Region of interest | MCE User’s Guide |
8 Region of interest
You can estimate the activity in certain brain area as a function of time by defining a region of interest (ROI). ROI is actually a weighting function that is used to cal- culate the sum amplitude from the estimate.
8.1 What is ROI?
The smoothed weighting is a function of position r, having form
f ( r) =
but you don't need to be interested in matrix calculation to understand how the ROIs work.
The ROI window (p. 34) visualizes the currently selected ROI with an ellipsoid. The weight in the center has weight 1 in the sum, and the points on border of the ellipsoid have the weight 0.60.
Activity at each source location is multiplied with the corresponding weight and added together. The weighted sum is shown in the Amplitude window (p. 31).
A ROI can also have hard edges. In this case, the weight is
f ( r) = 1, if
If no ROI is selected, normal sum of all the activity is used.
8.2 What is the orientation of a ROI?
Since MEG is sensitive to the orientation of the neural currents and this informa- tion is often very useful in differentiating nearby source areas, the ROIs can be used with a defined orientation. The weighting function depends in this case also on the orientation q0 and the direction of the estimated current q:
( , ) q ⋅ q0 g r q = f (r)-----------------
q q0
The amplitude is the projection of the estimated current on the selected orientation. If the estimated current is parallel with the ROI, the amplitude will be the same as for the
14 |
