Resting-State fMRI is a functional MRI (fMRI) technique in which, unlike in task-based fMRI, the patient is not stimulated by any paradigm. The techniques makes it possible to study the functional connectivity of the brain.
Resting-state fMRI is based on spontaneous low frequency fluctuations (0.1 Hz) in the BOLD signal (Biswal et al. 1995). Studying correlations between variations of the BOLD signal, it can identify regions which activate synchronously with each other (Lee et al. 2012).
Like for task-based fMRI, a T2*-weighted gradient-echo EPI sequence is used to acquire the images (pixel size: approx. 3*3*3 mm³) every few seconds (e.g. TR = 3s) during approximately 6 minutes (Van Dijk et al. 2010). Eyes are open, with a fixation target. ECG, breathing and refluxed CO2 are sometimes recorded in order to eliminate physiological noise.
The pre-processing of resting-state fMRI images is similar to the one of standard fMRI images: slice-timing correction, motion correction, spatial and temporal filtering, and normalization. In the seed-based analysis, ROIs are manually drawn and their temporal signals are correlated with each others and with other voxels in the brain. Another approach, which requires less a priori assumptions, uses independent component analysis, in which the user must select the relevant components. Alternative analysis methods include graphs theory, clustering algorithms, and multivariate pattern classification.
The functional network which has been the most studied by resting-state fMRI is the default mode network. This network has been shown to be spontaneously activated without stimulation, and linked with mind wandering (Christoff et al. 2009). On the contrary, some other systems are task-based, such as the somatosensory, visual, auditory, language, attention and cognitive control networks.
For patients with neurological disabilities, resting-state fMRI is less demanding than task-based fMRI, which requires cooperation. However, clinical applications of resting-state fMRI are still at an early stage (Lee et al. 2013). Studies have used the technique to identify functional networks for neurosurgical planning (Hart et al. 2016). It has also been used to identify epileptogenic networks in epileptic patients. Studies have shown that resting-state networks characteristics could identify patients with Alzheimer disease and distinguish patients with mild cognitive impairment from controls. Patients with disorders of consciousness and psychiatric patients may also benefit from the technique (Vanhaudenhuyse et al. 2010).
Resting-state fMRI is looking for correlations of these fluctuations in different brain regions, making it possible to identify functional networks, such as the default mode network, the auditory network, etc. Combo-fMRI substracts these fluctuations from the task-based fMRI signal, in order to purify it, leading to less noisy activation maps.
By Dr. Laurent Hermoye
Benefit from the most advanced brain imaging techniques.