The basic idea. According to the point process theory discrete BOLD events (i.e. pseudo-events in the absence of an external stimulus) govern the brain dynamics at rest (e.g. Tagliazucchi et al. 2012). The rsHRF toolbox is aimed to retrieve the neuronal onsets of these pseudo-events with no explicit stimulus and timing together with the hemodynamic response (rsHRF) it set off (Wu et al., 2013; Wu & Marinazzo, 2015; Wu & Marinazzo, 2016). To this end, the rsHRF toolbox first identifies the pseudo-events, i.e. when the standardized resting-state BOLD signal crosses a given threshold (1 SD). Thereafter, a model is fitted to retrieve:
- the optimal lag between the pseudo-events and the neuronal (rsHRF) onset;
- the shape of the estimated rsHRF which will depend on the by-the-toolbox predefined HRF basis functions. Users of the rsHRF toolbox can choose one of eight options:
Code HRF basis functions rsHRF_estimation_temporal_basis.mcanontd: a canonical HRF with its time derivative canontdd: a canonical HRF with its time and dispersion derivatives Gamma Functions with a variable number of basis functions (k), e.g. 3-5 Fourier Set with a default number of basis functions (k) equal to 3 Fourier Set (Hanning) with a default number of basis functions (k) equal to 3 rsHRF_estimation_FIR.mFIR: Finite Impulse Response sFIR: smoothed Finite Impulse Response ( rsHRF_estimation_impulseest.m)non-parametric impulse response estimation: not included in the rsHRF GUI
Once that the rsHRF has been retrieved for each voxel/vertex in the brain, you can:
use the rsHRF as a pathophysiological indicator (by mapping the rsHRF shape onto the brain surface and looking at the inter-subject variability);
The shape of the rsHRF can be characterized by three parameters, namely response height (RH), time to peak (TTP), and Full Width at Half Maximum (FWHM). Each of these parameters can be mapped onto the brain surface (see Figure for an example: full brain map of the response height estimated using the Finite Impulse Response basis functions). Note that the full brain map covers the full brain surface, including white matter and CSF. To consult some example data, head over to NeuroVault. The number of pseudo-events per voxel/vertex can also be mapped onto the brain surface. After mapping all parameters (i.e. RH, TTP, FWHM, number of pseudo-events) onto the brain surface for each voxel/vertex and subject, the subject-specific brain maps can be used to examine whether/how the rsHRF is modulated by psycho-physiological factors (i.e. inter-subject hemodynamic variability; e.g. post-traumatic stress disorder, autism spectrum disorder, chronic pain, consciousness). With the 3dMVM function embedded in AFNI, one can even run a multivariate analysis in which the three rsHRF parameters are modeled as multiple, simultaneous response variables (Chen, Adleman, Saad, Leibenluft, & Cox, 2014).
deconvolve the rsHRF from the resting-state fMRI BOLD signal (for example to improve lag-based connectivity estimates).
- Chen, G., Adleman, N. E., Saad, Z. S., Leibenluft, E., & Cox, R. W. (2014). Applications of multivariate modeling to neuroimaging group analysis: A comprehensive alternative to univariate general linear model. NeuroImage, 99, 571-588. https://doi.org/10.1016/j.neuroimage.2014.06.027
- Rangaprakash, D., Dretsch, M. N., Yan, W., Katz, J. S., Denney Jr, T. S., & Deshpande, G. (2017). Hemodynamic variability in soldiers with trauma: Implications for functional MRI connectivity studies. NeuroImage: Clinical, 16, 409-417. https://doi.org/10.1016/j.nicl.2017.07.016
- Rangaprakash, D., Wu, G. R., Marinazzo, D., Hu, X., & Deshpande, G. (2018). Hemodynamic response function (HRF) variability confounds resting‐state fMRI functional connectivity. Magnetic Resonance in Medicine, 80(4), 1697-1713. https://doi.org/10.1002/mrm.27146
- Tagliazucchi, E., Balenzuela, P., Fraiman, D., & Chialvo, D. R. (2012). Criticality in large-scale brain fMRI dynamics unveiled by a novel point process analysis. Frontiers in Physiology, 3, 15. https://doi.org/10.3389/fphys.2012.00015
- 📎 Wu, G. R., Di Perri, C., Charland-Verville, V., Martial, C., Carrière, M., Vanhaudenhuyse, A., ... & Marinazzo, D. (2019). Modulation of the spontaneous hemodynamic response function across levels of consciousness. NeuroImage, 200, 450-459. https://doi.org/10.1016/j.neuroimage.2019.07.011
- 📎 Wu, G. R., Liao, W., Stramaglia, S., Ding, J. R., Chen, H., & Marinazzo, D. (2013). A blind deconvolution approach to recover effective connectivity brain networks from resting state fMRI data. Medical Image Analysis, 17(3), 365-374. https://doi.org/10.1016/j.media.2013.01.003
- Wu, G. R., & Marinazzo, D. (2015). Point-process deconvolution of fMRI BOLD signal reveals effective connectivity alterations in chronic pain patients. Brain Topography, 28(4), 541-547.
- 📎📎 Wu, G. R., & Marinazzo, D. (2015). Retrieving the Hemodynamic Response Function in resting state fMRI: Methodology and applications (No. e1621). PeerJ PrePrints.
- 📎 Wu, G. R., & Marinazzo, D. (2016). Sensitivity of the resting-state haemodynamic response function estimation to autonomic nervous system fluctuations. Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, 374(2067), 20150190. https://doi.org/10.1098/rsta.2015.0190
- Yan, W., Rangaprakash, D., & Deshpande, G. (2018). Aberrant hemodynamic responses in autism: Implications for resting state fMRI functional connectivity studies. NeuroImage: Clinical, 19, 320-330. https://doi.org/10.1016/j.nicl.2018.04.013