A task-based figures out information about brain processing of short stimuli or tasks. The idea of task -based is to find out whether location in the brain are more or less active in different tasks and their correlation between activation and fluctuation. This technique depends on neural activation strength that is BOLD effect. Some of the problems with this technique would be improperly performing the task, an incorrect HDR for the part of the brain. If task is complex the outcome could be questionable. Multivariate Pattern analysis (MVPA) is a method relies on the idea of combining the data from multiple voxels to decode activation patterns. Activation in a region is weak or similar across conditions, but the pattern over voxels is informative. Limitation of MVPA is that it focuses on the patterns of activation and thus ignores interactions. A functional connectivity of a voxel may change even though it has stable activation across the conditions. Application of functional connectivity is examining intrinsic correlations while participant is rest. The resting state analysis is based on reading, ongoing, spontaneous functional signals across brain areas. Further, quantifying how similar those signal look over …show more content…
DMN comprise of a set of brain regions that are deactivated during a wide range of cognitive task. It is anchored in the posterior cingulate cortex and medial prefrontal cortex, with nodes in the medial temporal lobe and the angular gyrus. Research has suggested that the DMN is a robust network which can be readily identified in every individual. A failure to deactivate DMN activity during cognitively demanding tasks can lead to various diseases. An approach for processing rs-fMRI includes is hypothesis -driven that includes task-based GLM, seed based connectivity analysis (SCA). Another techniques is Data-driven analyses that comprise of principal component analysis (PCA), independent component analysis (ICA) and graph theory. The seed looks for the same data points within the ROI over different slices. The outcome of a SCA is a z-scores map depicting how well time series matched with the time series of the seed. Principal Component Analysis (PCA) uses the first and second moments of the measured data and hence rely heavily on Gaussian features. PCA is used to reduce a large set of data to a reduced set of principal components. It is used when data is Gaussian, linear and stationary. Independent Component Analysis (ICA) exploits inherently non-Gaussian features of the data and employs higher moments. It is used when data cannot be ensemble or when raw data appears to be very noisy. Graph theory is another way