updates 2016-08 August - September

Owned by Paul Macey (admin)
Oct 17, 2016
5 min read
2016-09-30
  • Filter and make new groups - fixed a bug with creating a new group (spelling mistake)
2016-09-20
  • FIRST - added visualization option
  • FIRST - now calculate uncorrected p-value files (good for pilot)
2016-09-19
  • DTI - had a weird tensor calculation problem; updated to use 2016 version of Diffusion toolbox tensor calculation. Could not replicate error, but it's still odd as to why the error occurred in the first place.
  • FIRST - option to select multiple regions, corrections, and normalizations, instead of just one/all.
  • FIRST - single region summaries will be saved in separate Excel files.
2016-09-18
  • Speeded up re-loading of FIRST stats (a lot of time was taken unzipping and zipping nii.gz files, so the intensity information is now saved).
2016-09-17
  • Tweaked FIRST run design to deal with Excel file problems, and show all stats not just first one
  • Tidied stats GUI
  • Edit sets now works with non-MRI studies
  • Added option to auto-create female and male subgroups in "Edit set" GUI (great for working in sex-specific studies)

2016-09-16
  • Fixed another minor metric bug (filtering)
  • Fixed bug in Project Add group (very odd, should have caused error for everyone, but seems to be java-dependant; Rania had issue but not other computers): Details: project_id was being converted to string.
  • Little tidying of Project GUI
  • Added some help for FIRST in Stats GUI.

2016-09-15
  • Fixed stats GUI bug introduced a couple of days ago.
2016-09-13
  • Fixed metric load from template bug (only on update)
  • Working on "metric" option, lots of GUI changes
  • Started tidying up font size in GUI's; CSPM is first completed one. When GUI changes size, it will still look ok, including when on different screen sizes.
2016-09-12
  • Fixed a bug in the project GUI
  • Several changes for metric, not ready yet.
2016-09-11
  • Fixed blob display bug in SPM overlays; new option to select interpolation method ("nearest neighbour" usuall works best)

  • Made global normalization options in basic stats slightly clearer on GUI
2016-09-08
  • Fixed bug with displaying blobs in SPM - sublet programming mistake (spm_orthviews, get_blob_data).
2016-09-04
  • Working on metric, debugging, types, loading, stats (descriptive). Still very early stage.
2016-09-03
  • "metric" load from Excel - complete, minimal testing.
2016-09-01
  • In the middle of "metric" changes, unfinished.
2016-08-24
  • Updated all "isconnection" to "isopen" to be compatible with newer matlab versions. This means older matlab versions won't work.
2016-08-23
  • Added ASL labeling efficiency option in CSPM - previously only had it in manual settings.
  • Flip option with DTI should work (not tested yet - Kasia?!)
  • Deals with inconsistent image dimensions/orientations in mean and standard deviation calculations; will show warning but not give error.
2016-08-22
  • Fixed a bug where basic stats did not "remember" initial setting - used to have to click button to get it to remember. Works ok now.
2016-08-19
  • Fixed a bug whereby DTI kept reslicing tensor
  • Fixed a bug - for multiple covariates, was not taking closest in time to the study. (It checks time of two covariate values, and time of MRI/phys, and takes closest covariate value).
  • Fixed a bug for Excel load covariates that comes on some computers (depending on database toolbox settings)
  • First pass at T1R: needs conversion with T1-related parameters, and optimization toolbox.

  • Option to median filter image. Defaults to 3 x 3 x 3 filter window; to chage, use command line (see im_medfilt3.m).

2016-08-18
  • New feature in Subjects GUI to view selected study (if path is there). The "selected" study is the row where the cursor has been clicked. "View" will open File Browser in nifti folder, and open first nifti image (if Windows is set to open nifti images); physiology will be shown, as well as metrics.

 

2016-08-17
2016-08-14
  • Fixed an error Luke had when he did cluster timetrends. Something to do with RMANOVA - it will skip now, with a message.
  • Added a check when installing CSPM for the first time, before db is connected.
2016-08-13
  • Fixed errors in basic stats
  • Tidied VBM/DTI/quantitative-specific global options

2016-08-11
  • Added a check in case accidentally copy incorrect local settings (cspm_local.mat should not be copied, as that is always the local settings and usually will not transfer from computer to computer).
  • Added a catch in case of a Windows file copy failure in FIRST design matrix processing. This error arises with out-of-date Windows or Internet Explorer.

2016-08-08
  • Added option to shift series subfolders to main folder after conversion.
  • Also will deal with subfolders that have "DICOM" 2nd level subfolders (weird format got from CNS). 

2016-08-05
  • Add repeated measure (paired sample t test) option in SPM basic stas; only works with two groups, and only works for sets with the same subjects in each group
    • Tested a couple of times, but still bleeding edge....
  • Added ANCOVA option in basic stats

  • Fixed VOI display bug introduced a couple of days ago ("spm_orthviews"). 
2016-08-04
  • Fixed ASL bug where sinc/surround option would cause error. 

NOTE: if you previously ran with default (simple) subtraction, then ran the sinc/surround, it would mistakenly copy the simple to the sinc/surround files.

2016-08-03
  • Fixed DTI bug in IXI conversion - gradients/b values were wrong, because had not accounted for last file being the average DWI.
  • Minor tweak in DTI display one.
  • Tidied DTI single subject visualization to work with run on, and to save pdf files in matpab\spm12\temp folder.
2016-08-02
  • Copied updated SPM files (otherwise visualization won't work)
  • DTI visualization - tweak to make b0 look good (intensity scaling). 
2016-08-01
  • Made tensor visualization look nice, and give scales to off-diagonal. 
  • Have T1-aligned option
  • Have option to include T1.