Back to the main page.
Bug 3350 - Error in sLoreta square root matrix
| Status | ASSIGNED |
| Reported | 2017-09-19 20:45:00 +0200 |
| Modified | 2017-11-09 17:34:06 +0100 |
| Product: | FieldTrip |
| Component: | inverse |
| Version: | unspecified |
| Hardware: | All |
| Operating System: | All |
| Importance: | P5 major |
| Assigned to: | Ricardo Bruña Fernández |
| URL: | |
| Tags: | |
| Depends on: | |
| Blocks: | |
| See also: |
Ricardo Bruña Fernández - 2017-09-19 20:45:19 +0200
In the sLoreta code (inverse/ft_sloreta.m), line 255 in the current version, the code reads: filt(ii,:) = pinv(sqrt(lf(:,ii)' * invG * lf(:,ii))) * lf(:,ii)' * invG; Pascual-Marqui in arXiv 0710.3341 equation 15 uses the inverse square root of the symmetric matrix. Right now the code is calculating the inverse of the matrix formed by the square root of each element. The code, thus, should read: filt(ii,:) = pinv(sqrtm(lf(:,ii)' * invG * lf(:,ii))) * lf(:,ii)' * invG; This completely perverts the calculated spatial filter.