Hemodynamic response function (HRF) variability confounds resting‐state fMRI functional connectivity

Rangaprakash, D.; Wu, Guo‐Rong; Marinazzo, Daniele; Hu, Xiaoping; Deshpande, Gopikrishna

doi:10.1002/mrm.27146

Cited by 111 publications

(101 citation statements)

References 47 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Direct evidence supporting this argument has been offered by studies showing that the vascular response latency identified by a breath-hold task closely resembled the task-evoked hemodynamic delay patterns in primary sensory regions (Chang C et al 2008;Li Y et al 2018). Along similar lines, a recent study observed a marked reduction in the strength of resting-state connectivity if regional HRF shapes were taken into account through blind de-convolution (Rangaprakash D et al 2018), which indeed implies a possible link between regionally varying HRFs and apparent RSN structures reported in the literature, similar to how RRFs/iCRFs determine apparent "physiological" networks reported here.…”

Section: Shared Spatial Patterns Between "Physiological" Dynamics Andsupporting

confidence: 86%

Resting-state “Physiological Networks”

Chen

Lewis²,

Chang

et al. 2019

Preprint

View full text Add to dashboard Cite

Slow changes in systemic brain physiology can elicit large fluctuations in fMRI time series, which may manifest as structured spatial patterns of temporal correlations between distant brain regions. These correlations can appear similar to large-scale networks typically attributed to coupled neuronal activity. However, little effort has been devoted to a systematic investigation of such "physiological networks"-sets of segregated brain regions that exhibit similar physiological responses-and their potential influence on estimates of resting-state brain networks. Here, by analyzing a large group of subjects from the 3T Human Connectome Project database, we demonstrate brain-wide and noticeably heterogenous dynamics attributable to either respiratory variation or heart rate changes. We show that these physiologic dynamics can give rise to apparent "connectivity" patterns that resemble previously reported resting-state networks derived from fMRI data. Further, we show that this apparent "physiological connectivity" cannot be removed by the use of a single nuisance regressor for the entire brain (such as global signal regression) due to the clear regional heterogeneity of the physiological responses.Possible mechanisms causing these apparent "physiological networks", and their broad implications for interpreting functional connectivity studies are discussed.

show abstract

Section: Shared Spatial Patterns Between "Physiological" Dynamics Andsupporting

confidence: 86%

Resting-state “Physiological Networks”

Chen

Lewis²,

Chang

et al. 2019

Preprint

View full text Add to dashboard Cite

show abstract

“…In addition to connectome specificity, other structural data features may drive functional inter-subject variability including foremost regional variance such as synaptic receptor type and density (39), but also methodological variations such as parcellation differences (3). Notwithstanding, we cannot exclude that the variations in hemodynamic response functions (HRF) across animals and brain location affect SC-FC relations, as it has been shown to contribute to individual variability in human FC estimation (40). In this study we aimed at reducing this variability by scanning awake mice, reducing the confounding effects of anesthesia and allowing collection of data over multiple sessions per mouse (41,42).…”

Section: Discussionmentioning

confidence: 99%

“…In this study we aimed at reducing this variability by scanning awake mice, reducing the confounding effects of anesthesia and allowing collection of data over multiple sessions per mouse (41,42). Moreover, we used spin-echo echo planar imaging (EPI), which is more sensitive to microvasculature relative to gradient-echo EPI, especially at high magnetic fields (43), further reducing the variability of HRF (40).…”

Section: Discussionmentioning

confidence: 99%

Individual structural features constrain the mouse functional connectome

Melozzi

Bergmann

Harris

et al. 2019

Preprint

View full text Add to dashboard Cite

Whole brain dynamics intuitively depends upon the internal wiring of the brain; butto which extent the individual structural connectome constrains the corresponding functional connectome is unknown, even though its importance is uncontested. After acquiring structural data from individual mice, we virtualized their brain networks and simulated in silico functional MRI data. Theoretical results were validated against empirical awake functional MRI data obtained from the same mice. We demonstrate that individual structural connectomes predict the functional organization of individual brains. Using a virtual mouse brain derived from the Allen Mouse Brain Connectivity Atlas, we further show that the dominant predictors of individual structure-function relations are the asymmetry and the weights of the structural links. Model predictions were validated experimentally using tracer injections, identifying which missing connections (not measurable with diffusion MRI) are important for whole brain dynamics in the mouse. Individual variations thus define a specific structural fingerprint with direct impact upon the functional organization of individual brains, a key feature for personalized medicine. SIGNIFICANCE STATEMENTThe structural connectome is a key determinant of brain function and dysfunction. The connectome-based model approach aims to understand the functional organization of the brain by modeling the brain as a dynamical system and then studying how the functional architecture rises from the underlying structural skeleton. Here, taking advantage of mice studies, we systematically investigated the informative content of different structural features in explaining the emergence of the functional ones. We demonstrate that individual variations define a specific structural fingerprint with a direct impact upon the functional organization of individual brains stressing the importance of using individualized models to understand brain function. We show how limitations of connectome reconstruction with the diffusion-MRI method restrict our comprehension of the structural-functional relation. PP Tracerfiltered = 0.461 ± 0.005 PP Tracer = 0.488 ± 0.005 P = 0.002 PP Individual−det = 0.415 ± 0.005 P = 2e -10 PP Tracersym = 0.418 ± 0.004 PP Tracer = 0.488 ± 0.005 P = 2e -20 P = 1.0 PP Tracersym = 0.418 ± 0.004

show abstract

“…In particular, variable delays between neural activity and peak haemodynamic response around the brain means that temporal precedence in the BOLD response does not necessarily imply neuronal causality. While approaches have been suggested to address this issue [68,69], we instead investigated information-theoretic signatures on simulated neural data, which has a much higher effective sampling frequency than BOLD, and is also relatively unaffected by the temporal convolution that masks neural activity in the BOLD response. In doing so, we highlight important multi-level organisation within the simulated neural time series, in which whole-brain topological signatures (measured using BOLD) overlap with specific signatures of regional (neural) effective connectivity.…”

Section: Discussionmentioning

confidence: 99%

Transitions in brain-network level information processing dynamics are driven by alterations in neural gain

Han

Aburn

et al. 2019

Preprint

View full text Add to dashboard Cite

A key component of the flexibility and complexity of the brain is its ability to dynamically adapt its functional network structure between integrated and segregated brain states depending on the demands of different cognitive tasks. Integrated states are prevalent when performing tasks of high complexity, such as maintaining items in working memory, consistent with models of a global workspace architecture. Recent work has suggested that the balance between integration and segregation is under the control of ascending neuromodulatory systems, such as the noradrenergic system. In a previous large-scale nonlinear oscillator model of neuronal network dynamics, we showed that manipulating neural gain led to a 'critical' transition in phase synchrony that was associated with a shift from segregated to integrated topology, thus confirming our original prediction. In this study, we advance these results by demonstrating that the gain-mediated phase transition is characterized by a shift in the underlying dynamics of neural information processing. Specifically, the dynamics of the subcritical (segregated) regime are dominated by information storage, whereas the supercritical (integrated) regime is associated with increased information transfer (measured via transfer entropy). Operating near to the critical regime with respect to modulating neural gain would thus appear to provide computational advantages, offering flexibility in the information processing that can be performed with only subtle changes in gain control. Our results thus link studies of whole-brain network topology and the ascending arousal system with information processing dynamics, and suggest that the constraints imposed by the ascending arousal system constrain low-dimensional modes of information processing within the brain. Author summaryHigher brain function relies on a dynamic balance between functional integration and segregation. Previous work has shown that this balance is mediated in part by alterations in neural gain, which are thought to relate to projections from ascending neuromodulatory nuclei, such as the locus coeruleus. Here, we extend this work by demonstrating that the modulation of neural gain alters the information processing dynamics of the neural components of a biophysical neural model. Specifically, we find March 11, 2019 1/26 that low levels of neural gain are characterized by high Active Information Storage, whereas higher levels of neural gain are associated with an increase in inter-regional Transfer Entropy. Our results suggest that the modulation of neural gain via the ascending arousal system may fundamentally alter the information processing mode of the brain, which in turn has important implications for understanding the biophysical basis of cognition. IntroductionAlthough there is a long history relating individual brain regions to specific and specialized functions, regions in isolation cannot perform meaningful physiological or cognitive processes [1]. Instead, starting at a lower scale, a few prominent features of ...

show abstract

Hemodynamic response function (HRF) variability confounds resting‐state fMRI functional connectivity

Abstract: HRFv, if ignored, could cause identification of false FC. FC findings from HRFv-ignored data should be interpreted cautiously. We suggest deconvolution to minimize HRFv.

Cited by 111 publications

References 47 publications

Resting-state “Physiological Networks”

Resting-state “Physiological Networks”

Individual structural features constrain the mouse functional connectome

Transitions in brain-network level information processing dynamics are driven by alterations in neural gain

Contact Info

Product

Resources

About