accumulative dose

[Esther]

Hi,

I tried to change my Matlab to R2009a version and finally get to see
the result of deformable registration through DIRART. What I want to
do is to accumulate daily dose (using CERR) and compare with plan
dose. Before I can do so, I have to deform each daily dose from daily
CT image. In DIRART, I use plan CT image as Fixed Image (image 2) and
daily CT image as Moving Image (image 1). After preprocessing, I run
the deformable registration then deform the Moving Image. Then load
the dose associate with the Moving Image and deform the dose. My
problem is what should I associate and align my deformed dose. Or
what is the reasonable way to do? Originally I choose the dose
associated with the Moving Image and align and associate the deformed
dose to the Moving Image. Then I add this dose into the daily CT
image. This way looks just fine. After I think about it, I think I
should choose the dose associated with the Moving Image and align and
associate the deformed dose to the Fixed Image because I want to add
all daily dose together in CERR by using Plan CT Image. However, this
does not work. Do I do any step wrong? I think the problem is
relative to image set not associated with right scan. I attached the
error code below. If you can help to clean up my thought, that would
be most grateful.

Cheers,
Esther


------------error
code-----------------------------------------------------------------------------
??? Subscript indices must either be real positive integers or
logicals.
Error in ==> getTransM at 60
planField = planC{indexS.scan}(number);
Error in ==> getTransM at 82
transM = getTransM('scan', aS, planC);
Error in ==> calcDoseSlice at 43
transM = getTransM(doseInfo, planC);
Error in ==> showDose at 97
[im, imageXVals, imageYVals] =
calcDoseSlice(doseSets(i), coord, dim, planC,compareMode);
Error in ==> CERRRefresh at 147
showDose(hAxis);
Error in ==> sliceCallBack at 2190
CERRRefresh
??? Error while evaluating uimenu Callback

[Deshan Yang]

You certainly should associated the deformed dose to the fixed image
so you can add multiple deformed dose together and compare them to the
planning dose.

I guess the error is something relating to your new CERR version, or
more precisely, DIRART does not set all the parameters correctly
before adding the deformed dose back to the CERR plan. It worked fine
with older CERR versions.

Which CERR version do you use?


Deshan Yang, PhD

[Esther]

When I posted this message, I used CERR3pt3Beta1_Nov_19_2009 with
older DIRART code. I did try to associated the deformed dose to the
fixed image, but when I add this dose back to CERR plan which contains
original plan CT/dose/structures, the deformed dose was shifted.
However, reviewing this deformed dose in DIRART was fine. I am not
sure where it can go wrong. Please give me a guide.

There is another problem. I found DIRART has an update on 06/02/2011
yesterday and I updated my DIRART and re-run previous plans again. I
cannot get results as good as older version of DIRART. I noticed when
I chose registration algorithm ---> Optical Flow Method ---> Iterative
Optical Flow Method in the older version, it would automatic choose
Fast Demon Method as well and this result is much better then what I
have in newest version. Can you please explain what's the issue?

Can you please let me know what version of CERR should work best with
DIRART? Also, I am not sure if I want to update the DIRART since the
result is poor unless I can figure out how to get the competitive
result for older and newest version of DIRART.

Thank you very much for your help.