hierarchy Vs. Graph

Questions about Anatomist manipulation

Moderators: denghien, riviere

Post Reply
User avatar
cedric clouchoux
Posts: 33
Joined: Tue Jul 06, 2004 2:54 pm
Location: BIC, MNI, Montreal

hierarchy Vs. Graph

Post by cedric clouchoux »

Hi!
(How) can I extract a hierarchy from a graph? Automatically, or have i to program it? (and is it possible?)
thx

cedric
Guest

Post by Guest »

What do you mean exactly?
User avatar
cedric clouchoux
Posts: 33
Joined: Tue Jul 06, 2004 2:54 pm
Location: BIC, MNI, Montreal

Post by cedric clouchoux »

ok sorry

i just want to know if i can get a hierarchy file (.hie) from a .arg file. I know that both of them follow quite a similar concept, but their attributes are different. So i guess that, from a graph, i can obtain a hierarchy.

My problem is : i have a set of sulci, and this set is actually a graph. But this graph is good for just 1 brain (the one i extracted the sulci). But now, i want this set to be "generic", i.e. directly select this set of sulci on every brain i want (assuming all the sulci are labelled, but this is another problem). A way to do this would be to have a hierarchy containing only the desired sulci, and link it to the "main graph" of sulci of a random brain. This system is already used (i.e. sulcal roots), and i can't use the graph i have for other brains.

C.
User avatar
Jean-Francois Mangin
Posts: 337
Joined: Mon Mar 01, 2004 10:24 am
Location: Neurospin, CEA, France
Contact:

Post by Jean-Francois Mangin »

I am not sure I completely understand what you have in mind, maybe because I am not a sigraph expert (Denis is in Taiwan for 2 weeks...). I understand you want to select a list of specific sulci in any brain, and you want to use the hierarchy mecanism for that? And this list of sulci corresponds to the names that can be found in a given brain, maybe manually labelled? Tt means your hierarchy is flat (no embedded level)?

Do you need the hierarchy in Anatomist for selection purpose and to give specific color codes? or do you need the hierarchy inside your own programs?

In general, I have the feeling that you can simply create this hierarchy manually as an ascii file, because it deos not change all the time?
User avatar
cedric clouchoux
Posts: 33
Joined: Tue Jul 06, 2004 2:54 pm
Location: BIC, MNI, Montreal

Post by cedric clouchoux »

The reason i want to use a hierarchy is the following: for now, i create a sub-graph, containing my set of specific sulci. But now i am thinking : it would be great (and it's actually necessary) if this set could be used as a template set. I think the only way (or at least the most efficient way) to do this is to create a hierarchy, not a graph. So, as you said, it would be a flat one, and i don't care about the color stuff, as it would be used inside my programs.
But as i already have a graph, i wanted to know if i could get a .hie from a .arg.
And, i would like to build the hierarchy not manually. I (almost) have a tool to create the sub-graph from the "main" graph of a brain, using anatomist. Should i create a hierarchy instead of a graph? Big question...
I'm sorry if it doesn't sound very clear...
User avatar
Jean-Francois Mangin
Posts: 337
Joined: Mon Mar 01, 2004 10:24 am
Location: Neurospin, CEA, France
Contact:

Post by Jean-Francois Mangin »

Like often, only Denis can clarify what you can or can not do.
However I even wonder why you need to create a subgraph.
As far as I am understanding, you just need to know the list
of sulci you are interested in?
I bet there is nothing to convert a graph into a hierarchy,
or even to do a hierarchy from a list. Me, Denis and Arnaud have
done most of the current hierarchies by hand.
denghien
Posts: 209
Joined: Mon Apr 05, 2004 1:35 pm
Location: IFR49 / SHFJ Orsay
Contact:

Post by denghien »

Hi Cedric,

In fact, I think too there is only one way to match .hie and .arg : You probably must write a hierarchy file.
This type of file allows match name and lable attributes and you can associate a color with each one of them.
What is origin of your graph ? Are you obtained by BrainVISA ?
Do you know structure of .hie file ?


Isabelle
User avatar
cedric clouchoux
Posts: 33
Joined: Tue Jul 06, 2004 2:54 pm
Location: BIC, MNI, Montreal

Post by cedric clouchoux »

The graph i have comes from anatomist, more precisely from a module of anatomist i wrote. I know the .hie file structure, that's why i thought it wouldn't be very hard to get a .hie from a .arg
obviously, i'd like to incorporate this in my program (i.e do not write .hie file by hand :-) )
User avatar
Jean-Francois Mangin
Posts: 337
Joined: Mon Mar 01, 2004 10:24 am
Location: Neurospin, CEA, France
Contact:

Post by Jean-Francois Mangin »

cedric clouchoux wrote: I know the .hie file structure, that's why i thought it wouldn't be very hard to get a .hie from a .arg
obviously, i'd like to incorporate this in my program (i.e do not write .hie file by hand :-) )
Good luck, I am sure you can do it:-) Do not think Denis did it before.
However Arnaud may have done something for the hierarchy of some gyri.
Ask him if there is a piece of I/O somewhere :idea:
User avatar
riviere
Site Admin
Posts: 1361
Joined: Tue Jan 06, 2004 12:21 pm
Location: CEA NeuroSpin, Saint Aubin, France
Contact:

Post by riviere »

Right, there is no tool to make a hierarchy from an existing graph. You can do it (I guess you know enough about graphs): build a list of used labels in a graph and make a (flat) hierarchy from it.
Denis
Post Reply