. Both simulated and real data results indicate that, compared with the widely used data reduction method, principal component analysis, the NPE-based data reduction method (a) shows superior performance on efficient data reduction, while enhancing group-level information, (b) develops a unique stratagem for selecting components based on an adjacency graph of eigenvectors, (c) generates more reliable and reproducible brain networks under different model orders when the outputs of NPE are used for ICA, (d) is more sensitive to revealing task-evoked activation for task fMRI, and (e) is extremely attractive and powerful for the increasingly popular fast fMRI and very large datasets.

Post-hoc physiological waveform extraction from motion estimation in simultaneous multislice (SMS) functional MRI using separate stack processing.

PURPOSE: Motion estimation is an essential step in functional MRI (fMRI) preprocessing. Usually, fMRI processing software packages (eg, FSL and AFNI) automatically estimate motion parameters in order to counteract the effects of motion. However, the time courses of the motion estimation for fMRI data also contain information about physiological processes. Here, we show that respiration and cardiac signals can be extracted from motion estimation at significantly higher bandwidth than is possible with current methods. METHOD: To detect motion at high effective temporal resolution (HighRes), the motion parameters of stacks of simultaneously acquired slices were estimated separately, then combined. This method was validated by extracting physiological motion signals from resting state fMRI (rsfMRI) data (Enhanced Nathan Kline Institute-Rockland Sample) and comparing them to respiration belt and pulse oximeter signals. RESULTS: HighRes motion time-courses with an effective sampling rate of 15.5 and 11.4 Hz were extracted from repetition time (TR) = 0.645 and 1.4 s data, respectively. Respiration waveforms were extracted with significantly higher accuracy than the original motion parameters. Even cardiac waveforms could be extracted, despite the fact that the sampling time or TR values were too long to sample cardiac frequencies. CONCLUSION: HighRes motion traces provide insight into the subjects' motion at higher frequencies than can be estimated using standard techniques. In its simplest form, this technique can recover accurate respiration signals and may reveal additional complexity in brain motion.

Increased cerebral blood flow in the right anterior cingulate cortex and fronto-orbital cortex during go/no-go task in children with ADHD.

OBJECTIVE: Arterial spin labeling (ASL) is a relatively new imaging modality in the field of the cognitive neuroscience. In the present study, we aimed to compare the dynamic regional cerebral blood flow alterations of children with ADHD and healthy controls during a neurocognitive task by using event-related ASL scanning. METHODS: The study comprised of 17 healthy controls and 20 children with ADHD. The study subjects were scanned on 3 Tesla MRI scanner to obtain ASL imaging data. Subjects performed go/no-go task during the ASL image acquisition. The image analyses were performed by FEAT (fMRI Expert Analysis Tool) Version 6. RESULTS: The mean age was 10.88 ± 1.45 and 11 ± 1.91 for the control and ADHD group, respectively (p = .112). The go/no-go task was utilized during the ASL scanning. The right anterior cingulate cortex (BA32) extending into the frontopolar and orbitofrontal cortices (BA10 and 11) displayed greater activation in ADHD children relative to the control counterparts (p < .001). With a lenient significance threshold, greater activation was revealed in the right-sided frontoparietal regions during the go session, and in the left precuneus during the no-go session. CONCLUSION: These results indicate that children with ADHD needed to over-activate frontopolar cortex, anterior cingulate as well as the dorsal and ventral attention networks to compensate for the attention demanded in a given cognitive task.

Dynamic functioning of transient resting-state coactivation networks in the Human Connectome Project.

Resting-state analyses evaluating large-scale brain networks have largely focused on static correlations in brain activity over extended time periods, however emerging approaches capture time-varying or dynamic patterns of transient functional networks. In light of these new approaches, there is a need to classify common transient network states (TNS) in terms of their spatial and dynamic properties. To fill this gap, two independent resting state scans collected in 462 healthy adults from the Human Connectome Project were evaluated using coactivation pattern analysis to identify (eight) TNS that recurred across participants and over time. These TNS spatially overlapped with prototypical resting state networks, but also diverged in notable ways. In particular, analyses revealed three TNS that shared cortical midline overlap with the default mode network (DMN), but these "complex" DMN states also encompassed distinct regions that fall beyond the prototypical DMN, suggesting that the DMN defined using static methods may represent the average of distinct complex-DMN states. Of note, dwell time was higher in "complex" DMN states, challenging the idea that the prototypical DMN, as a single unit, is the dominant resting-state network as typically defined by static resting state methods. In comparing the two resting state scans, we also found high reliability in the spatial organization and dynamic activities of network states involving DMN or sensorimotor regions. Future work will determine whether these TNS defined by coactivation patterns are in other samples, and are linked to fundamental cognitive properties.

Extraction of the cardiac waveform from simultaneous multislice fMRI data using slice sorted averaging and a deep learning reconstruction filter.

Cardiac signal contamination has long confounded the analysis of blood-oxygenation-level-dependent (BOLD) functional magnetic resonance imaging (fMRI). Cardiac pulsation results in significant BOLD signal changes, especially in and around blood vessels. Until the advent of simultaneous multislice echo-planar imaging (EPI) acquisition, the time resolution of whole brain EPI was insufficient to avoid cardiac aliasing (and acquisitions with repetition times (TRs) under 400-500 ms are still uncommon). As a result, direct detection and removal of the cardiac signal with spectral filters is generally not possible. Modelling methods have been developed to mitigate cardiac contamination, and recently developed techniques permit the visualization of cardiac signal propagation through the brain in undersampled data (e.g., TRs > 1s), which is useful in its own right for finding blood vessels. However, both of these techniques require data from which to estimate cardiac phase, which is generally not available for the data in many large databases of existing imaging data, and even now is not routinely recorded in many fMRI experiments. Here we present a method to estimate the cardiac waveform directly from a multislice fMRI dataset, without additional physiological measurements, such as plethysmograms. The pervasive spatial extent and temporal structure of the cardiac contamination signal across the brain offers an opportunity to exploit the nature of multislice imaging to extract this signal from the fMRI data itself. While any particular slice is recorded at the TR of the imaging experiment, slices are recorded much more quickly - typically from 10 to 20 Hz - sufficiently fast to fully sample the cardiac signal. Using the fairly permissive assumptions that the cardiac signal is a) pseudoperiodic b) somewhat coherent within any given slice, and c) is similarly shaped throughout the brain, we can extract a good estimate of the cardiac phase as a function of time from fMRI data alone. If we make further assumptions about the shape and consistency of cardiac waveforms, we can develop a deep learning filter to greatly improve our estimate of the cardiac waveform.

Vascular effects of caffeine found in BOLD fMRI.

The blood oxygen level-dependent (BOLD) signal in functional magnetic resonance imaging (fMRI) measures neuronal activation indirectly. Previous studies have found aperiodic, systemic low-frequency oscillations (sLFOs, ~0.1 Hz) in BOLD signals from resting state (RS) fMRI, which reflects the non-neuronal cerebral perfusion information. In this study, we investigated the possibility of extracting vascular information from the sLFOs in RS BOLD fMRI, which could provide complementary information to the neuronal activations. Two features of BOLD signals were exploited. First, time delays between the sLFOs of big blood vessels and brain voxels were calculated to determine cerebral circulation times and blood arrival times. Second, voxel-wise standard deviations (SD) of LFOs were calculated to represent the blood densities. We explored those features on the publicly available Myconnectome data set (a 2-year study of an individual subject (Male)), which contains 45 RS scans acquired after the subject had coffee, and 45 coffee-free RS scans, acquired on different days. Our results showed that shorter time delays and smaller SDs were detected in caffeinated scans. This is consistent with the vasoconstriction effects of caffeine, which leads to increased blood flow velocity. We also compared our results with previous findings on neuronal networks from the same data set. Our finding showed that brain regions with the significant vascular effect of caffeine coincide with those with a significant neuronal effect, indicating close interaction. This study provides methods to assess the physiological information from RS fMRI. Together with the neuronal information, we can study simultaneously the underlying correlations and interactions between vascular and neuronal networks, especially in pharmacological studies.

Impact of vessel wall lesions and vascular stenoses on cerebrovascular reactivity in patients with intracranial stenotic disease.

PURPOSE: To compare cerebrovascular reactivity (CVR) and CVR lagtimes in flow territories perfused by vessels with vs. without proximal arterial wall disease and/or stenosis, separately in patients with atherosclerotic and nonatherosclerotic (moyamoya) intracranial stenosis. MATERIALS AND METHODS: Atherosclerotic and moyamoya patients with >50% intracranial stenosis and <70% cervical stenosis underwent angiography, vessel wall imaging (VWI), and CVR-weighted imaging (n = 36; vessel segments evaluated = 396). Angiography and VWI were evaluated for stenosis locations and vessel wall lesions. Maximum CVR and CVR lagtime were contrasted between vascular territories with and without proximal intracranial vessel wall lesions and stenosis, and a Wilcoxon rank-sum was test used to determine differences (criteria: corrected two-sided P < 0.05). RESULTS: CVR lagtime was prolonged in territories with vs. without a proximal vessel wall lesion or stenosis for both patient groups: moyamoya (CVR lagtime = 45.5 sec ± 14.2 sec vs. 35.7 sec ± 9.7 sec, P < 0.001) and atherosclerosis (CVR lagtime = 38.2 sec ± 9.1 sec vs. 35.0 sec ± 7.2 sec, P = 0.001). For reactivity, a significant decrease in maximum CVR in the moyamoya group only (maximum CVR = 9.8 ± 2.2 vs. 12.0 ± 2.4, P < 0.001) was observed. CONCLUSION: Arterial vessel wall lesions detected on noninvasive, noncontrast intracranial VWI in patients with intracranial stenosis correlate on average with tissue-level impairment on CVR-weighted imaging. LEVEL OF EVIDENCE: 4 Technical Efficacy: Stage 3 J. Magn. Reson. Imaging 2017;46:1167-1176.

Comparison of peripheral near-infrared spectroscopy low-frequency oscillations to other denoising methods in resting state functional MRI with ultrahigh temporal resolution.

PURPOSE: Functional MRI (fMRI) blood-oxygen level-dependent (BOLD) signals result not only from neuronal activation, but also from nonneuronal physiological processes. These changes, especially in the low-frequency domain (0.01-0.2 Hz), can significantly confound inferences about neuronal processes. It is crucial to effectively identify these nuisance low-frequency oscillations (LFOs). METHOD: A high temporal resolution (repetition time, ∼0.5 s) fMRI resting state study was conducted with simultaneous physiological measurements to compare LFOs measured directly by near-infrared spectroscopy (NIRS) in the periphery and three methods that model LFOs from the respiration or cardiac signal: 1) the respiration volume per time (RVT), 2) the respiratory variation (RVRRF), and 3) the cardiac variation method (HRCRF). The LFO noise regressors from these methods were compared temporally and spatially as well as in their denoising efficiency. RESULTS: Methods were not highly correlated with one another, temporally or spatially. The set of two NIRS LFOs combined explained over 13% of BOLD signal variance and explained equal or more variance than HRCRF and RVRRF or RVT combined (in 14 of 16 participants). CONCLUSION: LFOs collected using NIRS in the periphery contain distinct temporal and spatial information about the LFOs in BOLD fMRI that is not contained in current low-frequency denoising methods derived from respiration and cardiac pulsation. Magn Reson Med 76:1697-1707, 2016. © 2016 International Society for Magnetic Resonance in Medicine.

Default Mode Network Functional Reorganization During Early Abstinence in Polysubstance-Using Emerging Adults Treated for Opioid Dependence.

This study examined default mode network connectivity within the first 30 days of abstinence in emerging adults entering treatment for opioid dependence. There were significant associations between abstinence duration and coupling strength with brain regions within and outside of the network.

Memory retrieval of smoking-related images induce greater insula activation as revealed by an fMRI-based delayed matching to sample task.

Nicotine dependence is a chronic and difficult to treat disorder. While environmental stimuli associated with smoking precipitate craving and relapse, it is unknown whether smoking cues are cognitively processed differently than neutral stimuli. To evaluate working memory differences between smoking-related and neutral stimuli, we conducted a delay-match-to-sample (DMS) task concurrently with functional magnetic resonance imaging (fMRI) in nicotine-dependent participants. The DMS task evaluates brain activation during the encoding, maintenance and retrieval phases of working memory. Smoking images induced significantly more subjective craving, and greater midline cortical activation during encoding in comparison to neutral stimuli that were similar in content yet lacked a smoking component. The insula, which is involved in maintaining nicotine dependence, was active during the successful retrieval of previously viewed smoking versus neutral images. In contrast, neutral images required more prefrontal cortex-mediated active maintenance during the maintenance period. These findings indicate that distinct brain regions are involved in the different phases of working memory for smoking-related versus neutral images. Importantly, the results implicate the insula in the retrieval of smoking-related stimuli, which is relevant given the insula's emerging role in addiction.

Tracking cerebral blood flow in BOLD fMRI using recursively generated regressors.

BOLD functional MRI (fMRI) data are dominated by low frequency signals, many of them of unclear origin. We have recently shown that some portions of the low frequency oscillations found in BOLD fMRI are systemic signals closely related to the blood circulation (Tong et al. [2013]: NeuroImage 76:202-215). They are commonly treated as physiological noise in fMRI studies. In this study, we propose and test a novel data-driven analytical method that uses these systemic low frequency oscillations in the BOLD signal as a tracer to follow cerebral blood flow dynamically. Our findings demonstrate that: (1) systemic oscillations pervade the BOLD signal; (2) the temporal traces evolve as the blood propagates though the brain; and, (3) they can be effectively extracted via a recursive procedure and used to derive the cerebral circulation map. Moreover, this method is independent from functional analyses, and thus allows simultaneous and independent assessment of information about cerebral blood flow to be conducted in parallel with the functional studies. In this study, the method was applied to data from the resting state scans, acquired using a multiband EPI sequence (fMRI scan with much shorter TRs), of seven healthy participants. Dynamic maps with consistent features resembling cerebral blood circulation were derived, confirming the robustness and repeatability of the method.

Short repetition time multiband echo-planar imaging with simultaneous pulse recording allows dynamic imaging of the cardiac pulsation signal.

PURPOSE: Recently developed simultaneous multislice echo-planar imaging (EPI) sequences permit imaging of the whole brain at short repetition time (TR), allowing the cardiac fluctuations to be fully sampled in blood-oxygen-level dependent functional MRI (BOLD fMRI). A novel low computational analytical method was developed to dynamically map the passage of the pulsation signal through the brain and visualize the whole cerebral vasculature affected by the pulse signal. This algorithm is based on a simple combination of fast BOLD fMRI and the scanner's own built-in pulse oximeter. METHODS: Multiple, temporally shifted copies of the pulse oximeter data (with 0.08 s shifting step and coverage of a 1-s span) were downsampled and used as cardiac pulsation regressors in a general linear model based analyses (FSL) of the fMRI data. The resulting concatenated z-statistics maps show the voxels that are affected as the cardiac signal travels through the brain. RESULTS: Many voxels were highly correlated with the pulsation regressor or its temporally shifted version. The dynamic and static cardiac pulsation maps obtained from both the task and resting state scans, resembled cerebral vasculature. CONCLUSION: The results demonstrated: (i) cardiac pulsation significantly affects most voxels in the brain; (ii) combining fast fMRI and this analytical method can reveal additional clinical information to functional studies.

Data-driven connectivity profiles relate to smoking cessation outcomes.

At a group level, nicotine dependence is linked to differences in resting-state functional connectivity (rs-FC) within and between three large-scale brain networks: the salience network (SN), default mode network (DMN), and frontoparietal network (FPN). Yet, individuals may display distinct patterns of rs-FC that impact treatment outcomes. This study used a data-driven approach, Group Iterative Multiple Model Estimation (GIMME), to characterize shared and person-specific rs-FC features linked with clinically-relevant treatment outcomes. 49 nicotine-dependent adults completed a resting-state fMRI scan prior to a two-week smoking cessation attempt. We used GIMME to identify group, subgroup, and individual-level networks of SN, DMN, and FPN connectivity. Regression models assessed whether within- and between-network connectivity of individual rs-FC models was associated with baseline cue-induced craving, and craving and use of regular cigarettes (i.e., "slips") during cessation. As a group, participants displayed shared patterns of connectivity within all three networks, and connectivity between the SN-FPN and DMN-SN. However, there was substantial heterogeneity across individuals. Individuals with greater within-network SN connectivity experienced more slips during treatment, while individuals with greater DMN-FPN connectivity experienced fewer slips. Individuals with more anticorrelated DMN-SN connectivity reported lower craving during treatment, while SN-FPN connectivity was linked to higher craving. In conclusion, in nicotine-dependent adults, GIMME identified substantial heterogeneity within and between the large-scale brain networks. Individuals with greater SN connectivity may be at increased risk for relapse during treatment, while a greater positive DMN-FPN and negative DMN-SN connectivity may be protective for individuals during smoking cessation treatment.

Brain-wide functional connectivity artifactually inflates throughout functional magnetic resonance imaging scans.

Functional magnetic resonance imaging (fMRI) is a central tool for investigating human brain function, organization and disease. Here, we show that fMRI-based estimates of functional brain connectivity artifactually inflate at spatially heterogeneous rates during resting-state and task-based scans. This produces false positive connection strength changes and spatial distortion of brain connectivity maps. We demonstrate that this artefact is driven by temporal inflation of the non-neuronal, systemic low-frequency oscillation (sLFO) blood flow signal during fMRI scanning and is not addressed by standard denoising procedures. We provide evidence that sLFO inflation reflects perturbations in cerebral blood flow by respiration and heart rate changes that accompany diminishing arousal during scanning, although the mechanisms of this pathway are uncertain. Finally, we show that adding a specialized sLFO denoising procedure to fMRI processing pipelines mitigates the artifactual inflation of functional connectivity, enhancing the validity and within-scan reliability of fMRI findings.

Prefrontal-Limbic Circuitry Is Associated With Reward Sensitivity in Nonhuman Primates.

BACKGROUND: Abnormal reward sensitivity is a risk factor for psychiatric disorders, including eating disorders such as overeating and binge-eating disorder, but the brain structural mechanisms that underlie it are not completely understood. Here, we sought to investigate the relationship between multimodal whole-brain structural features and reward sensitivity in nonhuman primates. METHODS: Reward sensitivity was evaluated through behavioral economic analysis in which monkeys (adult rhesus macaques; 7 female, 5 male) responded for sweetened condensed milk (10%, 30%, 56%), Gatorade, or water using an operant procedure in which the response requirement increased incrementally across sessions (i.e., fixed ratio 1, 3, 10). Animals were divided into high (n = 6) or low (n = 6) reward sensitivity groups based on essential value for 30% milk. Multimodal magnetic resonance imaging was used to measure gray matter volume and white matter microstructure. Brain structural features were compared between groups, and their correlations with reward sensitivity for various stimuli was investigated. RESULTS: Animals in the high sensitivity group had greater dorsolateral prefrontal cortex, centromedial amygdaloid complex, and middle cingulate cortex volumes than animals in the low sensitivity group. Furthermore, compared with monkeys in the low sensitivity group, high sensitivity monkeys had lower fractional anisotropy in the left dorsal cingulate bundle connecting the centromedial amygdaloid complex and middle cingulate cortex to the dorsolateral prefrontal cortex, and in the left superior longitudinal fasciculus 1 connecting the middle cingulate cortex to the dorsolateral prefrontal cortex. CONCLUSIONS: These results suggest that neuroanatomical variation in prefrontal-limbic circuitry is associated with reward sensitivity. These brain structural features may serve as predictive biomarkers for vulnerability to food-based and other reward-related disorders.

Evaluating the effects of systemic low frequency oscillations measured in the periphery on the independent component analysis results of resting state networks.

Independent component analysis (ICA) is widely used in resting state functional connectivity studies. ICA is a data-driven method, which uses no a priori anatomical or functional assumptions. However, as a result, it still relies on the user to distinguish the independent components (ICs) corresponding to neuronal activation, peripherally originating signals (without directly attributable neuronal origin, such as respiration, cardiac pulsation and Mayer wave), and acquisition artifacts. In this concurrent near infrared spectroscopy (NIRS)/functional MRI (fMRI) resting state study, we developed a method to systematically and quantitatively identify the ICs that show strong contributions from signals originating in the periphery. We applied group ICA (MELODIC from FSL) to the resting state data of 10 healthy participants. The systemic low frequency oscillation (LFO) detected simultaneously at each participant's fingertip by NIRS was used as a regressor to correlate with every subject-specific IC time course. The ICs that had high correlation with the systemic LFO were those closely associated with previously described sensorimotor, visual, and auditory networks. The ICs associated with the default mode and frontoparietal networks were less affected by the peripheral signals. The consistency and reproducibility of the results were evaluated using bootstrapping. This result demonstrates that systemic, low frequency oscillations in hemodynamic properties overlay the time courses of many spatial patterns identified in ICA analyses, which complicates the detection and interpretation of connectivity in these regions of the brain.

Brain-wide functional connectivity artifactually inflates throughout fMRI scans: a problem and solution.

The fMRI blood oxygen level-dependent (BOLD) signal is a mainstay of neuroimaging assessment of neuronal activity and functional connectivity in vivo. Thus, a chief priority is maximizing this signal's reliability and validity. To this end, the fMRI community has invested considerable effort into optimizing both experimental designs and physiological denoising procedures to improve the accuracy, across-scan reproducibility, and subject discriminability of BOLD-derived metrics like functional connectivity. Despite these advances, we discover that a substantial and ubiquitous defect remains in fMRI datasets: functional connectivity throughout the brain artifactually inflates during the course of fMRI scans - by an average of more than 70% in 15 minutes of scan time - at spatially heterogeneous rates, producing both spatial and temporal distortion of brain connectivity maps. We provide evidence that this inflation is driven by a previously unrecognized time-dependent increase of non-neuronal, systemic low-frequency oscillation (sLFO) blood flow signal during fMRI scanning. This signal is not removed by standard denoising procedures such as independent component analysis (ICA). However, we demonstrate that a specialized sLFO denoising procedure - Regressor Interpolation at Progressive Time Delays (RIPTiDe) - can be added to standard denoising pipelines to significantly attenuate functional connectivity inflation. We confirm the presence of sLFO-driven functional connectivity inflation in multiple independent fMRI datasets - including the Human Connectome Project - as well as across resting-state, task, and sleep-state conditions, and demonstrate its potential to produce false positive findings. Collectively, we present evidence for a previously unknown physiological phenomenon that spatiotemporally distorts estimates of brain connectivity in human fMRI datasets, and present a solution for mitigating this artifact.

Using carpet plots to analyze blood transit times in the brain during hypercapnic challenge magnetic resonance imaging.

Blood arrival time and blood transit time are useful metrics in characterizing hemodynamic behaviors in the brain. Functional magnetic resonance imaging in combination with a hypercapnic challenge has been proposed as a non-invasive imaging tool to determine blood arrival time and replace dynamic susceptibility contrast (DSC) magnetic resonance imaging, a current gold-standard imaging tool with the downsides of invasiveness and limited repeatability. Using a hypercapnic challenge, blood arrival times can be computed by cross-correlating the administered CO2 signal with the fMRI signal, which increases during elevated CO2 due to vasodilation. However, whole-brain transit times derived from this method can be significantly longer than the known cerebral transit time for healthy subjects (nearing 20 s vs. the expected 5-6 s). To address this unrealistic measurement, we here propose a novel carpet plot-based method to compute improved blood transit times derived from hypercapnic blood oxygen level dependent fMRI, demonstrating that the method reduces estimated blood transit times to an average of 5.32 s. We also investigate the use of hypercapnic fMRI with cross-correlation to compute the venous blood arrival times in healthy subjects and compare the computed delay maps with DSC-MRI time to peak maps using the structural similarity index measure (SSIM). The strongest delay differences between the two methods, indicated by low structural similarity index measure, were found in areas of deep white matter and the periventricular region. SSIM measures throughout the remainder of the brain reflected a similar arrival sequence derived from the two methods despite the exaggerated spread of voxel delays computed using CO2 fMRI.

Introduction to the shared near infrared spectroscopy format.

Significance: Functional near-infrared spectroscopy (fNIRS) is a popular neuroimaging technique with proliferating hardware platforms, analysis approaches, and software tools. There has not been a standardized file format for storing fNIRS data, which has hindered the sharing of data as well as the adoption and development of software tools. Aim: We endeavored to design a file format to facilitate the analysis and sharing of fNIRS data that is flexible enough to meet the community's needs and sufficiently defined to be implemented consistently across various hardware and software platforms. Approach: The shared NIRS format (SNIRF) specification was developed in consultation with the academic and commercial fNIRS community and the Society for functional Near Infrared Spectroscopy. Results: The SNIRF specification defines a format for fNIRS data acquired using continuous wave, frequency domain, time domain, and diffuse correlation spectroscopy devices. Conclusions: We present the SNIRF along with validation software and example datasets. Support for reading and writing SNIRF data has been implemented by major hardware and software platforms, and the format has found widespread use in the fNIRS community.