From minc-users@bic.mni.mcgill.ca Tue May 3 04:06:04 2005 From: minc-users@bic.mni.mcgill.ca (Anna Caroli) Date: Tue May 3 03:06:04 2005 Subject: [MINC-users] Windows-UNIX problems with Display Message-ID: <200505030705.j4375uPf6543174@shadow.bic.mni.mcgill.ca> Hi all, I have labels drawn using Display for Windows, and I need to use them on a SUN machine, because I need to process them with other MNI tools (labels are drawn over linearly registered images, and I need to apply a back transformation to have labels in the native space). The problem is I can't open with SUN and SGI machines labels.tag drawn using Display for Windows... What I get is the following error message: input_tag_points(): invalid header in file. (I have the same problem the other way around: when I have a label drawn over a UNIX machine, saved as filename.tag and I try to open it over a Windows machine, I can't) Using Display for Windows I also tried to open the label filename.tag and save it as filename.mnc to export it to the SUN (I don't have problems in importing/exporting and opening files.mnc): what I get from the correct file.tag is a file where all the drawn structures are labelled as "structure 1", irrespectively... Has anybody of you had the same problems? Can you please help me in overcoming them? Thanks a lot for your help and have a nice day! Anna Laboratory of Epidemiology, Neuroimaging & Telemedicine IRCCS San Giovanni di Dio FBF The National Center for Research and Care of Alzheimer's Disease via Pilastroni 4, 25125 - Brescia, Italy Tel: +39 030 3501 361, Fax: +39 02 700 435 727 www.centroAlzheimer.it From minc-users@bic.mni.mcgill.ca Tue May 3 04:57:05 2005 From: minc-users@bic.mni.mcgill.ca (Anna Caroli) Date: Tue May 3 03:57:05 2005 Subject: [MINC-users] Labels interpolation with Display Message-ID: <200505030756.j437uePf6585436@shadow.bic.mni.mcgill.ca> Hi all again, Just a quick question: is it possible to interpolate labels drawn with Display before calculating the volumes? I know it's possible to interpolate the input image.mnc over which to draw labels; is it posssible to interpolate also the drawn label to better fit the structure? Thanks for your help! Anna Laboratory of Epidemiology, Neuroimaging & Telemedicine IRCCS San Giovanni di Dio FBF The National Center for Research and Care of Alzheimer's Disease via Pilastroni 4, 25125 - Brescia, Italy Tel: +39 030 3501 361, Fax: +39 02 700 435 727 www.centroAlzheimer.it From minc-users@bic.mni.mcgill.ca Tue May 3 06:03:05 2005 From: minc-users@bic.mni.mcgill.ca (Michel Audette) Date: Tue May 3 05:03:05 2005 Subject: [MINC-users] z-y-x to x-y-z via mincreshape, orientation not preserved... Message-ID: <002e01c54fd2$e0afef00$3a751d96@surgsim2> Hi everyone, I am working with minc volumes that are originally z-y-x, but some of the processing software that I use, developped by one of my collaborators (Maxime Descoteaux), seems to depend on x-y-z orientation (or do "better" with thus oriented data), so I use "mincreshape -dimorder xspace, yspace, zspace +xdirection +ydirection +zdirection" to produce x-y-z volumes. Later on, I would like to use both the input and this processed volume, and to this effect, I resample my own input to also be x-y-z. However, I find that some of the volumes that I create in my own processing, using copy_volume_definition from my x-y-z input, does not preserve this x-y-z input and reverts to z-y-x. Is there a way of copying or opening that preserves x-y-z orientation input, even if the very original patient data was z-y-x? Just wondering... Michel Michel Audette, Ph.D., Research Fellow, Surgical Simulation, Surgical Assist Technology Group, AIST, Namiki 1-2, Tsukuba, Japan, 305-8564. -------------------------------------------------------- "If you think you can do it, you're right. If you think you can't do it, you're still right." - Henry Ford From Andrew Janke Tue May 3 06:21:05 2005 From: Andrew Janke (Andrew Janke) Date: Tue May 3 05:21:05 2005 Subject: [MINC-users] Windows-UNIX problems with Display In-Reply-To: <200505030705.j4375uPf6543174@shadow.bic.mni.mcgill.ca> References: <200505030705.j4375uPf6543174@shadow.bic.mni.mcgill.ca> Message-ID: On 5/3/05, Anna Caroli wrote: > I have labels drawn using Display for Windows, and I need to use them on a > SUN machine, because I need to process them with other MNI tools (labels are > drawn over linearly registered images, and I need to apply a back > transformation to have labels in the native space). > > The problem is I can't open with SUN and SGI machines labels.tag drawn using > Display for Windows... What I get is the following error message: > input_tag_points(): invalid header in file. > (I have the same problem the other way around: when I have a label drawn > over a UNIX machine, saved as filename.tag and I try to open it over a > Windows machine, I can't) Anna, At a quick guess without seeing the files I can only guess that you are having byte swapping problems. Byte swapping should not ever be a problem for MINC files but may be the case for .tag files. To check this byte swap the files on the SUN/IRIX machine before reading ie: dd if=tagfile.tag of=tagfile-bs.tag conv=swab -- Andrew Janke (a.janke@gmail.com || www.cmr.uq.edu.au/~rotor) Australia->Brisbane H: +61 7 3390 6332 || M: +61 4 2138 8581 From Andrew Janke Tue May 3 10:06:04 2005 From: Andrew Janke (Andrew Janke) Date: Tue May 3 09:06:04 2005 Subject: [MINC-users] z-y-x to x-y-z via mincreshape, orientation not preserved... In-Reply-To: <002e01c54fd2$e0afef00$3a751d96@surgsim2> References: <002e01c54fd2$e0afef00$3a751d96@surgsim2> Message-ID: On 5/3/05, Michel Audette wrote: > I am working with minc volumes that are originally z-y-x, but some of the > processing software that I use, developped by one of my collaborators > (Maxime Descoteaux), seems to depend on x-y-z orientation (or do "better" > with thus oriented data), so I use "mincreshape -dimorder xspace, yspace, > zspace +xdirection +ydirection +zdirection" to produce x-y-z volumes. If you really do want to flip all directions to positive remember to also use -dimsize xspace=-1 or the likes for the image directions (see man mincreshape for an explanation) > Later on, I would like to use both the input and this processed volume, and > to this effect, I resample my own input to also be x-y-z. However, I find > that some of the volumes that I create in my own processing, using > copy_volume_definition from my x-y-z input, does not preserve this x-y-z > input and reverts to z-y-x. Is there a way of copying or opening that > preserves x-y-z orientation input, even if the very original patient data > was z-y-x? copy_volume_definition will just copy attributes, not the dimension order, that depends on the inital volume creation IIRC. The easiest way I can think of to do this is just use open_volume on the volume you want to copy and then use output_modified_volume with the new filename. Somewhat dodgy I guess but it should copy all the associated data... a > Just wondering... > > Michel > > Michel Audette, Ph.D., > Research Fellow, Surgical Simulation, > Surgical Assist Technology Group, > AIST, > Namiki 1-2, > Tsukuba, Japan, > 305-8564. > -------------------------------------------------------- > "If you think you can do it, you're right. > If you think you can't do it, you're still right." > - Henry Ford > > _______________________________________________ > MINC-users@bic.mni.mcgill.ca > http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users > -- Andrew Janke (a.janke@gmail.com || www.cmr.uq.edu.au/~rotor) Australia->Brisbane H: +61 7 3390 6332 || M: +61 4 2138 8581 From minc-users@bic.mni.mcgill.ca Tue May 3 12:42:04 2005 From: minc-users@bic.mni.mcgill.ca (Robert VINCENT) Date: Tue May 3 11:42:04 2005 Subject: [MINC-users] Windows-UNIX problems with Display In-Reply-To: <200505030705.j4375uPf6543174@shadow.bic.mni.mcgill.ca> Message-ID: Hi Anna, I'm pretty certain this has to do with the differing newline conventions in text files in Windows and UNIX. On Windows and MS-DOS, there is typically an extra character placed at the end of each text file line. The library is not written to properly discard this extra character when it is encountered in a tag point file. So neither program is happy with the files produced by the other. There are a couple of workarounds for this that I can implement. It will be relatively easy to fix Register or Display for UNIX variants, but fixing the Windows version may take somewhat longer. In the meantime, if I am correct, it should be fairly easy to convert the files you import from Windows to UNIX using the "tr" command: tr -d "\r" < file-win.tag > file-unix.tag To go the other way, try this: perl -ne '$_ =~ s/\n/\r\n/; print $_;' < file-unix.tag > file-win.tag I know this isn't ideal but it should work. -bert On Tue, 3 May 2005, Anna Caroli wrote: > Hi all, > I have labels drawn using Display for Windows, and I need to use them on a > SUN machine, because I need to process them with other MNI tools (labels are > drawn over linearly registered images, and I need to apply a back > transformation to have labels in the native space). > > The problem is I can't open with SUN and SGI machines labels.tag drawn using > Display for Windows... What I get is the following error message: > input_tag_points(): invalid header in file. > (I have the same problem the other way around: when I have a label drawn > over a UNIX machine, saved as filename.tag and I try to open it over a > Windows machine, I can't) > > Using Display for Windows I also tried to open the label filename.tag and > save it as filename.mnc to export it to the SUN (I don't have problems in > importing/exporting and opening files.mnc): what I get from the correct > file.tag is a file where all the drawn structures are labelled as "structure > 1", irrespectively... > > Has anybody of you had the same problems? Can you please help me in > overcoming them? > Thanks a lot for your help and have a nice day! > > Anna > > Laboratory of Epidemiology, Neuroimaging & Telemedicine IRCCS San Giovanni > di Dio FBF > The National Center for Research and Care of Alzheimer's Disease > via Pilastroni 4, 25125 - Brescia, Italy > Tel: +39 030 3501 361, Fax: +39 02 700 435 727 www.centroAlzheimer.it > > _______________________________________________ > MINC-users@bic.mni.mcgill.ca > http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users > From minc-users@bic.mni.mcgill.ca Tue May 3 14:42:04 2005 From: minc-users@bic.mni.mcgill.ca (Joongnam Yang) Date: Tue May 3 13:42:04 2005 Subject: [MINC-users] Windows-UNIX problems with Display Message-ID: If the problem is due to "newline conventions in text files", the following two commands easily get rid of the unnecessay characters, dos2unix (when a Windows file was imported to Unix) unix2dos (the other way around). Nam. >>> bert@bic.mni.mcgill.ca 05/03/05 11:41 AM >>> Hi Anna, I'm pretty certain this has to do with the differing newline conventions in text files in Windows and UNIX. On Windows and MS-DOS, there is typically an extra character placed at the end of each text file line. The library is not written to properly discard this extra character when it is encountered in a tag point file. So neither program is happy with the files produced by the other. There are a couple of workarounds for this that I can implement. It will be relatively easy to fix Register or Display for UNIX variants, but fixing the Windows version may take somewhat longer. In the meantime, if I am correct, it should be fairly easy to convert the files you import from Windows to UNIX using the "tr" command: tr -d "\r" < file-win.tag > file-unix.tag To go the other way, try this: perl -ne '$_ =~ s/\n/\r\n/; print $_;' < file-unix.tag > file-win.tag I know this isn't ideal but it should work. -bert On Tue, 3 May 2005, Anna Caroli wrote: > Hi all, > I have labels drawn using Display for Windows, and I need to use them on a > SUN machine, because I need to process them with other MNI tools (labels are > drawn over linearly registered images, and I need to apply a back > transformation to have labels in the native space). > > The problem is I can't open with SUN and SGI machines labels.tag drawn using > Display for Windows... What I get is the following error message: > input_tag_points(): invalid header in file. > (I have the same problem the other way around: when I have a label drawn > over a UNIX machine, saved as filename.tag and I try to open it over a > Windows machine, I can't) > > Using Display for Windows I also tried to open the label filename.tag and > save it as filename.mnc to export it to the SUN (I don't have problems in > importing/exporting and opening files.mnc): what I get from the correct > file.tag is a file where all the drawn structures are labelled as "structure > 1", irrespectively... > > Has anybody of you had the same problems? Can you please help me in > overcoming them? > Thanks a lot for your help and have a nice day! > > Anna > > Laboratory of Epidemiology, Neuroimaging & Telemedicine IRCCS San Giovanni > di Dio FBF > The National Center for Research and Care of Alzheimer's Disease > via Pilastroni 4, 25125 - Brescia, Italy > Tel: +39 030 3501 361, Fax: +39 02 700 435 727 www.centroAlzheimer.it > > _______________________________________________ > MINC-users@bic.mni.mcgill.ca > http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users > _______________________________________________ MINC-users@bic.mni.mcgill.ca http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users From minc-users@bic.mni.mcgill.ca Tue May 3 23:44:07 2005 From: minc-users@bic.mni.mcgill.ca (Shahin Zangenehpour) Date: Tue May 3 22:44:07 2005 Subject: [MINC-users] (no subject) Message-ID: <9DBD01A8-DF6B-4F22-94D0-4BD869E84B92@bic.mni.mcgill.ca> Dear all, I am a recent post-doctoral fellow in Dr Zatorre's lab. By convention, I am a Mac user and I have enjoyed using the MINC tools and Dot on my PowerBook G4 over the past couple of months since my arrival at Robert's lab. As of last weekend, I have been having trouble running Dot on my machine and I suspect it has to do with upgrading my operating system from OS X Panther (10.3) to Tiger (10.4). I keep getting the following message after trying to launch Dot from an X-term subsequent to logging in: shadow 101% dot fork process 6472031 shadow 102% X Error of failed request: BadAtom (invalid Atom parameter) Major opcode of failed request: 18 (X_ChangeProperty) Atom id in failed request: 0xb5 Serial number of failed request: 12 Current serial number in output stream: 15 Could anyone help me get back on track? Many thanks, SZ From minc-users@bic.mni.mcgill.ca Wed May 4 04:25:05 2005 From: minc-users@bic.mni.mcgill.ca (Anna Caroli) Date: Wed May 4 03:25:05 2005 Subject: R: [MINC-users] Windows-UNIX problems with Display In-Reply-To: Message-ID: <200505040724.j447OdPf6837963@shadow.bic.mni.mcgill.ca> Thanks Bert! Anna > -----Messaggio originale----- > Da: minc-users-admin@bic.mni.mcgill.ca > [mailto:minc-users-admin@bic.mni.mcgill.ca] Per conto di > Robert VINCENT > Inviato: martedì 3 maggio 2005 17.42 > A: MINC list > Oggetto: Re: [MINC-users] Windows-UNIX problems with Display > > Hi Anna, > > I'm pretty certain this has to do with the differing newline > conventions in text files in Windows and UNIX. On Windows > and MS-DOS, there is typically an extra character placed at > the end of each text file line. > The library is not written to properly discard this extra > character when it is encountered in a tag point file. So > neither program is happy with the files produced by the other. > > There are a couple of workarounds for this that I can > implement. It will be relatively easy to fix Register or > Display for UNIX variants, but fixing the Windows version may > take somewhat longer. > > In the meantime, if I am correct, it should be fairly easy to > convert the files you import from Windows to UNIX using the > "tr" command: > > tr -d "\r" < file-win.tag > file-unix.tag > > To go the other way, try this: > > perl -ne '$_ =~ s/\n/\r\n/; print $_;' < file-unix.tag > file-win.tag > > I know this isn't ideal but it should work. > > -bert > > On Tue, 3 May 2005, Anna Caroli wrote: > > > Hi all, > > I have labels drawn using Display for Windows, and I need > to use them > > on a SUN machine, because I need to process them with other > MNI tools > > (labels are drawn over linearly registered images, and I > need to apply > > a back transformation to have labels in the native space). > > > > The problem is I can't open with SUN and SGI machines > labels.tag drawn > > using Display for Windows... What I get is the following > error message: > > input_tag_points(): invalid header in file. > > (I have the same problem the other way around: when I have a label > > drawn over a UNIX machine, saved as filename.tag and I try > to open it > > over a Windows machine, I can't) > > > > Using Display for Windows I also tried to open the label > filename.tag > > and save it as filename.mnc to export it to the SUN (I don't have > > problems in importing/exporting and opening files.mnc): what I get > > from the correct file.tag is a file where all the drawn > structures are > > labelled as "structure 1", irrespectively... > > > > Has anybody of you had the same problems? Can you please help me in > > overcoming them? > > Thanks a lot for your help and have a nice day! > > > > Anna > > > > Laboratory of Epidemiology, Neuroimaging & Telemedicine IRCCS San > > Giovanni di Dio FBF The National Center for Research and Care of > > Alzheimer's Disease via Pilastroni 4, 25125 - Brescia, Italy > > Tel: +39 030 3501 361, Fax: +39 02 700 435 727 > www.centroAlzheimer.it > > > > _______________________________________________ > > MINC-users@bic.mni.mcgill.ca > > http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users > > > > > > _______________________________________________ > MINC-users@bic.mni.mcgill.ca > http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users > From Andrew Janke Wed May 4 12:28:04 2005 From: Andrew Janke (Andrew Janke) Date: Wed May 4 11:28:04 2005 Subject: [MINC-users] Labels interpolation with Display In-Reply-To: <200505030756.j437uePf6585436@shadow.bic.mni.mcgill.ca> References: <200505030756.j437uePf6585436@shadow.bic.mni.mcgill.ca> Message-ID: On 5/3/05, Anna Caroli wrote: > Just a quick question: is it possible to interpolate labels drawn with > Display before calculating the volumes? > I know it's possible to interpolate the input image.mnc over which to draw > labels; is it posssible to interpolate also the drawn label to better fit > the structure? Presuming your label is a .mnc file and not a .tag file, you could easily do this using mincresample to whatever resolution you desire. But I don't think this is the question you are asking... If you increase the sampling of your anatomical file before tracing, the output resolution from Display of your label should always match the input file. -- Andrew Janke (a.janke@gmail.com || www.cmr.uq.edu.au/~rotor) Australia->Brisbane H: +61 7 3390 6332 || M: +61 4 2138 8581 From Andrew Janke Wed May 4 12:32:04 2005 From: Andrew Janke (Andrew Janke) Date: Wed May 4 11:32:04 2005 Subject: [MINC-users] (no subject) In-Reply-To: <9DBD01A8-DF6B-4F22-94D0-4BD869E84B92@bic.mni.mcgill.ca> References: <9DBD01A8-DF6B-4F22-94D0-4BD869E84B92@bic.mni.mcgill.ca> Message-ID: On 5/4/05, Shahin Zangenehpour wrote: > I have been having trouble running Dot on my machine and I suspect it has to do with > upgrading my operating system from OS X Panther (10.3) to Tiger > (10.4). I keep getting the following message after trying to launch > Dot from an X-term subsequent to logging in: > > shadow 101% dot > fork process 6472031 > shadow 102% X Error of failed request: BadAtom (invalid Atom parameter) > Major opcode of failed request: 18 (X_ChangeProperty) > Atom id in failed request: 0xb5 > Serial number of failed request: 12 > Current serial number in output stream: 15 > > Could anyone help me get back on track? At a wild guess this looks suspiciously like a Tcl/Tk error, it rings some bells anyhow. Perhaps go hunting for a new version of tcl/tk for Tiger? tcl/tk was onthe apple developer website IIRC. -- Andrew Janke (a.janke@gmail.com || www.cmr.uq.edu.au/~rotor) Australia->Brisbane H: +61 7 3390 6332 || M: +61 4 2138 8581 From minc-users@bic.mni.mcgill.ca Wed May 4 12:43:04 2005 From: minc-users@bic.mni.mcgill.ca (Jason Lerch) Date: Wed May 4 11:43:04 2005 Subject: [MINC-users] (no subject) In-Reply-To: References: <9DBD01A8-DF6B-4F22-94D0-4BD869E84B92@bic.mni.mcgill.ca> Message-ID: On May 4, 2005, at 11:31 AM, Andrew Janke wrote: > On 5/4/05, Shahin Zangenehpour wrote: >> I have been having trouble running Dot on my machine and I suspect it >> has to do with >> upgrading my operating system from OS X Panther (10.3) to Tiger >> (10.4). I keep getting the following message after trying to launch >> Dot from an X-term subsequent to logging in: >> >> shadow 101% dot >> fork process 6472031 >> shadow 102% X Error of failed request: BadAtom (invalid Atom >> parameter) >> Major opcode of failed request: 18 (X_ChangeProperty) >> Atom id in failed request: 0xb5 >> Serial number of failed request: 12 >> Current serial number in output stream: 15 >> >> Could anyone help me get back on track? > > At a wild guess this looks suspiciously like a Tcl/Tk error, it rings > some bells anyhow. > > Perhaps go hunting for a new version of tcl/tk for Tiger? tcl/tk was > onthe apple developer website IIRC. I doubt that tcl/tk is at fault since the job is launched from shadow. It thus has to do with how the X forwarding is being handled. Jason > > > -- > Andrew Janke (a.janke@gmail.com || www.cmr.uq.edu.au/~rotor) > Australia->Brisbane H: +61 7 3390 6332 || M: +61 4 2138 8581 > > _______________________________________________ > MINC-users@bic.mni.mcgill.ca > http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users From minc-users@bic.mni.mcgill.ca Wed May 4 12:53:04 2005 From: minc-users@bic.mni.mcgill.ca (Jean-Francois Malouin) Date: Wed May 4 11:53:04 2005 Subject: [MINC-users] (no subject) In-Reply-To: References: <9DBD01A8-DF6B-4F22-94D0-4BD869E84B92@bic.mni.mcgill.ca> Message-ID: <20050504155222.GA6677525@shadow.bic.mni.mcgill.ca> * Jason Lerch [20050504 11:49]: > > On May 4, 2005, at 11:31 AM, Andrew Janke wrote: > > >On 5/4/05, Shahin Zangenehpour wrote: > >>I have been having trouble running Dot on my machine and I suspect it > >>has to do with > >>upgrading my operating system from OS X Panther (10.3) to Tiger > >>(10.4). I keep getting the following message after trying to launch > >>Dot from an X-term subsequent to logging in: > >> > >>shadow 101% dot > >>fork process 6472031 > >>shadow 102% X Error of failed request: BadAtom (invalid Atom > >>parameter) > >> Major opcode of failed request: 18 (X_ChangeProperty) > >> Atom id in failed request: 0xb5 > >> Serial number of failed request: 12 > >> Current serial number in output stream: 15 > >> > >>Could anyone help me get back on track? > > > >At a wild guess this looks suspiciously like a Tcl/Tk error, it rings > >some bells anyhow. > > > >Perhaps go hunting for a new version of tcl/tk for Tiger? tcl/tk was > >onthe apple developer website IIRC. > > I doubt that tcl/tk is at fault since the job is launched from shadow. > It thus has to do with how the X forwarding is being handled. That would be my guess too. It's the X server/display on the Mac. Which X have you installed? jf > > Jason > > > > > > >-- > >Andrew Janke (a.janke@gmail.com || www.cmr.uq.edu.au/~rotor) > >Australia->Brisbane H: +61 7 3390 6332 || M: +61 4 2138 8581 > > > >_______________________________________________ > >MINC-users@bic.mni.mcgill.ca > >http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users > > > _______________________________________________ > MINC-users@bic.mni.mcgill.ca > http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users -- <° >< From minc-users@bic.mni.mcgill.ca Wed May 4 13:06:04 2005 From: minc-users@bic.mni.mcgill.ca (Jason Lerch) Date: Wed May 4 12:06:04 2005 Subject: [MINC-users] Fwd: Question re Tiger Message-ID: --Apple-Mail-3--523464641 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset=ISO-8859-1; delsp=yes; format=flowed It was X forwarding. See below. Cheers, Jason Begin forwarded message: > From: Shahin Zangenehpour > Date: May 4, 2005 12:00:25 PM EDT > To: Jason Lerch > Subject: Re: Question re Tiger > > Hi Jason, > Thanks for getting back to me. I log onto shadow with ssh and X =20 > forwarding enabled. For example: > shahin@wireless152108:~$ ssh -X shahin@shadow.bic.mni.mcgill.ca > > I have managed to find a solution to the problem, based on some =20 > postings I found on the internet. Basically, by changing the -X to -Y =20= > the problem disappears. That is: > shahin@wireless152108:~$ ssh -Y shahin@shadow.bic.mni.mcgill.ca > > I have no real grasp of what is happening, but so long as it works I =20= > am happy and continue to be productive. > Best, > SZ > > > On 4-May-05, at 8:23 AM, Jason Lerch wrote: > > Greetings, > > I've not had a chance to try 10.4 yet, so can only vaguely guess about = =20 > what might be going wrong. How do you log into shadow? With ssh and X =20= > forwarding enabled? Or do you handle your X forwarding differently. > > Jason > > On May 3, 2005, at 1:00 AM, Shahin Zangenehpour wrote: > > > Dear Jason, > I am a recent post-doctoral fellow in Robert Zatorre's lab. By =20 > convention, I am a Mac user and I have enjoyed using the MINC tools =20= > and Dot on my PowerBook G4 over the past couple of months since my =20 > arrival at Robert's lab. As of last weekend, I have been having =20 > trouble running Dot on my machine and I suspect it has to do with =20 > upgrading my operating system from OS X Panther (10.3) to Tiger =20 > (10.4). I keep getting the following message after trying to launch =20= > Dot from an X-term subsequent to logging in: > > shadow 101% dot > fork process 6472031 > shadow 102% X Error of failed request:=A0 BadAtom (invalid Atom =20 > parameter) > =A0 Major opcode of failed request:=A0 18 (X_ChangeProperty) > =A0 Atom id in failed request:=A0 0xb5 > =A0 Serial number of failed request:=A0 12 > =A0 Current serial number in output stream:=A0 15 > > > Is there any way you may be able to help me get back on track? > Many thanks, > SZ > = _______________________________________________________________________=20= > _______________ > Shahin Zangenehpour, PhD=A0 =A0 Postdoctoral Fellow =A0 |=A0 P.=A0 =A0 = 514=A0 398=A0 =20 > 1717=A0 |=A0 F.=A0 =A0 514=A0 398=A0 1338 > Cognitive Neuroscience Unit=A0 |=A0 MNI=A0 |=A0 3801 University = Street=A0 |=A0 =20 > Montr=E9al =A0 QC =A0 Canada =A0 H3A 2B4 > > > --Apple-Mail-3--523464641 Content-Transfer-Encoding: quoted-printable Content-Type: text/enriched; charset=ISO-8859-1 It was X forwarding. See below. Cheers, Jason Begin forwarded message: 0000,0000,0000From: Shahin Zangenehpour < 0000,0000,0000Date: May 4, 2005 12:00:25 PM EDT 0000,0000,0000To: Jason Lerch < 0000,0000,0000Subject: Re: Question re Tiger Hi Jason, Thanks for getting back to me. I log onto shadow with ssh and X forwarding enabled. For example: shahin@wireless152108:~$ ssh -X = 0000,0000,EEEEshahin@shadow.bic.mni.mcgill.ca I have managed to find a solution to the problem, based on some postings I found on the internet. Basically, by changing the -X to -Y the problem disappears. That is: shahin@wireless152108:~$ ssh -Y = 0000,0000,EEEEshahin@shadow.bic.mni.mcgill.ca I have no real grasp of what is happening, but so long as it works I am happy and continue to be productive. Best, SZ On 4-May-05, at 8:23 AM, Jason Lerch wrote: OptimaGreetings, OptimaI've not had a chance to try 10.4 yet, so can only vaguely guess about what might be going wrong. How do you log into shadow? With ssh and X forwarding enabled? Or do you handle your X forwarding differently. OptimaJason OptimaOn May 3, 2005, at 1:00 AM, Shahin Zangenehpour wrote: OptimaDear Jason, OptimaI am a recent post-doctoral fellow in Robert Zatorre's lab. By convention, I am a Mac user and I have enjoyed using the MINC tools and Dot on my PowerBook G4 over the past couple of months since my arrival at Robert's lab. As of last weekend, I have been having trouble running Dot on my machine and I suspect it has to do with upgrading my operating system from OS X Panther (10.3) to Tiger (10.4). I keep getting the following message after trying to launch Dot from an X-term subsequent to logging in: Optimashadow 101% dot Optimafork process 6472031 Optimashadow 102% X Error of failed request:=A0 BadAtom (invalid Atom parameter) Optima=A0 Major opcode of failed request:=A0 = 18 (X_ChangeProperty) Optima=A0 Atom id in failed request:=A0 = 0xb5 Optima=A0 Serial number of failed request:=A0 = 12 Optima=A0 Current serial number in output stream:=A0 15 OptimaIs there any way you may be able to help me get back on track? OptimaMany thanks, OptimaSZ = Optima0000,0000,8080_____= __________________________________________________________________________= _______ = Optima0000,0000,8080Shahin = ZangenehpourOptima<= color>0000,0000,8080, PhD=A0 =A0 Postdoctoral Fellow =A0 B3B3,B3B3,B3B3|=A0 P.=A0 =A0 0000,0000,8080514=A0 398=A0 1717=A0 B3B3,B3B3,B3B3|=A0 F.=A0 =A0 0000,0000,8080514=A0 398=A0 = 1338 = Optima0000,0000,8080Cogni= tive Neuroscience Unit=A0 = B3B3,B3B3,B3B3|0000,00= 00,8080=A0 MNI=A0 = B3B3,B3B3,B3B3|0000,00= 00,8080=A0 3801 University Street=A0 = B3B3,B3B3,B3B3|0000,00= 00,8080=A0 Montr=E9al =A0 QC =A0 Canada =A0 H3A 2B4 = --Apple-Mail-3--523464641-- From minc-users@bic.mni.mcgill.ca Wed May 4 14:47:05 2005 From: minc-users@bic.mni.mcgill.ca (minc-users@bic.mni.mcgill.ca) Date: Wed May 4 13:47:05 2005 Subject: [MINC-users] hires vs. lowres registration In-Reply-To: <200505041601.j44G1DPf6893533@shadow.bic.mni.mcgill.ca> Message-ID: Hello, I've encountered a curious situation with image registration and would like to know if anyone has any insight or comments: I'm aligning two images that represent MRI scans of the same object (i.e. the human brain) but in two different modalities (i.e. T2 and T1). The registration process is essentially mritoself which calculates the transform function needed to register one image to the other, followed by mincresample, which applies the transform and places one image in the same space as the other. Initially we decided to use the hi-res T1 images and to align them to the lower resolution T2 images, with the rationale that providing more information about the image should lead to a better fit. But we have since found that using the lo-res T1 images actually produces better registration (i.e. alignment) results. Does this make sense and what could be happening here within the mritoself and mincresample algorithms to cause this? The lo-res T1 image resolution matches the T2 image resolution, but theoretically the hi-res T1 image should be less prone to partial volume etc. errors. Any comments or ideas would be appreciated! Ernest Lo NeuroRX Research Montreal, Canada From Andrew Janke Wed May 4 21:25:04 2005 From: Andrew Janke (Andrew Janke) Date: Wed May 4 20:25:04 2005 Subject: [MINC-users] hires vs. lowres registration In-Reply-To: References: <200505041601.j44G1DPf6893533@shadow.bic.mni.mcgill.ca> Message-ID: On 5/5/05, elo@neurorx.com wrote: > > I've encountered a curious situation with image registration and would > like to know if anyone has any insight or comments: I'm sure Louis is the one who can provide the most insight here, but nevertheless here's a few things to think about. > I'm aligning two images that represent MRI scans of the same object (i.e. > the human brain) but in two different modalities (i.e. T2 and T1). The > registration process is essentially mritoself which calculates the > transform function needed to register one image to the other, followed by > mincresample, which applies the transform and places one image in the same > space as the other. > > Initially we decided to use the hi-res T1 images and to align them to the > lower resolution T2 images, with the rationale that providing more > information about the image should lead to a better fit. But we have > since found that using the lo-res T1 images actually produces better > registration (i.e. alignment) results. Does this make sense and what > could be happening here within the mritoself and mincresample algorithms > to cause this? First, this difference has (well should have!) nothing to do with mincresample so let's concentrate on minctracc/mritoself. >From what I can gather you are registering the T1 to the T2 as such: minctracc/mritoself T1.mnc T2.mnc T1-T2.xfm Also given that you are using differing modalities, I presume you are using mutual information: minctracc -mi T1.mnc T2.mnc T1-T2.xfm Given this information a few things could be happening: * mritoself as part of its own internal fitting process does do some blurring. however the size of this blurring is somewhat dependant on the input resolution. As such it may just happen that the higher FWHM blurring caused by initially downsampling the data just happens to be getting over a local minima in the high res version. * Given that you are using mutual information and that the input T1 may be somewhat noisy, the mutual information "image" that is calculated internally as part of the fit may be somewhat spurious. Down sampling the data first is effectively blurring it. Thus your input T1 is more smooth in the low-res version. * Lots of other things.... You can easily test the second by blurring the input high-res data before fitting at the same level as the downsampling would achieve. -- Andrew Janke (a.janke@gmail.com || www.cmr.uq.edu.au/~rotor) Australia->Brisbane H: +61 7 3390 6332 || M: +61 4 2138 8581 From minc-users@bic.mni.mcgill.ca Thu May 5 11:47:04 2005 From: minc-users@bic.mni.mcgill.ca (minc-users@bic.mni.mcgill.ca) Date: Thu May 5 10:47:04 2005 Subject: [MINC-users] Re: hires vs. lowres registration In-Reply-To: Message-ID: Hi Andrew, Thanks very much for your comments. We're continuing to look at this issue. >> First, this difference has (well should have!) nothing to do with >> mincresample so let's concentrate on minctracc/mritoself. I'm not sure if mincresample does not have an effect. For example, the process of reducing the resolution and changing coordinates to match the T1 image to the T2 space takes place when this tool is run. So perhaps mritoself calculates the correct transformation (.xfm) but could error or noise be then introduced afterward during the resampling? >> * Given that you are using mutual information and that the input T1 >> may be somewhat noisy, the mutual information "image" that is >> calculated internally as part of the fit may be somewhat spurious. >> Down sampling the data first is effectively blurring it. Thus your >> input T1 is more smooth in the low-res version. >> >> * Lots of other things.... >> >> You can easily test the second by blurring the input high-res data >> before fitting at the same level as the downsampling would achieve. I plan to go ahead and try this - to manually reduce the hi-res data, and to see if this also improves the registration/alignment quality. Recently I tested the registration (i.e. mritoself and mincresample) on a series of scans with varying quality from different scanners. In all cases registering using the lo-res T1 image worked extremely well for some reason. So if we can be sure that this is happening for a valid reason, it would represent a good technique or strategy for us (and perhaps in general), during the image registration stage of analysis. All comments or thoughts are greatly appreciated!! Ernest Lo NeuroRX Research Montreal, Canada On Wed, 4 May 2005 elo@neurorx.com wrote: > Hello, > > I've encountered a curious situation with image registration and would > like to know if anyone has any insight or comments: > > I'm aligning two images that represent MRI scans of the same object (i.e. > the human brain) but in two different modalities (i.e. T2 and T1). The > registration process is essentially mritoself which calculates the > transform function needed to register one image to the other, followed by > mincresample, which applies the transform and places one image in the same > space as the other. > > Initially we decided to use the hi-res T1 images and to align them to the > lower resolution T2 images, with the rationale that providing more > information about the image should lead to a better fit. But we have > since found that using the lo-res T1 images actually produces better > registration (i.e. alignment) results. Does this make sense and what > could be happening here within the mritoself and mincresample algorithms > to cause this? > > The lo-res T1 image resolution matches the T2 image resolution, but > theoretically the hi-res T1 image should be less prone to partial volume > etc. errors. > > Any comments or ideas would be appreciated! > > > Ernest Lo > NeuroRX Research > Montreal, Canada > > > From Andrew Janke Fri May 6 12:38:03 2005 From: Andrew Janke (Andrew Janke) Date: Fri May 6 11:38:03 2005 Subject: [MINC-users] Re: hires vs. lowres registration In-Reply-To: References: Message-ID: On 5/6/05, elo@neurorx.com wrote: > > I'm not sure if mincresample does not have an effect. For example, the > process of reducing the resolution and changing coordinates to match the > T1 image to the T2 space takes place when this tool is run. So perhaps > mritoself calculates the correct transformation (.xfm) but could error or > noise be then introduced afterward during the resampling? Yes, resampling errors will be introduced when mincresample is run. But these same resampling errors are going to occur regardless of which transformation you apply. In any case, these errors are nothing to do with registration error. It is these registration errors/differences that I believe you are interested in. > Recently I tested the registration (i.e. mritoself and mincresample) on a > series of scans with varying quality from different scanners. In all > cases registering using the lo-res T1 image worked extremely well for some > reason. So if we can be sure that this is happening for a valid reason, > it would represent a good technique or strategy for us (and perhaps in > general), during the image registration stage of analysis. Again, I suspect this is due to the inherrent blurring that is performed as part of downsampling data. What you have described here (low res fitting etc) is one of the fundamental ideas of a hierachial fitting strategy -- fitting large features first followed by high-res features. Have a read of Louis Collins' (the author of minctracc) paper in JCAT regarding this. (man minctracc will spit up the reference). [2] Collins DL, Neelin P, Peters TM, and Evans AC, "Automatic 3-D intersubject registration of MR volumetric data in standardized Talairach space", J. Comput. Assis. Tomogr, 18(2):192-205. -- Andrew Janke (a.janke@gmail.com || www.cmr.uq.edu.au/~rotor) Australia->Brisbane H: +61 7 3390 6332 || M: +61 4 2138 8581 From minc-users@bic.mni.mcgill.ca Mon May 9 12:08:04 2005 From: minc-users@bic.mni.mcgill.ca (minc-users@bic.mni.mcgill.ca) Date: Mon May 9 11:08:04 2005 Subject: [MINC-users] Problem with nu_estimate_np_and_em Message-ID: <20050509110709.5t95epdzmass08gc@web.mail.umich.edu> when I run the following command nu_estimate_np_and_em -parzen -log -sharpen 0.15 0.01 -iterations 50 -stop 0.01 -shrink 4 -auto_mask -nonotify -b_spline 1 -distance 200 -quiet -execute -clobber -nokeeptmp manconv_02_t1_final_tmp.mnc /manconv_02_t1_final.imp I get this message: Transforming slices:..............................................Done Error: Composite mask volume is empty. Check for one of the following: 1) the user supplied mask is empty (all zeros) 2) the values within the region of interest are entirely less than one 3) the user supplied background threshold is too high 4) the intersection of the various masking options yields an empty mask I tried changing some of the options in my command but with no luck. I also noticed someone had a similar problem: http://www.bic.mni.mcgill.ca/pipermail/minc-users/2002-September/000034.html Suggestions? Thanks in advance, Haytham From minc-users@bic.mni.mcgill.ca Tue May 10 04:06:04 2005 From: minc-users@bic.mni.mcgill.ca (Michel Audette) Date: Tue May 10 03:06:04 2005 Subject: [MINC-users] specifying different output ranges under voxel_loop Message-ID: <000f01c55542$bd977770$3a751d96@surgsim2> Hi everyone, I am working with voxel_loop, and I would like to find out if different output files can be specified as having different voxel-ranges. As far as I can see, set_loop_datatype() seems to set one kind of output range. Is there another way? Cheers, Michel Michel Audette, Ph.D., Research Fellow, Surgical Simulation, Surgical Assist Technology Group, AIST, Namiki 1-2, Tsukuba, Japan, 305-8564. -------------------------------------------------------- "If you think you can do it, you're right. If you think you can't do it, you're still right." - Henry Ford From Andrew Janke Tue May 10 11:05:03 2005 From: Andrew Janke (Andrew Janke) Date: Tue May 10 10:05:03 2005 Subject: [MINC-users] specifying different output ranges under voxel_loop In-Reply-To: <000f01c55542$bd977770$3a751d96@surgsim2> References: <000f01c55542$bd977770$3a751d96@surgsim2> Message-ID: On 5/10/05, Michel Audette wrote: > > I am working with voxel_loop, and I would like to find out if different > output files can be specified as having different voxel-ranges. As far as I > can see, set_loop_datatype() seems to set one kind of output range. Is there > another way? Are you trying to set the voxel or real range? you say voxel range but I suspect you mean real-range. The only way that I know of is to use set_loop_datatype(), however this will not rescale the images, only set the valid range. Might I ask why? -- Andrew Janke (a.janke@gmail.com || www.cmr.uq.edu.au/~rotor) Australia->Brisbane H: +61 7 3390 6332 || M: +61 4 2138 8581 From Andrew Janke Tue May 10 11:10:03 2005 From: Andrew Janke (Andrew Janke) Date: Tue May 10 10:10:03 2005 Subject: [MINC-users] Problem with nu_estimate_np_and_em In-Reply-To: <20050509110709.5t95epdzmass08gc@web.mail.umich.edu> References: <20050509110709.5t95epdzmass08gc@web.mail.umich.edu> Message-ID: On 5/10/05, haytham@umich.edu wrote: > when I run the following command > > nu_estimate_np_and_em -parzen -log -sharpen 0.15 0.01 -iterations 50 -stop 0.01 > -shrink 4 -auto_mask -nonotify -b_spline 1 -distance 200 -quiet -execute > -clobber -nokeeptmp manconv_02_t1_final_tmp.mnc /manconv_02_t1_final.imp > > I get this message: > > Transforming slices:..............................................Done > Error: Composite mask volume is empty. Check for one of the following: > 1) the user supplied mask is empty (all zeros) > 2) the values within the region of interest are entirely less than one > 3) the user supplied background threshold is too high > 4) the intersection of the various masking options yields an empty mask I have seen this when the generated (automatic in your case -- -automask) mask is indeed all 0 only. Perhaps remove the nokeeptmp and have a look at the intermediate mask file. You may have to provide a manually selected mask threshold. -- Andrew Janke (a.janke@gmail.com || www.cmr.uq.edu.au/~rotor) Australia->Brisbane H: +61 7 3390 6332 || M: +61 4 2138 8581 From minc-users@bic.mni.mcgill.ca Tue May 10 11:17:05 2005 From: minc-users@bic.mni.mcgill.ca (Christophe Grova) Date: Tue May 10 10:17:05 2005 Subject: [MINC-users] Question regarding fmr_preprocess References: <1060291078.4670.13.camel@tosa-nb.nmr.mgh.harvard.edu> <003801c35da4$48497820$d03384c8@PUCRS.BR> <1060350006.18989.5.camel@dennis.bic.mni.mcgill.ca> Message-ID: <001a01c5557b$7b8d66d0$c636d884@cgrova> Hello everybody, in what units (voxel or mm) are given the motion parameters provided by the log files of fmr_preprocess ? Thanks in advance, Christophe From Andrew Janke Tue May 10 11:30:04 2005 From: Andrew Janke (Andrew Janke) Date: Tue May 10 10:30:04 2005 Subject: [MINC-users] Question regarding fmr_preprocess In-Reply-To: <001a01c5557b$7b8d66d0$c636d884@cgrova> References: <1060291078.4670.13.camel@tosa-nb.nmr.mgh.harvard.edu> <003801c35da4$48497820$d03384c8@PUCRS.BR> <1060350006.18989.5.camel@dennis.bic.mni.mcgill.ca> <001a01c5557b$7b8d66d0$c636d884@cgrova> Message-ID: On 5/11/05 Christophe Grova wrote: > in what units (voxel or mm) are given the motion parameters provided by the > log files of fmr_preprocess ? Morning Christophe, >From a quick read of the script, and assuming you mean the bit where it is spitting out information about "Max X translation" and the likes, the answer is real (mm). -- Andrew Janke (a.janke@gmail.com || www.cmr.uq.edu.au/~rotor) Australia->Brisbane H: +61 7 3390 6332 || M: +61 4 2138 8581 From minc-users@bic.mni.mcgill.ca Wed May 11 03:56:04 2005 From: minc-users@bic.mni.mcgill.ca (Michel Audette) Date: Wed May 11 02:56:04 2005 Subject: [MINC-users] specifying different output ranges under voxel_loop References: <000f01c55542$bd977770$3a751d96@surgsim2> Message-ID: <005a01c5560a$7314c590$3a751d96@surgsim2> Hi Andrew, thanks for the reply, and you're right, I meant real range. Now, does set_loop_datatype set real or voxel range? Because what I would like to do, is set different real ranges for various outputs of voxel_loop(), because one output has is a measure of sheetlike structure that should be between 0 and 1, and the other outputs are not restricted to this range. Or is does voxel_loop take care of real ranges automatically? Just wondering... Michel Michel Audette, Ph.D., Research Fellow, Surgical Simulation, Surgical Assist Technology Group, AIST, Namiki 1-2, Tsukuba, Japan, 305-8564. -------------------------------------------------------- "If you think you can do it, you're right. If you think you can't do it, you're still right." - Henry Ford ----- Original Message ----- From: "Andrew Janke" To: Sent: Tuesday, May 10, 2005 11:04 PM Subject: Re: [MINC-users] specifying different output ranges under voxel_loop > On 5/10/05, Michel Audette wrote: > > > > I am working with voxel_loop, and I would like to find out if different > > output files can be specified as having different voxel-ranges. As far as I > > can see, set_loop_datatype() seems to set one kind of output range. Is there > > another way? > > Are you trying to set the voxel or real range? you say voxel range > but I suspect you mean real-range. > > The only way that I know of is to use set_loop_datatype(), however > this will not rescale the images, only set the valid range. > > Might I ask why? > > -- > Andrew Janke (a.janke@gmail.com || www.cmr.uq.edu.au/~rotor) > Australia->Brisbane H: +61 7 3390 6332 || M: +61 4 2138 8581 > > _______________________________________________ > MINC-users@bic.mni.mcgill.ca > http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users From minc-users@bic.mni.mcgill.ca Wed May 11 12:28:04 2005 From: minc-users@bic.mni.mcgill.ca (Anna Caroli) Date: Wed May 11 11:28:04 2005 Subject: [MINC-users] tool to reorient images along AC-PC? Message-ID: <200505111527.j4BFR7Pf7740529@shadow.bic.mni.mcgill.ca> Dear MINC user, Does anybody know if there is any software tool to reorient images along AC-PC? I need to draw some structures over native images.mnc (in the native space, and not in the talairach space), before linear registration, and I would like to reorient them along AC-PC to draw labels more easily. Thanks for your help! Anna Laboratory of Epidemiology, Neuroimaging & Telemedicine IRCCS San Giovanni di Dio FBF The National Center for Research and Care of Alzheimer's Disease via Pilastroni 4, 25125 - Brescia, Italy Tel: +39 030 3501 361, Fax: +39 02 700 435 727 www.centroAlzheimer.it From minc-users@bic.mni.mcgill.ca Tue May 17 08:42:04 2005 From: minc-users@bic.mni.mcgill.ca (Oliver Winz) Date: Tue May 17 07:42:04 2005 Subject: [MINC-users] align PET to MRT using minctracc (stripscalp missing) Message-ID: <4289D88F.7050902@fz-juelich.de> Hi everyone, I want to align a individual PET to an individual MRT using minctracc. I've got a perl script "mritopet" where i read, that it is beneficial to autocrop the MRT and after that to strip the scalp befor using minctracc. To strip the scalp the script uses "stripscalp", but i don't have this program. Where can i get it? or what other tools can i used to strip the scalp. Greeting, Oliver -- Oliver Winz, M.Sc. AG Molekulares Neuroimaging Institut für Medizin (IME) Forschungszentrum Jülich D-52425 Jülich Tel.: (02461) 61-6493 Internet: http://www.fz-juelich.de/ime/mni Ort: Gebäude 15.9, Raum 3010 From Andrew Janke Tue May 17 10:08:04 2005 From: Andrew Janke (Andrew Janke) Date: Tue May 17 09:08:04 2005 Subject: [MINC-users] tool to reorient images along AC-PC? In-Reply-To: <200505111527.j4BFR7Pf7740529@shadow.bic.mni.mcgill.ca> References: <200505111527.j4BFR7Pf7740529@shadow.bic.mni.mcgill.ca> Message-ID: On 5/12/05, Anna Caroli wrote: > Does anybody know if there is any software tool to reorient images along > AC-PC? > I need to draw some structures over native images.mnc (in the native space, > and not in the talairach space), before linear registration, and I would > like to reorient them along AC-PC to draw labels more easily. Well there is no tool to do specifically this that I know of, however all you need do is first register all your images to talairach space (using mritotal) and then from here just apply a standard (precomputed) xfm that aligns a talairach image along its AC-PC line. I suspect it will take a few guesses to get the re-orientation xfm right but once you have it you can just resample your native images with the concatenation of their talairach xfm and the reorientation xfm. -- Andrew Janke (a.janke@gmail.com || www.cmr.uq.edu.au/~rotor) Australia->Brisbane H: +61 7 3390 6332 || M: +61 4 2138 8581 From minc-users@bic.mni.mcgill.ca Fri May 20 05:33:05 2005 From: minc-users@bic.mni.mcgill.ca (Michel Audette) Date: Fri May 20 04:33:05 2005 Subject: [MINC-users] easiest way to do large, isotropic erosion and dilation with mincmorph? Message-ID: <002901c55d2a$a7907900$3a751d96@surgsim2> Hello, I want to run mincmorph to produce a morphological opening with a large isotropic structuring element, and I would like to keep the results of the erosion and dilation separately, so in fact I want to run separate erode and dilate operations with this element and keep each result. What is the simplest way? Must I put together a file that expresses this element, or can I run this with a simple parameter that indicates the size of element? If I absolutely need the file, are there some that I can get ready-made? Cheers, Michel Michel Audette, Ph.D., Research Fellow, Surgical Simulation, Surgical Assist Technology Group, AIST, Namiki 1-2, Tsukuba, Japan, 305-8564. -------------------------------------------------------- "If you think you can do it, you're right. If you think you can't do it, you're still right." - Henry Ford ----- Original Message ----- From: "Andrew Janke" To: Sent: Tuesday, May 10, 2005 11:04 PM Subject: Re: [MINC-users] specifying different output ranges under voxel_loop > On 5/10/05, Michel Audette wrote: > > > > I am working with voxel_loop, and I would like to find out if different > > output files can be specified as having different voxel-ranges. As far as I > > can see, set_loop_datatype() seems to set one kind of output range. Is there > > another way? > > Are you trying to set the voxel or real range? you say voxel range > but I suspect you mean real-range. > > The only way that I know of is to use set_loop_datatype(), however > this will not rescale the images, only set the valid range. > > Might I ask why? > > -- > Andrew Janke (a.janke@gmail.com || www.cmr.uq.edu.au/~rotor) > Australia->Brisbane H: +61 7 3390 6332 || M: +61 4 2138 8581 > > _______________________________________________ > MINC-users@bic.mni.mcgill.ca > http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users From minc-users@bic.mni.mcgill.ca Fri May 20 05:54:04 2005 From: minc-users@bic.mni.mcgill.ca (Michel Audette) Date: Fri May 20 04:54:04 2005 Subject: Fw: [MINC-users] easiest way to do large, isotropic erosion and dilation with mincmorph? Message-ID: <004901c55d2d$a2025320$3a751d96@surgsim2> Hi everyone, please disregard my previous message: I've found the documentation on the use of the "-successive" option for mincmorph. Cheers, Michel Michel Audette, Ph.D., Research Fellow, Surgical Simulation, Surgical Assist Technology Group, AIST, Namiki 1-2, Tsukuba, Japan, 305-8564. -------------------------------------------------------- "If you think you can do it, you're right. If you think you can't do it, you're still right." - Henry Ford ----- Original Message ----- From: "Michel Audette" To: Sent: Friday, May 20, 2005 7:57 PM Subject: [MINC-users] easiest way to do large, isotropic erosion and dilation with mincmorph? > Hello, > > I want to run mincmorph to produce a morphological opening with a large > isotropic structuring element, and I would like to keep the results of the > erosion and dilation separately, so in fact I want to run separate erode and > dilate operations with this element and keep each result. What is the > simplest way? Must I put together a file that expresses this element, or can > I run this with a simple parameter that indicates the size of element? If I > absolutely need the file, are there some that I can get ready-made? > > Cheers, > > Michel > > Michel Audette, Ph.D., > Research Fellow, Surgical Simulation, > Surgical Assist Technology Group, > AIST, > Namiki 1-2, > Tsukuba, Japan, > 305-8564. > -------------------------------------------------------- > "If you think you can do it, you're right. > If you think you can't do it, you're still right." > - Henry Ford > ----- Original Message ----- > From: "Andrew Janke" > To: > Sent: Tuesday, May 10, 2005 11:04 PM > Subject: Re: [MINC-users] specifying different output ranges under > voxel_loop > > > > On 5/10/05, Michel Audette wrote: > > > > > > I am working with voxel_loop, and I would like to find out if different > > > output files can be specified as having different voxel-ranges. As far > as I > > > can see, set_loop_datatype() seems to set one kind of output range. Is > there > > > another way? > > > > Are you trying to set the voxel or real range? you say voxel range > > but I suspect you mean real-range. > > > > The only way that I know of is to use set_loop_datatype(), however > > this will not rescale the images, only set the valid range. > > > > Might I ask why? > > > > -- > > Andrew Janke (a.janke@gmail.com || www.cmr.uq.edu.au/~rotor) > > Australia->Brisbane H: +61 7 3390 6332 || M: +61 4 2138 8581 > > > > _______________________________________________ > > MINC-users@bic.mni.mcgill.ca > > http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users > > _______________________________________________ > MINC-users@bic.mni.mcgill.ca > http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users From Andrew Janke Fri May 20 09:25:04 2005 From: Andrew Janke (Andrew Janke) Date: Fri May 20 08:25:04 2005 Subject: Fw: [MINC-users] easiest way to do large, isotropic erosion and dilation with mincmorph? In-Reply-To: <004901c55d2d$a2025320$3a751d96@surgsim2> References: <004901c55d2d$a2025320$3a751d96@surgsim2> Message-ID: On 5/20/05, Michel Audette wrote: > > please disregard my previous message: I've found the documentation on the > use of the "-successive" option for mincmorph. :) In response to the first part of the question about structuring elements, yes, you must make a file to express the element. Look in the .tar.gz version of mincmorph for some examples. -- Andrew Janke (a.janke@gmail.com || www.cmr.uq.edu.au/~rotor) Australia->Brisbane H: +61 7 3390 6332 || M: +61 4 2138 8581 From minc-users@bic.mni.mcgill.ca Wed May 25 13:43:04 2005 From: minc-users@bic.mni.mcgill.ca (Robert VINCENT) Date: Wed May 25 12:43:04 2005 Subject: [MINC-users] MINC 1.4 Message-ID: Hi everyone, So I've officially "released" MINC 1.4. The source code is available for download here: http://www.bic.mni.mcgill.ca/software/distribution/packages/minc-1.4.tar.gz The new features are as follows: * Converters for DICOM, ECAT, microPET, and NIfTI-1 (Analyze 7.5) format have been included in the package. See the man pages for details. * The code has been tested to build on Windows using Microsoft C/C++. See the file "Makefile.msvc-win32" * Optional DICOM (ACR/NEMA) tools and library are now part of the distribution. * The "simplified" MINC programming interface is now available. See minc_simple.h for details. This interface is still somewhat preliminary, we would welcome feedback and suggestions for how it can be improved. * For software developers, there is a new macro, VIO_PREFIX_NAMES, which can be defined to force the volume_io headers to avoid defining commonly used types and symbols by prepending the string "VIO_" to all such types and constants: #define VIO_PREFIX_NAMES 1 #include * New, corrected bimodal threshold algorithms and calculations are now available in mincstats. * Other minor bug fixes. See the ChangeLog for more details. As always, please direct any questions or problems to me! -bert