Server error 500 with GE Pristina 3D Tomo RDSR

Julio Almansa - JALmansa

unread,

Jul 8, 2024, 1:30:47 PM7/8/24

to OpenREM

Hi,

I'm using 0.10.0 version (Ubuntu 18.04.5 LTS) and I have the following error.

When working with a GE Pristina 3D mammography unit OpenRem import and render correctly the summary of studies.

But when I try to go into the detail of a study I have a "Server Error 500". In Debug mode seems to indicate an error in the number of filters for each exposure.

Is there way to fix this ?

Thanks

JALmansa

PD: Associated with this error, I suppose, is that when I try to export to csv, the export process freezes without finishing.

Ed McDonagh

unread,

Jul 10, 2024, 3:37:35 AM7/10/24

to OpenREM

Hi Julio, thanks for raising this.

I've just looked in the 1.0 beta release code, and I think the same problem would occur. Is there any possibility of sending me a QA RDSR for this system that falls over in the same way, so that I can test against it?

If so, you can send it to me directly rather than to the group. An anonymised one can work too, but a QA one is always safer and avoids the anonymisation tool changing the RDSR too much.

Kind regards

Ed

--
You received this message because you are subscribed to the Google Groups "OpenREM" group.
To unsubscribe from this group and stop receiving emails from it, send an email to openrem+u...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/openrem/93629dc1-5f91-4086-9a11-35918de78c07n%40googlegroups.com.

Julio Almansa - JALmansa

unread,

Jul 10, 2024, 7:06:10 PM7/10/24

to OpenREM

Thanks for assistance Ed,

I'll send you some QA RDSR examples as soon as I find them.

JALmansa

Ed McDonagh

unread,

Jul 14, 2024, 2:37:24 PM7/14/24

to OpenREM

Just to note that I have created the following issue to track this issue: https://bitbucket.org/openrem/openrem/issues/1029/mammo-rdsr-import-kv-ma-pulse-width-filter

To view this discussion on the web, visit https://groups.google.com/d/msgid/openrem/934277f0-5bca-4a5e-8afb-ed75d92f8f5bn%40googlegroups.com.

Ed McDonagh

unread,

Jul 15, 2024, 1:20:42 PM7/15/24

to OpenREM

Hi Julio, Hi Marco

Please see the latest update on issue 1029 as it affects both of you.

If we are receiving 10 or more pulses per tomo exposure, each with kV, mA and filter information, how do we expect that to be displayed and exported? Pulse width isn't currently on the mammo detail page - should it be? (exposure time is).

We need to establish how to export them too - what should be in the csv or Excel file?

For fluoro, I think we ended up with a mean of the non-zero values, though I think we did have median values at one point. I think in this example, a mean would not be representative of these 10 exposures, but a median would be. I don't have a variety of these RDSRs to review to see how the numbers usually vary.

Your thoughts, or anyone elses, would be very welcome here or on the bitbucket page linked to at the top (issue 1029)

Kind regards

Ed

Ed McDonagh

unread,

Jul 16, 2024, 5:32:35 PM7/16/24

to OpenREM

The issue has been updated with some more discussion and ideas - I'd appreciate it if anyone can review issue #1029 on bitbucket and comment there, or comment in this thread via email.