Difference between revisions of "DBP:Harvard:Software:Minutes:3"

From NAMIC Wiki
Jump to: navigation, search
m (Update from Wiki)
m (Update from Wiki)
 
Line 1: Line 1:
−
== DTI Meeting Minutes: 3-29-06 ==
+
== DTI Meeting 3/8/2006 ==
  
−
<br /> New Masking Method
+
Bugs:
  
−
* Raul finished creating a new method for masking (see in slicer2-nightly); the mask is created during the conversion of the DWI volume, and lets you set a parameter to control how aggressive the mask is, and whether or not to remove islands within the brain (e.g., ventricles).
+
* Scalar maps: slicer26 crashes when created scalar maps from tensor volumes; this can be addressed at a DTI friday meeting (3/17 is the next one)
 +
* Masks: masks created in DTMRI are too aggressive, while masks created using Stephan's LSDI reconstruction are ideal. Doug met with Raul and added to BugTracker.
 +
* Seeding Parameters: changes to settings in DTMRI > Tracts > Settings do not affect tracts seeded from ROIs; added to BugTracker and assigned to Raul.
  
−
<br /> DTMRI Scalar Maps
+
<br /> Software Testing:
  
−
* L1[xyz] have been added to the Scalar map creation menu (now as "MaxEigenvalueProjection[xyz]"
+
* there are still discrepancies between the DTMRI and Matlab results, and between DTMRI and LSDIrecon; see most recent data (~2 months old) [[DBP:Harvard:Software:Testing:DTI:Slicer|here]]
 +
* PNL needs to have accurate calcalations before converting data to NRRD; this will also be addressed at the next DTI Friday meeting
  
−
<br /> NRRD Volumes
+
<br /> Reading Compressed Volumes:
  
−
* In addition to the DWI nrrd volume, there should be a separate baseline nrrd and a movement correction (FIL) nrrd. If anyone needs other separate volumes, they should let Sylvain know asap.
+
* high priority as PNL needs to conserve disk space for new data (which is ultimately used by everybody)
 +
* Writing compressed files is already supported for structural and labelmap volumes
  
−
<br /> Tensor estimation
+
<br /> Scanning:
  
−
* Still not settled regarding negative eigenvalues; this will be addressed at the LMI meeting
+
* Application was submitted yesterday; it should be about 10 days before we can start scheduling time to scan.
 +
 
 +
<br /> Diffusion Spectrum Imaging
 +
 
 +
* CF went to talk yesterday at MGH on DSI; will be inviting the speaker here in 2-3 weeks
 +
* See related paper here: [http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=16247738&query_hl=1&itool=pubmed_docsum http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=16247738&query_hl=1&itool=pubmed_docsum]

Latest revision as of 13:55, 18 December 2006

Home < DBP:Harvard:Software:Minutes:3

DTI Meeting 3/8/2006

Bugs:

  • Scalar maps: slicer26 crashes when created scalar maps from tensor volumes; this can be addressed at a DTI friday meeting (3/17 is the next one)
  • Masks: masks created in DTMRI are too aggressive, while masks created using Stephan's LSDI reconstruction are ideal. Doug met with Raul and added to BugTracker.
  • Seeding Parameters: changes to settings in DTMRI > Tracts > Settings do not affect tracts seeded from ROIs; added to BugTracker and assigned to Raul.


Software Testing:

  • there are still discrepancies between the DTMRI and Matlab results, and between DTMRI and LSDIrecon; see most recent data (~2 months old) here
  • PNL needs to have accurate calcalations before converting data to NRRD; this will also be addressed at the next DTI Friday meeting


Reading Compressed Volumes:

  • high priority as PNL needs to conserve disk space for new data (which is ultimately used by everybody)
  • Writing compressed files is already supported for structural and labelmap volumes


Scanning:

  • Application was submitted yesterday; it should be about 10 days before we can start scheduling time to scan.


Diffusion Spectrum Imaging

Retrieved from "https://www.na-mic.org/w/index.php?title=DBP:Harvard:Software:Minutes:3&oldid=4430"