fusion of CT and SPECT

[roland makay]

Hi,
For preoperative planning I am trying to fusion
CT/SPECT images. The fusion image gives anatomical
information (CT) with an overlay of SPECT showing
functions in different areas. I used the commercially
also available software of Hermes Medical Systems for
that. Now, to make a free solution, I'd like to do it
with ImageJ. Formerly, Arnaud Dieudonné already
created a Fusion J, which works fine (thanks!) if the
images are registered, meaning every voxel of the CT
matches each corresponding voxel of the SPECT (hope I
am describing more less right, I am not imaging
expert, just doing surgery). So, the main problem is
this registration process, which is basically
downgrading CT resolution to SPECT. So far I tried
TransformJ, but it works only for reslicing, and the
implemented size adjust capability of ImageJ is not
enough.
Does anybody have an ideia for the problem?
Sincerely,
Roland Makay

__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around
http://mail.yahoo.com 

[J. Anthony Parker]

Roland,

My plugin, Align3_TP, may help.  "MoreSites with Plugins" / "Tony  
Parker" will lead you to http://www.med.harvard.edu/JPNM/ij/ 
plugins/.  Select "Alignment".

This plugin aligns two stacks using a series of 2D operations.  It is  
largely manual; thus, not well suited for production work.  If you  
are used to Hermes, it may be laborious.  There is a manual which  
explains the operations.

"Show Alignment" provides some tools for visualizing two slices which  
are already aligned.

Hope this is of some help.

Tony
J. Anthony Parker, MD PhD
[hidden email]
tel [phone] fax -2185 comp -5275
page 617 632-7243 #31978
330 Brookline Ave, Boston, MA, 02215-5400


On Jun 27, 2006, at 1:27 PM, roland makay wrote:

> Hi,
> For preoperative planning I am trying to fusion
> CT/SPECT images. The fusion image gives anatomical
> information (CT) with an overlay of SPECT showing
> functions in different areas. I used the commercially
> also available software of Hermes Medical Systems for
> that. Now, to make a free solution, I'd like to do it
> with ImageJ. Formerly, Arnaud Dieudonné already
> created a Fusion J, which works fine (thanks!) if the
> images are registered, meaning every voxel of the CT
> matches each corresponding voxel of the SPECT (hope I
> am describing more less right, I am not imaging
> expert, just doing surgery). So, the main problem is
> this registration process, which is basically
> downgrading CT resolution to SPECT. So far I tried
> TransformJ, but it works only for reslicing, and the
> implemented size adjust capability of ImageJ is not
> enough.
> Does anybody have an ideia for the problem?
> Sincerely,
> Roland Makay
>
> __________________________________________________
> Do You Yahoo!?
> Tired of spam?  Yahoo! Mail has the best spam protection around
> http://mail.yahoo.com

[roland makay]

Thanks Tony, I'll try...

--- "J. Anthony Parker"
<[hidden email]> wrote:

> Roland,
>
> My plugin, Align3_TP, may help.  "MoreSites with
> Plugins" / "Tony  
> Parker" will lead you to
> http://www.med.harvard.edu/JPNM/ij/ 
> plugins/.  Select "Alignment".
>
> This plugin aligns two stacks using a series of 2D
> operations.  It is  
> largely manual; thus, not well suited for production
> work.  If you  
> are used to Hermes, it may be laborious.  There is a
> manual which  
> explains the operations.
>
> "Show Alignment" provides some tools for visualizing
> two slices which  
> are already aligned.
>
> Hope this is of some help.
>
> Tony
> J. Anthony Parker, MD PhD
> [hidden email]
> tel [phone] fax -2185 comp -5275
> page 617 632-7243 #31978
> 330 Brookline Ave, Boston, MA, 02215-5400
>
>
> On Jun 27, 2006, at 1:27 PM, roland makay wrote:
>
> > Hi,
> > For preoperative planning I am trying to fusion
> > CT/SPECT images. The fusion image gives anatomical
> > information (CT) with an overlay of SPECT showing
> > functions in different areas. I used the
> commercially
> > also available software of Hermes Medical Systems
> for
> > that. Now, to make a free solution, I'd like to do
> it
> > with ImageJ. Formerly, Arnaud Dieudonné already
> > created a Fusion J, which works fine (thanks!) if
> the
> > images are registered, meaning every voxel of the
> CT
> > matches each corresponding voxel of the SPECT
> (hope I
> > am describing more less right, I am not imaging
> > expert, just doing surgery). So, the main problem
> is
> > this registration process, which is basically
> > downgrading CT resolution to SPECT. So far I tried
> > TransformJ, but it works only for reslicing, and
> the
> > implemented size adjust capability of ImageJ is
> not
> > enough.
> > Does anybody have an ideia for the problem?
> > Sincerely,
> > Roland Makay
> >
> > __________________________________________________
> > Do You Yahoo!?
> > Tired of spam?  Yahoo! Mail has the best spam
> protection around
> > http://mail.yahoo.com
>

__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around
http://mail.yahoo.com 

[Arnaud DIEUDONNE]

Hi Roland,
I don't have a solution for you on ImageJ.
But I can advise you to use AIR (Automated Image Registration),
http://bishopw.loni.ucla.edu/AIR5/index.html
This is the best free registration software I know.
To install this software you need to compile the sources, the
documentation can help you.

Arnaud

roland makay <[hidden email]> a écrit :

> Hi,
> For preoperative planning I am trying to fusion
> CT/SPECT images. The fusion image gives anatomical
> information (CT) with an overlay of SPECT showing
> functions in different areas. I used the commercially
> also available software of Hermes Medical Systems for
> that. Now, to make a free solution, I'd like to do it
> with ImageJ. Formerly, Arnaud Dieudonné already
> created a Fusion J, which works fine (thanks!) if the
> images are registered, meaning every voxel of the CT
> matches each corresponding voxel of the SPECT (hope I
> am describing more less right, I am not imaging
> expert, just doing surgery). So, the main problem is
> this registration process, which is basically
> downgrading CT resolution to SPECT. So far I tried
> TransformJ, but it works only for reslicing, and the
> implemented size adjust capability of ImageJ is not
> enough.
> Does anybody have an ideia for the problem?
> Sincerely,
> Roland Makay
>
> __________________________________________________
> Do You Yahoo!?
> Tired of spam?  Yahoo! Mail has the best spam protection around
> http://mail.yahoo.com
>

--
Arnaud Dieudonné
Doctorant en Physique Médicale
Laboratoire QUANT.I.F.
Faculté de Médecine et Pharmacie de Rouen
Tel: [phone]