Effect of image offset on DIR
Jeremy Lim
Sharp, Gregory C.,PHD
Hi Jeremy,
I would think so too. The only thing that gives me pause is that most of the calculations are done in single precision floating point, so there will be slight differences due to rounding.
I'll also point out that the calculations have a non-deterministic order of operations between threads, which can also exhibit rounding differences. You can test single threading by setting the environment OMP_NUM_THREADS to 1.
Did you find there to be differences? How large are the differences?
Greg
From: plast...@googlegroups.com <plast...@googlegroups.com>
On Behalf Of Jeremy Lim
Sent: Tuesday, October 11, 2022 1:08 AM
To: Plastimatch <plast...@googlegroups.com>
Subject: [Plastimatch] Effect of image offset on DIR
External Email - Use Caution
--
You received this message because you are subscribed to the Google Groups "Plastimatch" group.
To unsubscribe from this group and stop receiving emails from it, send an email to
plastimatch...@googlegroups.com.
To view this discussion on the web visit
https://groups.google.com/d/msgid/plastimatch/3df6d69c-ffd3-47a3-9d82-3c1cd6390f2dn%40googlegroups.com.
The information in this e-mail is intended only for the person to whom it is addressed. If you believe this e-mail was sent to you in error and the e-mail contains patient information, please contact the Mass General Brigham Compliance HelpLine at https://www.massgeneralbrigham.org/complianceline .
Please note that this e-mail is not secure (encrypted). If you do not wish to continue communication over unencrypted e-mail, please notify the sender of this message immediately. Continuing to send or respond to e-mail after receiving this message means you understand and accept this risk and wish to continue to communicate over unencrypted e-mail.
Jeremy Lim
Sharp, Gregory C.,PHD
Very interesting data. Thanks!
Qualitatively, the data are consistent with numeric rounding. But the absolute differences are larger than I would expect.
Unfortunately I don't have a short term solution. Let me know if you discover more. I will work on a test plan.
Greg
________________________________________
From: plast...@googlegroups.com <plast...@googlegroups.com> on behalf of Jeremy Lim <junlim...@gmail.com>
Sent: Tuesday, October 11, 2022 9:08 PM
To: Plastimatch
Subject: Re: [Plastimatch] Effect of image offset on DIR
External Email - Use Caution
Thanks for your response. Yes, judging by the MSE after DIR, I found differences in the registration. However, they are not very large.
Here are the results for two different patients (4 stage DIR):
Curious about the pattern seen in Patient1, I did more registrations with fixed intervals for the offsets changes (but with only 1 stage DIR). There was this oscillating pattern and I tried looking at finer and finer intervals to see if I could find the boundary but was not successful.
Kind regards,
Jeremy
On Wednesday, October 12, 2022 at 1:21:03 AM UTC+11 Sharp, Gregory C.,PHD wrote:
Hi Jeremy,
I would think so too. The only thing that gives me pause is that most of the calculations are done in single precision floating point, so there will be slight differences due to rounding.
I'll also point out that the calculations have a non-deterministic order of operations between threads, which can also exhibit rounding differences. You can test single threading by setting the environment OMP_NUM_THREADS to 1.
Did you find there to be differences? How large are the differences?
Greg
From: plast...@googlegroups.com <plast...@googlegroups.com> On Behalf Of Jeremy Lim
Sent: Tuesday, October 11, 2022 1:08 AM
To: Plastimatch <plast...@googlegroups.com>
Subject: [Plastimatch] Effect of image offset on DIR
External Email - Use Caution
Hi,
I was wondering how the value of the image offset is changing the result of the DIR.
I'm registering two images which have the same offset. What I have done is modify the offsets of the two images in the header by the same amount and nothing else, i.e. simply moving the images together in space. Shouldn't the placement of grid points on the image and hence the DIR results be exactly the same?
Kind regards,
Jeremy
--
You received this message because you are subscribed to the Google Groups "Plastimatch" group.
To unsubscribe from this group and stop receiving emails from it, send an email to plastimatch...@googlegroups.com.
The information in this e-mail is intended only for the person to whom it is addressed. If you believe this e-mail was sent to you in error and the e-mail contains patient information, please contact the Mass General Brigham Compliance HelpLine at https://www.massgeneralbrigham.org/complianceline<https://secure-web.cisco.com/1IBpYVQiP1WIOWUXUoy3c2RuAvu9fznH-3ANT8gmGnn30mKLoCEqtqsq95GjBPL2GQMqlUJa76ORk2ckIw_68CkKpY74ZdjEP663GsZvKwYDyeWqErPyf1wyX1pCpfS6VyIQWNRVFzHS22iVmBNQ3Z5xKt8Uznj6ftctG9x-_oMnZByHRL1Gr_5kpCVMkVM0gQnBdd9V1fxE5WmomyDP261jEiEDKEdpLb1C3MBOvyRxy_okg745F6I5CYei567djfxJDMXTx0kuy8i9un91DblxzTmgJP-GF3cmGLN3fwJRCU-XREdXM5tzqoEJxqno63Yb3sLgkdwKKZEq_Tl90gw/https%3A%2F%2Fwww.massgeneralbrigham.org%2Fcomplianceline> .
Please note that this e-mail is not secure (encrypted). If you do not wish to continue communication over unencrypted e-mail, please notify the sender of this message immediately. Continuing to send or respond to e-mail after receiving this message means you understand and accept this risk and wish to continue to communicate over unencrypted e-mail.
You received this message because you are subscribed to the Google Groups "Plastimatch" group.
To view this discussion on the web visit https://groups.google.com/d/msgid/plastimatch/9b357a3d-25e0-46d0-97bb-6ef164c0456fn%40googlegroups.com<https://secure-web.cisco.com/14PiveOvX8k8_RgusIZucVKWxsxA9UjcpnbV-u3QoAde3VtRmvFzvkkrvoq4fxNvfk9A0uWnfjVNcxGUR6fb0PTefEIXcCFL7gMHSwEvJjStGuYkqyeLfH6zcoaIVXnvUH-M7IDw4W9tUOJKe9L7HQoj7R_a6dR36wXW5GfNlzui6_OhLzXrPexZfJSPCuGwL1d17OehICKZRY_txA3VcNNXPcB2agLUvJmFXGhFDyfDcT2ZxF0uoQ6IrZT8Ko-B4B1V7xk4M7YVkjtxnn8GI5Bv8at8xbkZyoE2dDE804NE9bE7a9INkSpVKbjO73QN_JKXHgw_wCQhQMd4bjC6D8w/https%3A%2F%2Fgroups.google.com%2Fd%2Fmsgid%2Fplastimatch%2F9b357a3d-25e0-46d0-97bb-6ef164c0456fn%2540googlegroups.com%3Futm_medium%3Demail%26utm_source%3Dfooter>.
The information in this e-mail is intended only for the person to whom it is addressed. If you believe this e-mail was sent to you in error and the e-mail contains patient information, please contact the Mass General Brigham Compliance HelpLine at https://www.massgeneralbrigham.org/complianceline <https://www.massgeneralbrigham.org/complianceline> .
