[Intl_DxMedPhys] Older Siemens EI question

[Douglas Pfeiffer]

Thanks, Ryan. Yes, it’s the physical EXI that I’m looking at. I’m fascinated by the requirement to set the cal factor to one, as that seems to defeat the purpose of a cal factor! To Mike’s comment, I think the reason that they don’t match is because of the assumptions made for clinical EXI, that there’s inherently non-uniformity due to the expected anatomy. I don’t know how the calculation is done, but I think that’s where the delta comes from. This incident is from my consulting practice. My hospital is almost all Siemens R/F, and I’ve not run into this at the hospital, which is part of why I’m so confused at this response. When the EXI is out of calibration at the hospital, Siemens fixes it with no issue. I’m thinking that in this case is not OEM service and they just don’t understand in any detail. I have a feeling that I, too, will end up just throwing up my hands. Sigh…

Best regards,
Doug



Douglas Pfeiffer, MS, DABR FACR, FAAPM (he/him)
Medical Physicist, Radiation Safety Officer
Boulder Community Health
xraydoug at me.com
303.415.7515

+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-

The opinions expressed in this message are the product of the gray and white matter loitering in my cranium. I speak for myself and no one else, unless I say otherwise.

> On Oct 8, 2026, at 6:24 AM, Ryan Fisher <rfisher1 at metrohealth.org> wrote:
> 
> Siemens is weird on this, which makes things messy from our side. They kinda sorta follow the IEC definition and now can report both a “physical EXI” and a “clinical EXI” on their modern systems. I believe the older systems just report what is now the Physical EXI as “EXI”.
>  
> Definitions attached, but the physical EXI is the average of the central area of the detector, regardless of what’s there, and with no histogram analysis/pixel rejection. The Clinical EXI considers the pixel values and is expecting some kind of clinical anatomy and does histogram analysis to try to determine the relevant area, and then is reporting the average of only those pixels. For IEC style verification, the physical EXI is what you want, as the clinical one gets weird with flat fields sometimes and can pick a weird clump of pixels somewhere skewing results. Naturally there’s no indication of what pixels are going into the calculation. To make things even more fun, on the modern systems there are calibration factors under the hood to adjust the accuracy of both, but all the Siemens PM procedures require the FSE to set the cal factor for one of them back to 1.
>  
> It’s been awhile since I dealt with this, but remember basically just throwing up my hands about it because even if we adjusted the cal factor, the next year you’d come back and it would be set back to 1. Good luck. 
>  
>  
> Ryan 
>  
>  
> From: Intl_dxmedphys_wd_osu_list <intl_dxmedphys_wd_osu_list-bounces at lists.osu.edu <mailto:intl_dxmedphys_wd_osu_list-bounces at lists.osu.edu>> On Behalf Of Douglas Pfeiffer via Intl_dxmedphys_wd_osu_list
> Sent: Wednesday, October 7, 2026 6:20 PM
> To: DxMedPhys List <intl_dxmedphys_wd_osu_list at lists.osu.edu <mailto:intl_dxmedphys_wd_osu_list at lists.osu.edu>>
> Subject: [Intl_DxMedPhys] Older Siemens EI question
>  
> STOP!
> THIS IS AN EXTERNAL EMAIL
> This email was originated outside of the MetroHealth System
> Take a 3-second pause and ask yourself the following questions:
> 1. Who is sending me this email? Do I know who this is?
> 2. Am I expecting this email? Is this out of the ordinary?
> 3. Why is there a link in the email? Do I trust this link?
> 
> THINK
> Before You 
> Click the Link
> Dear all, I recently tested a 2009 Siemens Axiom Luminos TF and found that the reported EI (or EXI in Siemens parlance) was about 50% of from what it should be. The service engineer replied thusly: In Siemens we calculate the EXI as the arithmetic
> Dear all, 
>  
> I recently tested a 2009 Siemens Axiom Luminos TF and found that the reported EI (or EXI in Siemens parlance) was about 50% of from what it should be. The service engineer replied thusly:
> In Siemens we calculate the EXI as the arithmetic mean of gray value of a central area (refer to the example picture below) under consideration of the actual collimation.
> EXI value is not calculated for dull detector image.
>  
> For the Cannon detector in Luminos TF system, there is no adjustment for the EXI value. The maximum we can do is to calibrate the detector.
>  
> The IEC standard [62494-1 (2008)] is very specifically calibrated at RQA5 in a uniform field. Has anyone else been told anything like what this person is saying?
>  
> Best regards,
> Doug
>  
>  
>  
> Douglas Pfeiffer, MS, DABR FACR, FAAPM (he/him)
> Medical Physicist, Radiation Safety Officer
> Boulder Community Health
> xraydoug at me.com <mailto:xraydoug at me.com>
> 303.415.7515
>  
> +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-
>  
> The opinions expressed in this message are the product of the gray and white matter loitering in my cranium. I speak for myself and no one else, unless I say otherwise.
>  
> 
> MetroHealth’s Mission: Leading the way to a healthier you and a healthier community through service, teaching, discovery, and teamwork. This email and all attachments that may have been included are intended only for the use of the party to whom/which the email is addressed and may contain information that is privileged, confidential, or exempt from disclosure under applicable law. If you are not the addressee or the employee or agent of the intended recipient, you are hereby notified that you are strictly prohibited from printing, storing, disseminating, distributing, or copying this communication.Contact us <https://urldefense.com/v3/__https://fs26.formsite.com/8dBVlg/sjvmjw2v0c/index.html__;!!KGKeukY!3Md5eCYfVpzqql2RkXbi4ZjYw-qmDJvZYGx9ZU11P82gs14vTPapYsRP5afqr8im6kgKuc0f5gro4-5l0lxpx8CCd6uF6p5vS_DC$ ><Siemens EXI Definitions.docx>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osu.edu/pipermail/intl_dxmedphys_wd_osu_list/attachments/20261008/741e856e/attachment.html>