On 19.03.2013 20:22, Federico Milano wrote: > Hi Lucas, > > I've been working on this problem and I've found that the > DicomSeriesReader::GantryTiltInformation constructor detects a gantry > tilt in many of the MRI images I've been working with. This happens > because the code in that ctor does not use the gantry tilt attribute > (0018, 1120), that is present in the header, but instead it tries to > identify if a gantry tilt is present by taking two consecutive slices > and measuring the distance between the line passing through the origin > of the second slice in the direction of the normal (the normal is > calculated using the director cosines) and the origin of the first > slice. If this distances is more than 0.001 mm, then there is a tilt; > since gantry tilt is not managed right now, each slice is loaded as a > separate data node. I suppose this is a reasonable way to calculate the > tilt in CT...but I'm not sure about MRI.
Hi Federico, I guess this calculation is reasonable for any kind of sliced data, especially because the tag you mention is described as "not for use in calculations". > > Now, if I modify this behaviour, mitkWorkbench is behaving like the 3M3 > app. In both apps there is still a problem: if you look at the axial and > the sagital windows in the attached screenshot you'll see that the > images are a little rotated. This should not happen (if I load them, for > example, in ITK-Snap, this rotation is corrected). I don't know if this > is the "gantry tilt" correction (mentioned in the code comments) not yet > implemented, but now I'll try to focus in solving this issue. what you see in the attached screenshot is probably a result of the image being not aligned with the world coordinate axes. Try right-clicking the image in the data list at the left of the window, then select "Reinit" (not "Global reinit"). This option sets the render windows to the orientation of the image. Regarding the tilt correction, I am curious to hear if the images are displayed correctly after you try "reinit". If there still is a problem I'd be keen to get a look at the images. I'd be motivated to fix the problem. Kind regards Daniel -- Dr. Daniel Maleike Telefon: + 49 6221 647976 3 Mint Medical GmbH, Friedrich-Ebert-Straße 2, 69221 Dossenheim/Heidelberg Geschäftsführer: Dr. Matthias Baumhauer Registergericht Mannheim, HRB 709351 ------------------------------------------------------------------------------ Everyone hates slow websites. So do we. Make your web apps faster with AppDynamics Download AppDynamics Lite for free today: http://p.sf.net/sfu/appdyn_d2d_mar _______________________________________________ mitk-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/mitk-users
