[Insight-users] Using Bacterial Colonies for segmentation : Morphogenesis Application

[Luis Ibanez]

Hi Yasser,

Note first that for using the Morphogenesis
application in 3D you must go to the file

   InsightApplications/Morphogenesis/Cell.h

and change line 41 to be

   Dimension = 3

instead of

   Dimension = 2


---

Also, the Morphogenesis application requires you
to write two small text files with parameters.

This is far from being an ideal interface...
but it is the way it is working right now...


So,
please write text files with names

    CellularAggregate.dat
    GradientEatingBacteria.dat


In the file CellularAggregate.dat you put
values defining the X,Y,Z coordinates of
the first egg cell.

This is plain ascii text just like

          124  34  68

where those values are in millimeters
(e.g.  physical units of your image,
not pixel units).  [ This should be replaced
with an interactive  mouse click... whenever
we got time to work further in this application...]


In the file GradientEatingBacteria.dat you
put values

1) Default cell radius
2) Growth radius increment
3) Growth radius limit
4) Chemo attractant low value
5) Chemo attractant high value

The meaning of these values is:

1) the size in millimeters of the
    cell at birth time. A typical
    value is 2.5

2) how much the cell grows at each
    iteration. A typical value is 0.1

3) The adult size radius of the
    bacterial cell. A typical value is
    5.0.  This values defines how
    coarse the mesh of the cellular
    colony will be by the end of the
    segmentation.

4 & 5) are threshold values of gray
    levels from your image.
    The bacteria will grow and reproduce
    happily in the regions of the image
    having gray level values in the range
    defined by these two values.

So, a typical file will look like

2.5
0.1
5.0
120
180

For filling in a region with gray level
intensities in the range [120:180].


Put these two files in the directory
from which you are executing the Morphogenesis
executable.


When you run the executable, do the following:


1) Load your dataset using the option
    "Load Substrate", give it a string name,
    and click on the option "make visible".

    You can zoom in/out by using Ctrl+rightmousebutton
    and dragging.

2) Go to the "View" menu and select "Species",
    the smarter bacteria at this point is the
    "Gradient Eating Bacteria", select this species
    and click on "Create Egg".

    At this point a sphere should appear on the
    3D view.

3) then click on the "play" button. The one with
    the black triangular label. The bacteria should
    start growing and reproducing filling in the
    region of the image defined by the two threshold.

    Make sure that the initial position of the egg
    cell is in a region with gray values in the
    range defined by the two threshold values.


    You can stop the activity of the bacterial colony
    by clicking on the "stop" button. (the black square).

    Note that the method will never stop by itself.
    Bacteria will keep trying to reproduce. You simply
    stop it when you judge that no new cells are being
    born (a cell counter is in the interface)
    [This could also be easily fixed by setting a timeout
     for the next cellular division. E.g. stop the colony
     if no cells are divided in a period of 3 seconds]



Please let us know if you have any further question.


   Thanks



      Luis




-------------------------
Yasser Bashir wrote:

> Thank you Luis.
..
> 
> I have a question regarding the Morphogenesis application. Can you please
> tell me how can i generate a bacterial colony inside my 3D image.  It seems
> to work fine for 2D slices but when i load my segmented 3D image and try
> generating the bacterial colony, nothing happens i.e. im unable to see the
> bacterial colony grow.
> 
> Regards
> 
> yasser.
> 
> 
> 
>