Wound Healing Example

Hi,
I am trying to use CellProfiler version 2.0 on Windows XP OS. For some reason I am having error messages. Here is the log on the execution of the ExampleWoundHealing on the images from the examples posted on the site (executed on the first image only). I tried to run another example (ExampleVitra) and got similar errors. Can you help me figure out what is the problem?

Pipeline saved with CellProfiler SVN revision 9722
Wed May 26 09:46:20 2010: Image # 1, module LoadImages # 1: 0.67 sec (bg)
Wed May 26 09:46:20 2010: Image # 1, module ColorToGray # 2: 0.83 sec (bg)
Wed May 26 09:46:21 2010: Image # 1, module Smooth # 3: 2.47 sec (bg)
C:\Program Files\CellProfiler\library.zip\numpy\lib\function_base.py:324: Warnin
g:
The new semantics of histogram is now the default and the new
keyword will be removed in NumPy 1.4.

Wed May 26 09:46:26 2010: Image # 1, module IdentifyPrimaryObjects # 4: 7.31 sec
(bg)
Traceback (most recent call last):
File “cellprofiler\pipeline.pyc”, line 1133, in run_with_yield
File “cellprofiler\modules\identifyprimaryobjects.pyc”, line 1179, in display
File “cellprofiler\gui\cpfigure.pyc”, line 810, in subplot_imshow
MemoryError
Traceback (most recent call last):
File “cellprofiler\gui\cpfigure.pyc”, line 292, in on_paint
File “matplotlib\backends\backend_wxagg.pyc”, line 59, in draw
File “matplotlib\backends\backend_agg.pyc”, line 314, in draw
File “matplotlib\artist.pyc”, line 46, in draw_wrapper
File “matplotlib\figure.pyc”, line 773, in draw
File “matplotlib\artist.pyc”, line 46, in draw_wrapper
File “matplotlib\axes.pyc”, line 1701, in draw
File “matplotlib\artist.pyc”, line 46, in draw_wrapper
File “matplotlib\image.pyc”, line 237, in draw
File “matplotlib\image.pyc”, line 182, in make_image
File “matplotlib\cm.pyc”, line 167, in to_rgba
File “matplotlib\colors.pyc”, line 527, in call
MemoryError
Traceback (most recent call last):
File “matplotlib\backends\backend_wx.pyc”, line 1155, in _onPaint
File “matplotlib\backends\backend_wxagg.pyc”, line 59, in draw
File “matplotlib\backends\backend_agg.pyc”, line 314, in draw
File “matplotlib\artist.pyc”, line 46, in draw_wrapper
File “matplotlib\figure.pyc”, line 773, in draw
File “matplotlib\artist.pyc”, line 46, in draw_wrapper
File “matplotlib\axes.pyc”, line 1701, in draw
File “matplotlib\artist.pyc”, line 46, in draw_wrapper
File “matplotlib\image.pyc”, line 237, in draw
File “matplotlib\image.pyc”, line 182, in make_image
File “matplotlib\cm.pyc”, line 167, in to_rgba
File “matplotlib\colors.pyc”, line 527, in call
MemoryError

When I get those MemoryError messages it is because I ask the computer to perform a task its RAM can’t handle. This usually happends to me when the image is to large or you require too many cycles in the step at which you get the memoryerror message. Have you tried with smaller or lower resolution images?

It works on lower resolution images (at least on the wound healing example) - thanks! Althoug it is still strange that the application rquires so much memory. My computer has 2GB RAM, it should be sufficient to hold manipulation of a single 1024x1024 RBG images (the size of my data). Is it a common problem when using CellProfiler? I am thinking about using it, and plan to spend some time in exploring it, so it is a crutial question for me.

Thanks.

Well, it usually depends on the choices in the module and the amount of iformation on the image.
I remember running into memoryerror issues when aplying certain thresholds, I think lapladician or something like that, special when you rise the number of…(I don’t remeber what was that value exactly hehe). The same goes for radial distribution and neighbours meassurements.
On the other hand, I may be wrong, but establishing a threshold early may help you to reduce the amount of RAM required in later modules as it simplifies the amount of data to analyse and in general reduces the overlaping between objects.
By the way I’m analysing 512x512 images on a computer with Win XP and 250MB of RAM. But it takes its time of course, in fact I only use this for finetunning the pipeline. I just use this computer because it’s in the lab but it does the job.
You can check if there are processes taking to much RAM like antivirus or just to many processes.
In my case I use a modified Windows XP which has many services that usually nobody uses disabled, it’s called Suricata Mangosta. It runs on something like 50Mb RAM so you still have cap space for extra processing.