Acquisition time and DAP values and units for fluoroscopy in OpenREM

[Bolstad, Kirsten Hansine Helene Nygaard]

Dear Ed and OpenREM team,

 

First of all, thank you for creating the OpenREM program. My colleagues Njål Brekke and Daniel Aadnevik installed OpenREM in our hospital (Haukeland University Hospital, Bergen, Norway) about one and a half year ago, and it works better and is more useful than we imagined. At the moment we have 64.313 studies in our database (39229 CT, 12000 Fluoroscopy and 7929 Radiography). I know Njål has been in touch with you on several occasions regarding some OpenREM features.

 

I am currently looking into dose values, fluoro and acquisition times for our fluoroscopy Siemens Artis Q system, and it appears to me that the units and values does not correspond to each other in the exported files.

 

Some examples:

 

Acquisition time:

Exported XLSX file column U: "A Fluoro time total: 35”

Exported CSV file " Total fluoro time (ms): 35.00”

Copy-paste of the exposure incidents matrix from OpenREM in the web-browser, then sum of the "Exposure time (ms)" column: 3461,8

It looks to me like the values in the exported file is not actually [ms].

 

 

DAP total

Exported CSV file, then imported to XLSX:

DAP total (Gy.m2): 0,00199307

Fluoro DAP total (Gy.m2): 0,11069

Acquisition DAP total (Gy.m2): 1224

 

Shouldn’t the "Fluoro DAP total" + "Acquisition DAP total" = "DAP Total" ? it looks to me like 3 different units are used in the exported CSV.

 

Copy-paste of the exposure incidents matrix from OpenREM web-browser into XLSX, then sum of the "DAP (cGy.cm2)" column: 1992.95, so the “DAP total (Gy.m2)” in the exported file seems to be correct.

But neither the "Fluoro DAP total (Gy.m2)" nor "Acquisition DAP total (Gy.m2)" in the exported file has a value that corresponds to [Gy.m2].

 

I have currently not looked into other vendors.

 

Do you think this is a Siemens issue or an OpenREM issue?

 

Hope to hear back from you.

 

PS: We met in Bergen in October 2013 when I invited you to give a talk at a physicist course about optimization.

 

Kind regards

Kirsten Nygaard Bolstad
Medical physicist

Head of X-ray physics

Haukeland University Hospital

Helse Bergen HF
+47 55 97 60 78

kirsten...@helse-bergen.no
www.helse-bergen.no

 

 

 

 

[Ed McDonagh]

Hi Kirsten, it is nice to hear from you and really good to have the feedback :-)

Thank you also for bringing these discrepancies to my attention.

Regarding the time, you are right. The accumulated times in an RDSR are in seconds, the event times are in milliseconds. I didn't take this into account when labelling the column headings, so I will make that change for the next release (see ticket 573 here). 

Regarding the DAP, that is not so clear.

I have exported a study from my production server that has fluoro and acquisition events, and for me the fluoro DAP total and the acquisition DAP total add up to the DAP total, with each one being in Gy.m2. This also corresponds to the value on the web interface for the study in cGy.cm2, once you multiply it by 1 million.

Therefore I suspect something else is going on here. Are you able to compare 'in-room' values to RDSR values to OpenREM values to see where the difference comes in? If you have an example RDSR that is reporting strange DAP values in OpenREM I would like to take a look at it - QA files are usually ideal.

We have had a much more subtle issue on Siemens kit recently where the event DAP exceeds the Cumulative DAP, both on the screen in the room and in the exported RDSR. But this is of the order of a few percent - nothing like what you have shown above.

Let me know what you find or if you have anything for me to look at.

Kind regards

Ed

P.S. The development version has faster exports, with much nicer code structure, and the fluoro study detail page has a nice summary table of the cumulative DAP, reference point dose and times at the top - have a look at http://dev.openrem.org/openrem/rf/2/ for example :-)

[Ed McDonagh]

Hello again.

I have just committed the fix to the develop version, and it has reminded me of some of the other things to look out for.

Can I ask - in your example was the fluoroscopy continuous?

At summary level, the only time recorded is the irradiation duration. That is, the time someone's foot was on the peddle, which is usually longer than the sum of all the irradiation pulses.

At event level, there are different fields for irradiation duration (in s) and for exposure time (in ms). On my Siemens units, the exposure time is recorded at event level, and the irradiation duration is not. Which means that you would not expect the sum of the series event level exposure times (in ms) to be equal to the summary level irradiation duration (in s).

As well as correcting the units for the columns that were already there, I have added the irradiation duration column (it is populated by some other vendors), and the number of pulses too.

Kind regards

Ed