Back to the main page.
Bug 1042 - leadfields do not scale appropriately with the conductivity
| Status | CLOSED FIXED |
| Reported | 2011-10-14 14:30:00 +0200 |
| Modified | 2014-02-24 10:56:29 +0100 |
| Product: | FieldTrip |
| Component: | forward |
| Version: | unspecified |
| Hardware: | PC |
| Operating System: | Windows |
| Importance: | P1 normal |
| Assigned to: | Robert Oostenveld |
| URL: | |
| Tags: | |
| Depends on: | 1426 |
| Blocks: | 1414 |
| See also: |
Cristiano Micheli - 2011-10-14 14:30:23 +0200
Following up on bug 686 make a similar test file to check the dependenncy of lead fields on different conducitivity values and units
Cristiano Micheli - 2011-10-14 15:40:15 +0200
(In reply to comment #0) > Following up on bug 686 make a similar test file to check the dependenncy of > lead fields on different conducitivity values and units EEG leadfields scale with conductivity, but geometrical and conductivity units are independent at the moment. The units of conductivity should depend on the geometrical ones, but this is not implemented yet. EEG OM leadfields seem not to behave very well MEG leadfields do not depend at all from conductivity whereas the percentile of the magnetic field due to the return currents should account for these changes.
Robert Oostenveld - 2011-11-11 11:19:47 +0100
I closed all bugs that were recently resolved.
Robert Oostenveld - 2012-04-15 13:06:51 +0200
The test script has regressed and now does not work any more, therefore I am reopening this bug. Although the reason for the test script failing does not seem to be due to the conductivity scaling, it should be checked that the conductivity scaling is (still) correct. See also bug #1414.
Boris Reuderink - 2012-10-10 14:15:24 +0200
Moved bugs from Christiano Michelli to joint development user.
Robert Oostenveld - 2013-11-19 14:14:59 +0100
% dim1 = the "slices" correspond to different conductivities % dim2 = the rows correspond to the different volume conduction models % dim3 = the columns correspond to m, cm, mm -------------------------------------------------------- MEG results -------------------------------------------------------- ans(:,:,1) = 1.000000000000000 0.000100000000000 0.000001000000000 0.999999998126793 0.000099999999813 0.000000999999998 1.020168870872125 0.000102016887087 0.000001020168871 1.010098940468291 0.000101009894046 0.000001010098940 0.006483640594510 0.000109093182336 0.000051250774482 ans(:,:,2) = 0.333333333333333 0.000033333333333 0.000000333333333 0.333333332708931 0.000033333333271 0.000000333333333 0.340056290290709 0.000034005629029 0.000000340056290 0.336699659143620 0.000033669965914 0.000000336699659 0.002161213531503 0.000036364394112 0.000017083591494 ans(:,:,3) = 0.142857142857143 0.000014285714286 0.000000142857143 0.142857142589542 0.000014285714259 0.000000142857143 0.145738410124589 0.000014573841012 0.000000145738410 0.144299852741964 0.000014429985274 0.000000144299853 0.000926234370644 0.000015584740334 0.000007321539212 -------------------------------------------------------- EEG results -------------------------------------------------------- ans(:,:,1) = 1.000000000000000 0.000100000000000 0.000001000000000 1.000000000000009 0.000100000000000 0.000001000000000 0.993643521960103 0.000099364352196 0.000000993643522 ans(:,:,2) = 1.000000000000000 0.000100000000000 0.000001000000000 1.000000000000009 0.000100000000000 0.000001000000000 0.993643521960103 0.000099364352196 0.000000993643522 ans(:,:,3) = 1.000000000000000 0.000100000000000 0.000001000000000 1.000000000000009 0.000100000000000 0.000001000000000 0.993643521960103 0.000099364352196 0.000000993643522
Robert Oostenveld - 2013-11-19 14:24:25 +0100
(In reply to comment #5) hmm, it should just have been the other way around in the previous comment. The first table contains EEG results for 'eegvol_singlesphere' 'eegvol_concentricspheres' 'eegvol_bem_cp' 'eegvol_bem_dipoli' 'eegvol_bem_openmeeg' The EEG results are consistent with each other, except for openmeeg (the last one). That is also what I see in test_eeg_leadfield_units (after the fix that was needed, please use revision 8816 or later). The 2nd table contains MEG results for 'megvol_singlesphere' 'megvol_localspheres' 'megvol_singleshell' The MEG results are consistent with each other. The test_meg_leadfield_units script shows that the units are also correct for SI inputs, i.e. if the input is SI, the output leadfield is also in SI. In short: the test_bug1042 script runs through without errors (on revision 8816). Only the OpenMEEG EEG leadfield is incorrect (when inputs are in SI units). I will file the OpenMEEG case as a separate bug.
Robert Oostenveld - 2013-11-19 14:27:54 +0100
(In reply to comment #6) For reference, I have posted the OpenMEEG bug as bug #2380.