[Insight-users] MR Object Extraction and Registering with SPECT.

[valli gummadi]

 This is a multipart mime message


--Next_1079687397---0-202.54.124.153-7369
Content-type: text/html;
	charset=iso-8859-1
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline

<P>=0A<BR>=0A Hi,<BR>=0A <BR>=0A&nbsp; &nbsp; &nbsp; I am working on Regist=
ration of Single and Multi modality images. I am facing problem in register=
ing multi modality images.<BR>=0A <BR>=0ALet me explain the procedure and s=
teps I followed to process multi modlaity images.<BR>=0A<BR>=0ARegistration=
 of MR and SPECT images:<BR>=0A<BR>=0A1.MRimage will be in CORONAL view and=
 SPECT will be in Transverse view.<BR>=0AI converted the SPECT to CORONAL b=
efore giving it for registration.<BR>=0A<BR>=0A2.Actual requirement is to e=
xtract the Brain object from the MR and to register that object with the sp=
ect. I extracted Object from the MR using itkNeighbourhoodConnectionImageFi=
lter and used the exctracted brain as Fixed image and Registered MROBject w=
ith SPECT.<BR>=0A<BR>=0A3.The registered spect is transformed and fuse it w=
ith the original MR so that spect fuses with the brain and not with the oth=
er tissues.<BR>=0A<BR>=0AI am&nbsp; using following itk classes for registr=
ation and transform <BR>=0Aproces:<BR>=0A1. MultiResolutionImageRegistratio=
nMethod<BR>=0A2.MattesMutualInformationImageToImageMetric<BR>=0A3.LinearInt=
erpolateImageFunction<BR>=0A4.RecursiveMultiResolutionPyramidImageFilter<BR=
>=0A5.CenteredAffineTransform &amp; centered Affine Transform Initializer<B=
R>=0A6.RegularStepGradientDescentOptimizer<BR>=0A7.ResampleImageFilter<BR>=
=0A<BR>=0Ano of iterations =3D450;<BR>=0ATranslateScale =3D1000000;<BR>=0Ao=
ptimizer Scales =3D1;<BR>=0AOptimizer maximum step length =3D16;<BR>=0AOpti=
mizer minimum step length =3D2.5;<BR>=0ASpatialSamples =3D1000;<BR>=0ABins =
=3D50;<BR>=0ANo of Levels =3D3;<BR>=0ASetMinimize =3DTRUE;<BR>=0A<BR>=0A<BR=
>=0AAfter fusion, I observed that SPECT is not exactly alighned with brain =
object of the MR Volume.<BR>=0APlease,Anybody help me to implement the regi=
stration to acieve the expected output.<BR>=0A<BR>=0AThanking You,<BR>=0ARe=
gards,<BR>=0ASrivalli.<BR>=0A<BR>=0A<BR>=0A<BR>=0A<BR>=0A=0A</P>=0A<br><br>=
=0A<A target=3D"_blank" HREF=3D"http://clients.rediff.com/signature/track_s=
ig.asp"><IMG SRC=3D"http://ads.rediff.com/RealMedia/ads/adstream_nx.cgi/www=
.rediffmail.com/inbox.htm at Bottom" BORDER=3D0 VSPACE=3D0 HSPACE=3D0 HEIGHT=
=3D74 WIDTH=3D496></a>=0A
--Next_1079687397---0-202.54.124.153-7369
Content-type: text/plain;
	charset=iso-8859-1
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline

=0A Hi,=0A =0A      I am working on Registration of Single and Multi modali=
ty images. I am facing problem in registering multi modality images.=0A =0A=
Let me explain the procedure and steps I followed to process multi modlaity=
 images.=0A=0ARegistration of MR and SPECT images:=0A=0A1.MRimage will be i=
n CORONAL view and SPECT will be in Transverse view.=0AI converted the SPEC=
T to CORONAL before giving it for registration.=0A=0A2.Actual requirement i=
s to extract the Brain object from the MR and to register that object with =
the spect. I extracted Object from the MR using itkNeighbourhoodConnectionI=
mageFilter and used the exctracted brain as Fixed image and Registered MROB=
ject with SPECT.=0A=0A3.The registered spect is transformed and fuse it wit=
h the original MR so that spect fuses with the brain and not with the other=
 tissues.=0A=0AI am  using following itk classes for registration and trans=
form =0Aproces:=0A1. MultiResolutionImageRegistrationMethod=0A2.MattesMutua=
lInformationImageToImageMetric=0A3.LinearInterpolateImageFunction=0A4.Recur=
siveMultiResolutionPyramidImageFilter=0A5.CenteredAffineTransform & centere=
d Affine Transform Initializer=0A6.RegularStepGradientDescentOptimizer=0A7.=
ResampleImageFilter=0A=0Ano of iterations =3D450;=0ATranslateScale =3D10000=
00;=0Aoptimizer Scales =3D1;=0AOptimizer maximum step length =3D16;=0AOptim=
izer minimum step length =3D2.5;=0ASpatialSamples =3D1000;=0ABins =3D50;=0A=
No of Levels =3D3;=0ASetMinimize =3DTRUE;=0A=0A=0AAfter fusion, I observed =
that SPECT is not exactly alighned with brain object of the MR Volume.=0APl=
ease,Anybody help me to implement the registration to acieve the expected o=
utput.=0A=0AThanking You,=0ARegards,=0ASrivalli.=0A=0A=0A=0A=0A
--Next_1079687397---0-202.54.124.153-7369--