I just splitted the bug into two separate ones:

• doxygen user manual with detailed usage description (T2705)
• "advertisement" page on mitk.org (T2704)

Segmentation perspective consisting of the Data Manager view and the Segmentation view:

therappy support -> theraPy support
similar for many parts -> similar in...
However, the distinction between a foreground ... -> Please stretch into more sentences, kind of confusing

Segmentation perspective makes a number -> makes = is based on
Segmentation perspective makes use of the -> perspektive uses

Segment : Data Manager is used for selection of the current segmentation

absolute control over which voxels are -> control of those voxels that...
The drawback of manual contouring is that might need to -> is the possible need to... ; remove 2nd 'manually' in that sentence
Notice that the name input suggests -> Add quotation marks

Segment: Correction tool
The user does not have to draw a closed contour to use the Correction tool and does not need -> You.. and you don't need (since it is used in the rest of the article)
segmenation -> segmentation,
fully inside the segmentation -> , <- the marked region

if more complex contours are drawn, (a) and/or (b) operations are applied segment-wise - didn't get what you mean

Segment: Interpolation

time-consuming because strucutres of -> structures ; comma in front of because
The Segmentation View -> Quotation marks
has neighboring slices which have segmentations -> has got ... with segmentations

is completly clear of a manual segmentation - completely ... clear of manual segmentations

if there is even only a single pixel of segmentation in the current slice. -> didn't get it

are displayed in different way than manual segmentations -> in a different way than manual segmentations ARE -> , <- until

If you have segmented a whole organ in a every-x-slices way, you may also review the interpolations and then accept all of them at once by clicking "... all slices". - How do i review, what is every-x-slices-way ?

Segment: Organ segmentation

The manual contouring described - Quotation marks

Segment: Liver on CT images
All you have to do is load and - have to do is to load and to select

Segment: Heart and lung on MRI
is performed fully automatic -> automatically
tools for segmentation of the heart and the - of heart and lungs
minimum - a bit, small or something like that - is the sentence required anyway ?
Click one of the MR to add -> Click on - what is MR
the then-automatic segmentation process - didnt get it, maybe quotation marks ?, remove 'then'?

The algorithms used for segmentation of the heart and lung are method which need training - of heart and lungs are a method, which needs...

Why does it need training ? Can it be done before selling the program ?

Segment other Organs:

on all kind - on all kinds
than offerEEd by the current version of the Segmentation view - than those offered , Quotation Marks
cooperation - cooperations

Segment: Lesion segmentation

differnt - different

in that lesions are not part of the healthy body - didnt get it
and oftenly they are found if varying places - and they are often found in...

Segment: Segmentation of hypodense or hyperdense lesions on CT images

ways to do so using the - so by using, quotation marks

the algorithms has automatically - have

[SVN revision 20158]
DOC (#2683): minor editing

Klaus and Tobias, could you please have a look at lines 201-203 of

mbi-sb/BundlesQt/org.mitk.gui.qt.lymphnodevolumetry/documentation/org_mitk_gui_qt_segmentation.dox

These lines should explain what kind of MRI images are needed for the shape models to work correctly. I copied the information that Tobias already has put into the wiki but I'm not sure if the names of these sequences are understandable for outside-DKFZ users.

Forgot to CC Klaus/Tobias. Please see comment #4

Marco or Jan, do you have an idea where to put the Doxygen files of the user manual for the Segmentation perspective? Currently I put them into

mbi-sb/BundlesQt/org.mitk.gui.qt.lymphnodevolumetry/documentation/org_mitk_gui_qt_segmentation.dox

This is because Marco suggested not to mention closed-source parts of the program that might be unavailable for users who build the open source MITK by themselves. (that makes sense and avoids confusion)

If you already know someone who coordinates collecting, generating and integrating all manual pages, please reassign this bug to him/her.

Sorry, I do not no more about the training datasets. Tobi?

Klaus or Sarah: Sarah is writing her thesis about these datasets, right? She should know what kind of images these are. Klaus, I don't know if Sarah reads her Bugzilla-Mails, could you perhaps ask her when she's in the office?

sorry, my checkout was from Friday, the version did not include the line "Hippocampus segmentation". I just added some information about the datasets.

Thanks! This closes this bug.

Added some more info, too.
Answering comment #4: The MRI Sequence names are unique and not DKFZ-internal.
Every radiologist should know them. Can also be found by Google.
These names are actually the only way to describe the used sequences.

Reopened for commit

[SVN revision 20172]
FIX (#2683): Added docu for heart MRI sequences