integration of Baudrexel's imperfect spoiling correction, SDAM B1mapping, and visual quality tool #19
Conversation
Three types of visualisation are possible: - automated intensity scaling for all images (to check e.g. reorientation comparing input image and template) - identical intensity scaling for all images (to check e.g. series of echoes from a T1w acquisition) - standard scaling as defined for each output map (R1, R2*, MT, PD, B1, RFsens). Scaling is automated (between percentiles 5 and 95) if image type cannot be determined. Allows for quick visual check of any obvious anomaly in the output maps.
- range for RF sensitivity maps [0 2] - backward/forward compatibility between very last version of SPM and previous ones - to handle B1+ and RFsens maps imtype
Bit of cosmetic changes...
- Depending on the history of manipulating the st.centre field, the latter can either be a column or a line vector. Elements processed one by one to avoid problems. - strsplit and strjoin were added to Matlab from version R2013a. Alternative implementation for that specific case.
A start...
Two options can be selected: no correction (default) or select a json file containing the correction coefficients.
…s in config/local/ISC_parameters
…eters in batch system
… hmri-group-master
|
…nd fix few bugs in that context
|
I've changed the code to have the ISC options in local defaults only (not in the GUI) - could you check descriptions in hmri_defaults.m please? @mfcallaghan @siawoosh |
| % f_T correction is applied either if: | ||
| % - f_T has been provided as separate B1 mapping measurement (not | ||
| % UNICORT!) or | ||
| % - f_T has been calculated using UNICORT *AND* the UNICORT.MT flag | ||
| % is enabled (advanced user only! method not validated yet!) | ||
| if (~isempty(f_T))&&(~mpm_params.UNICORT.R1 || mpm_params.UNICORT.MT) | ||
| if (~isempty(f_T)&&~isfield(ISC,'iscphase'))&&(~mpm_params.UNICORT.R1 || mpm_params.UNICORT.MT) |
There was a problem hiding this comment.
Why is Gunther's MT correction not applied if imperfect spoiling correction is applied? They should be independent, I think.
|
@tleutritz-cbs I've had a brief look through the changes, and in my opinion it would be better as three separate pull requests. I think the SDAM and visual quality tool implementations are things that (I think) can be easily tested and pulled, but the imperfect spoiling correction (ISC) changes will be a lot harder to verify and test. |
…f multi-echo data per contrast
|
QC tool is now part of pull request #39 |
|
SDAM B1 mapping moved into pull request #41. |
The recent development contains
my integration of Baudrexel's imperfect spoiling correction, which was in use for the multi-center multi-vendor study [1], including the new file-reader for the new parameter calculator introduced by M. Callaghan (for ease of use not json files will be read in, but a separate m file; the wiki will/has to be changed accordingly after merge),
an implementation of SDAM B1-mapping (double angle method applied mainly at Philips sites within the clinical trial NISCI [2]), including some bug fixes within several of the B1-mapping functions (regarding data type),
and a visual quality tool developed by E. Balteau.
[1] Multiparameter mapping of relaxation (R1 , R2 *), proton density and magnetization transfer saturation at 3 T : A multicenter dual‐vendor reproducibility and repeatability study, T. Leutritz & M. Seif et al., HBM 2020, http://dx.doi.org/10.1002/hbm.25122
[2] http://nisci-2020.eu/