<DIV>Hello Luis,</DIV>
<DIV>Thanks for the reply. I have chanegd interataions to 100. But with same png image, I get different result (I have used ctr. as(79,93). The center that you have specified (30,63) seems to be out of brain area. Also, if i use original raw image, output is blank. I am not sure what is going wrong. (I saved original raw image as .png image by making it 8-bits per channel from 16-bit per channel in photoshop , and then sent to you)</DIV>
<DIV>Neha.<BR><BR><B><I>Luis Ibanez <luis.ibanez@kitware.com></I></B> wrote:</DIV>
<BLOCKQUOTE class=replbq style="PADDING-LEFT: 5px; MARGIN-LEFT: 5px; BORDER-LEFT: #1010ff 2px solid"><BR>Hi Neha,<BR><BR>Thanks for posting your image.<BR><BR>The ThresholdSegmentationLevelSet worked fine in this image,<BR><BR>I used the example code in:<BR><BR><BR>Insight/Examples/Segmentation/<BR>ThresholdSegmentationLevelSetImageFilter.cxx<BR><BR><BR>Just replaced the maximum number of iteration in line 200.<BR><BR>The current value is 1200 iteration,<BR>you must use here 100 iterations.<BR><BR><BR>Then in the command line use<BR><BR><BR>ThresholdSegmentationLevelSetImageFilter<BR>tumor57.png segmented.png 30 63 5 70 207<BR><BR><BR>(in a single command line)<BR><BR>The numbers: 30 63 are respectively the X,Y approximate coordinates<BR>of the tumor center.<BR><BR>The number: 5 is a radius to be used for creating a initial level set<BR>as a circle of radius 5 pixels centered at (30,63).<BR><BR>The numbers: 70 and 207 and the lower and upper thresholds that define<BR>a border for the
tumor.<BR><BR><BR><BR>In the future you may find useful to use the interactive application<BR><BR><BR>InsightApplications/<BR>ThresholdSegmentationLevelSetFltkGui<BR><BR><BR>That allows you to set all this parameters interactively and see the<BR>evolution of the level set iteration by iteration.<BR><BR>Note that it is *very* important to monitor/supervise the level set<BR>because if you run it for too few iterations the tumor will be under<BR>segmented, an if you run it for too many, the level set will leak and<BR>start invading other brain structures.<BR><BR>Looking at the level set as it is evolving is the best way to figure<BR>out how many iterations are enough for stopping the level set.<BR><BR>By fine tunning the values of the lower/upper thresholds, the curvature<BR>scaling and the number of iterations you will be able to refine the<BR>borders of the segmented region.<BR><BR>You could also post-process the segmented image with a filter such as<BR>the BinaryMedianImageFilter or
the new itkVotingBinaryImageFilter.<BR><BR><BR><BR>Please find attached the PNG image with the tumor segmented from your<BR>image.<BR><BR><BR><BR>Regards,<BR><BR><BR>Luis<BR><BR><BR><BR>---------------------------------------------------------------------------<BR>neha k wrote:<BR><BR>> Hello Luis,<BR>> <BR>> Thanks for your reply. While using this filter, with my Brain tumor <BR>> Image attached, I am not getting any luck with application, i have tried <BR>> different parameter values. <BR>> I am not sure what Level Set filters would be good to delineate tumor in <BR>> my image attached. Smoothing did not seem to help for using <BR>> ThresholdLevelSet and ShapeDetection Level Set. My original Image is <BR>> RAW with .mhd header file.<BR>> Any suggestions? How can I tune pramaters to separate tumor? What type <BR>> of preprocessing and other filters might help?<BR>> <BR>> Thanks,<BR>> Neha<BR>> <BR>> */Luis Ibanez
<LUIS.IBANEZ@KITWARE.COM>/* wrote:<BR>> <BR>> <BR>> Hi Neah,<BR>> <BR>> Thanks a lot for posting such a detailed message.<BR>> <BR>> <BR>> We attempted to reproduce the result that you reported,<BR>> but the program worked fine for us.<BR>> <BR>> <BR>> <BR>> Here are the command line parameters that we used,<BR>> and the output from the program:<BR>> <BR>> <BR>> A) Command line:<BR>> <BR>> ShapeDetectionLevelSetFilter BrainT1Slice.png T1.png<BR>> 81 113 5 1 -0.5 3 0.05 1<BR>> <BR>> Output:<BR>> <BR>> Max. no. iterations: 800<BR>> Max. RMS error: 0.02<BR>> <BR>> No. elpased iterations: 108<BR>> RMS change: 0.0199696<BR>> <BR>> <BR>> <BR>> B) Command line:<BR>> <BR>> ShapeDetectionLevelSetFilter BrainProtonDensitySlice.png PD.png<BR>> 81 113 5 1 -0.5 3 0.05 1<BR>> <BR>> Output:<BR>> <BR>> Max. no. iterations: 800<BR>> Max. RMS error: 0.02<BR>> <BR>> No. elpased
iterations: 125<BR>> RMS change: 0.0189581<BR>> <BR>> <BR>> The output images (T1.png and PD.png) are attached.<BR>> <BR>> <BR>> <BR>> <BR>> Did you modified th! e code of this Example<BR>> in any way before running it ?<BR>> <BR>> <BR>> <BR>> Please let us know,<BR>> <BR>> <BR>> Thanks<BR>> <BR>> <BR>> <BR>> Luis<BR>> <BR>> <BR>> --------------------------<BR>> neha k wrote:<BR>> <BR>> > Hello,<BR>> ><BR>> > I am trying to check with<BR>> ITK1.8\Example\ShapeDetectionLevelSetFilter,<BR>> > and do not get desired output. I tried this compiled example with<BR>> > following input images and parameters -<BR>> ><BR>> > *1. BrainT1Slice.png(input image from \ITK\Examples\Data) - with *<BR>> ><BR>> > Seed Point - (81 113) , Distance - 5.0,<BR>> ><BR>> > Sigma SigmoidAlpha SigmoidBeta are -- 1 .0 -0.5 3.0<BR>> ><BR>> > Curv. Scaling
-- 0.05 Propogation Scaling -- 1<BR>> ><BR>> > *Output File Name Attached - ShapeDetectionLevelSetFilterOutput4.png*<BR>> ><BR>> > *2. BrainProtonDensitySlice.png with Seed Point - (81,114) and<BR>> rest of<BR>> > the parameters *have same values as above.<BR>> ><BR>> > Problem - Result of SigmoidFilter is Blank Image while output ! of<BR>> > ShapeDetectionLevelSet Filter for both images is something weird.<BR>> > (attached files)<BR>> ><BR>> > Any input on this appreciated.<BR>> ><BR>> > Thanks,<BR>> ><BR>> > Neha<BR>> ><BR>> ><BR>> ><BR>> <BR>> <BR>> > ATTACHMENT part 2 image/png name=T1.png<BR>> <BR>> <BR>> > ATTACHMENT part 3 image/png name=PD.png<BR>> <BR>> __________________________________________________<BR>> Do You Yahoo!?<BR>> Tired of spam? Yahoo! Mail has the best spam protection around<BR>> http://mail.yahoo.com<BR>>
<BR>> ------------------------------------------------------------------------<BR>> <BR><BR><BR><BR>> ATTACHMENT part 2 image/png name=tumor57Segmented.png<BR></BLOCKQUOTE><p>
<hr size=1>Do you Yahoo!?<br>
<a href="http://us.rd.yahoo.com/mail_us/taglines/security/*http://promotions.yahoo.com/new_mail/static/protection.html">Yahoo! Mail</a> - You care about security. So do we.