[Intl_DxMedPhys] MQSA - Manufacturer PM vs ACR DM QC manual

Tao, Ashley T attao at emplifyhealth.org
Tue Jun 2 07:15:38 EDT 2026 

yes, QC is and should be based on either the manufacturer QC manual or the ACR DM QC manual. In my opinion, the MQSA inspectors are misunderstanding and misinterpreting the regulations and the ACR DM QC manual. While the ACR DM QC manual covers QC for all manufacturers, it does not provide guidance or clarity when manufacturer PM specs are not met. The MP is qualified and should be able to determine whether or not a PM deficiency rises to the requirements of MQSA and the ACR DM QC manual. Our recent inspector disagreed and cited us but also acknowledged that she didn't fully understand what a PM was and why there was so much detail in the PM report. We have submitted a letter to the FDA with our justification for why they should consider withdrawing the citations and are awaiting their response.

Ashley


________________________________
From: Milman, Rebecca <REBECCA.MILMAN at CUANSCHUTZ.EDU>
Sent: Monday, June 1, 2026 3:54 PM
To: Tao, Ashley T <attao at emplifyhealth.org>; Megan Brown <megan at centennialmedphys.com>
Cc: Chikezie C. Onyema <onyema.cc at gmail.com>; DxMedPhys List <intl_dxmedphys_wd_osu_list at lists.osu.edu> <intl_dxmedphys_wd_osu_list at lists.osu.edu>
Subject: Re: [Intl_DxMedPhys] MQSA - Manufacturer PM vs ACR DM QC manual

CAUTION: This email originated from outside of Emplify Health. Do not click on any links or open any attachments unless you recognize the sender and know the contents are safe.
We will NEVER ask you to provide your user name or password in reply to an email.
Questions? Contact the Bellin Service Desk at x258600 / (920) 436-8600 or the Gundersen Help Desk at x53070 / (608) 775-3070.




Okay, still trying to make sense of all of this…

The Hologic QC manual references MQSA Alternative Standard 9<https://urldefense.com/v3/__https://www.fda.gov/radiation-emitting-products/regulations-mqsa/mqsa-alternative-standard-9-correction-period-when-components-selenia-full-field-digital-mammography__;!!IqJknsWg5-ALmw!SJvAusa0dASolA6d5QLzkwA_RKh3mwcCEDpbEhGHp4GQHYXum2sPFOPcPd_XwZ-Ysm6ijQhpdx8l72IHLFptrB5cSBZT5lnZ$>.

This states that “If any of the following quality control tests that evaluate the performance of the image acquisition components…produce results that fall outside the action limits as specified by the manufacturer, the source of the problem shall be identified and corrective action must be taken before any further examinations are performed.” The list of applicable QC tests includes Breast Entrance Exposure and Average Glandular Dose.

Under the QC tests that require corrective action within 30 days is Radiation Output Rate, HVL Measurement, and Automatic Exposure Control Function Performance.

It seems that this would apply to the “action limits as specified by the manufacturer” that are in the QC Manual, not some other value the manufacturer decides on. The criteria in the QC Manual are either sufficient for evaluating system performance or they aren’t. (As far as I can tell the internal performance requirements aren’t published anywhere, which makes the whole thing even more confusing.)

What am I missing here?



Rebecca.





From: Tao, Ashley T <attao at emplifyhealth.org>
Date: Monday, June 1, 2026 at 12:28 PM
To: Milman, Rebecca <REBECCA.MILMAN at cuanschutz.edu>; Megan Brown <megan at centennialmedphys.com>
Cc: Chikezie C. Onyema <onyema.cc at gmail.com>; DxMedPhys List
<intl_dxmedphys_wd_osu_list at lists.osu.edu>
Subject: Re: [Intl_DxMedPhys] MQSA - Manufacturer PM vs ACR DM QC manual

[External Email - Use Caution]
So there's two follow-up questions about AEC recalibration:

  1.
Hologic AEC performance fails - example below - fails @ 2cm.  Hologic's AEC performance metrics are different than the ACR DM QC manual.  my MEE SNR changed by 0.8%. One could argue that any recalibration affects dose and so would need on-site physics testing.
[cid:805cebcb-38c4-4463-9aa4-dcb8fb5cb328]
  2.
Hologic HVL fails - many of you probably aren't aware, but my biomed team showed me how they do the HVL cals. It prompts them to re-calibrate AEC. While AEC may have been within cal standards before the HVL cal, the service engineer is prompted to recal AEC  - which according to the ACR DM QC manual would require on-site physics testing.
[cid:37f4be00-634b-4ce9-928f-e53226f562b9]

I think part of the struggle is that our biomed team includes a lot of detail in their service reports (and this is something we are trying to clean up)- in some ways, more transparency and detail is better  - there have been times in the past when they didn't notify us of something that needed physics oversight/on-site testing.  At the same time, inspectors only want to see what is relevant to them and as soon as they see a fail, they automatically assume that it's a failure for MQSA.





________________________________
From: Milman, Rebecca <REBECCA.MILMAN at CUANSCHUTZ.EDU>
Sent: Monday, June 1, 2026 1:02 PM
To: Megan Brown <megan at centennialmedphys.com>; Tao, Ashley T <attao at emplifyhealth.org>
Cc: Chikezie C. Onyema <onyema.cc at gmail.com>; DxMedPhys List <intl_dxmedphys_wd_osu_list at lists.osu.edu> <intl_dxmedphys_wd_osu_list at lists.osu.edu>
Subject: Re: [Intl_DxMedPhys] MQSA - Manufacturer PM vs ACR DM QC manual

CAUTION: This email originated from outside of Emplify Health. Do not click on any links or open any attachments unless you recognize the sender and know the contents are safe.
We will NEVER ask you to provide your user name or password in reply to an email.
Questions? Contact the Bellin Service Desk at x258600 / (920) 436-8600 or the Gundersen Help Desk at x53070 / (608) 775-3070.




Thanks, Megan. That’s interesting.

I would think we could monitor some of these things by looking at the weekly QC phantom images. If the mR/mAs increases, then either the noise should decrease or the AEC-determined mAs should decrease.

However, I will wait for Doug Pfeiffer to point out why I’m wrong about that and for Dustin Gress to point out why that approach is inadequate for accreditation/regulatory purposes. 😉 (Please note that this last sentence is in jest and that I highly respect both of these physicists!!)

Rebecca.


From: Megan Brown <megan at centennialmedphys.com>
Date: Monday, June 1, 2026 at 11:50 AM
To: Tao, Ashley T <attao at emplifyhealth.org>
Cc: Chikezie C. Onyema <onyema.cc at gmail.com>; Milman, Rebecca <REBECCA.MILMAN at cuanschutz.edu>; DxMedPhys List
<intl_dxmedphys_wd_osu_list at lists.osu.edu>
Subject: Re: [Intl_DxMedPhys] MQSA - Manufacturer PM vs ACR DM QC manual

[External Email - Use Caution]
This came up for me recently - Hologic declared the dose was too high and they would have to do adjustments to the tube output. I was sent this if anyone is curious about the vendor spec:
[IMG_20260601_104856.heic]

I guess a question to ask is what is meant by "AEC calibration that effects dose" in the ACR QC Manual. Are we adjusting the AEC tables from standard to low dose or contrast screening? In that case yes I think physicist testing is justified. The minor adjustments that they have been making during PMs are meant to bring it back within their own tight specs for performance. That I don't know requires us to be there to test. The other question is if none of the tech or physicist QC tests fail, then how are we applying the corrective action timeframe for those tests? IMO if none of the QC tests fail, and no major parts are replaced then physicist oversight should be sufficient. The HVL failures which seem to occur at every PM in particular - I've been doing oversight for these adjustments and have not had an issue with MQSA.


Thanks,

Megan Brown, M.S., DABR
Medical Physicist
Centennial Medical Physics
megan at centennialmedphys.com<mailto:megan at centennialmedphys.com>
303.916.5305


[https://urldefense.com/v3/__https://outlook.office.com/mail/AAMkADQ0Njk5YTBlLThiZjQtNDc1ZC1iNjJjLTQwZTVlMTA4YjMyMQAuAAAAAABoBNNn36mdRZ9ap2iVUZnxAQAdO6M3kPxXToQtjzwOM1*2B5AAAEFwyKAAA*3D/id/AAkALgAAAAAAHYQDEapmEc2byACqAC*2FEWg0AHTujN5D8V06ELY88DjNfuQAHHG2AwQAA?nativeVersion=1.2026.520.400__;JSUl!!KGKeukY!yCAEaOoL-rhhKBmQRIjd9V5tPzaRYp2fYlwfGEdS--vT86wP4MHOdFcqOk3bBnKtRgo6O3NMl8kUSsdr2mZ9ymHXBebMF32LSsxLDX5HFg$ ]<https://urldefense.com/v3/__https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail__;!!IqJknsWg5-ALmw!Tg_VahfT9e113dvju1UKcMoNU1HX_LLp5uYuWkKsoZRwH1XaP0rCTkJvNLnJOE561wvM1d4UEXq7Gy_N9UxocAIeJoLM4DsD$>
Virus-free.www.avast.com<https://urldefense.com/v3/__https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail__;!!IqJknsWg5-ALmw!Tg_VahfT9e113dvju1UKcMoNU1HX_LLp5uYuWkKsoZRwH1XaP0rCTkJvNLnJOE561wvM1d4UEXq7Gy_N9UxocAIeJoLM4DsD$>

On Mon, Jun 1, 2026 at 11:25 AM Tao, Ashley T via Intl_dxmedphys_wd_osu_list <intl_dxmedphys_wd_osu_list at lists.osu.edu<mailto:intl_dxmedphys_wd_osu_list at lists.osu.edu>> wrote:
I agree with what you have described Chikezie, however, our MQSA inspectors are holding us to PM failures and citing us for it for not fixing HVL immediately before clinical use.

To complicate matters, when they recalibrate HVL, they technically have to recalibrate AEC. Per the ACR DM QC manual, if there is AEC recalibration that affects dose - technically any recalibration would affect dose and a physicist would need to do on-site testing post recalibration.


________________________________
From: Chikezie C. Onyema <onyema.cc at gmail.com<mailto:onyema.cc at gmail.com>>
Sent: Monday, June 1, 2026 12:14 PM
To: Milman, Rebecca <REBECCA.MILMAN at cuanschutz.edu<mailto:REBECCA.MILMAN at cuanschutz.edu>>
Cc: Tao, Ashley T <attao at emplifyhealth.org<mailto:attao at emplifyhealth.org>>; DxMedPhys List <intl_dxmedphys_wd_osu_list at lists.osu.edu<mailto:intl_dxmedphys_wd_osu_list at lists.osu.edu>> <intl_dxmedphys_wd_osu_list at lists.osu.edu<mailto:intl_dxmedphys_wd_osu_list at lists.osu.edu>>
Subject: Re: [Intl_DxMedPhys] MQSA - Manufacturer PM vs ACR DM QC manual

CAUTION: This email originated from outside of Emplify Health. Do not click on any links or open any attachments unless you recognize the sender and know the contents are safe.
We will NEVER ask you to provide your user name or password in reply to an email.
Questions? Contact the Bellin Service Desk at x258600 / (920) 436-8600 or the Gundersen Help Desk at x53070 / (608) 775-3070.




Hi,

My understanding is that the tolerance level set by Hologic and checked by FSE during PMs is tighter than the one specified in the QC manual. The image below shows a snapshot of a tube output measurement done by FSE during PM. An FSE typically uses a dosimeter (in our case, Raysafe X2) to measure the tube output and HVL at different kV settings, then compares them against set specification values. It is my understanding that if it is out of spec, the FSE can increase/decrease the X-ray tube output to meet the specification (and pass the tube output tests), which may require a total recalibration of the unit, including dose measurement. It is possible that some measurements may pass per the QC manual, whereas during PM the same measurement may be out of spec due to tighter tolerances.
[IMG_1368.jpeg]



Thanks
Chikezie Onyema, PhD, DABR
Diagnostic Medical Physicist
Department of Radiology
Marshfield Clinic | Sanford Health
Marshfield, WI 54449

[https://urldefense.com/v3/__https://outlook.office.com/mail/AAMkADQ0Njk5YTBlLThiZjQtNDc1ZC1iNjJjLTQwZTVlMTA4YjMyMQAuAAAAAABoBNNn36mdRZ9ap2iVUZnxAQAdO6M3kPxXToQtjzwOM1*2B5AAAEFwyKAAA*3D/id/AAkALgAAAAAAHYQDEapmEc2byACqAC*2FEWg0AHTujN5D8V06ELY88DjNfuQAHHG2AwQAA?nativeVersion=1.2026.520.400__;JSUl!!KGKeukY!yCAEaOoL-rhhKBmQRIjd9V5tPzaRYp2fYlwfGEdS--vT86wP4MHOdFcqOk3bBnKtRgo6O3NMl8kUSsdr2mZ9ymHXBebMF32LSsxLDX5HFg$ ]<https://urldefense.com/v3/__http://www.avg.com/email-signature?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail__;!!IqJknsWg5-ALmw!QfYFW_Ujhei4YmgkqNDpG2yZw0XUwcoHJzYZ9WeGy8m8PIQMqV3D8xZkBgoRbFuH6q5xRynObaqx0nczGw$>
Virus-free.www.avg.com<https://urldefense.com/v3/__http://www.avg.com/email-signature?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail__;!!IqJknsWg5-ALmw!QfYFW_Ujhei4YmgkqNDpG2yZw0XUwcoHJzYZ9WeGy8m8PIQMqV3D8xZkBgoRbFuH6q5xRynObaqx0nczGw$>

On Mon, Jun 1, 2026 at 11:21 AM Milman, Rebecca via Intl_dxmedphys_wd_osu_list <intl_dxmedphys_wd_osu_list at lists.osu.edu<mailto:intl_dxmedphys_wd_osu_list at lists.osu.edu>> wrote:
Update:

I just received a PM service report (from Hologic) that stated that, “System passed all tests…to the exception of tube output and dose for all techniques.”

This raises a couple of questions, and I’m interested in other people’s experiences.

As a physicist, I don’t really know what it means if the “tube output and dose for all techniques” doesn’t pass during a PM. Without more information - including their internal pass/fail criteria - I’m not sure how to interpret this service report.

The other issue is related to Ashley’s original post. If the manufacturer’s tolerance for HVL is 0.01 mm Al, then it is problematic when you’re measuring HVL with a device with a measurement error > 0.01 mm Al. My experience has been that Hologic FSEs use the Raysafe X2 systems, which have a measurement error of 5% (below 25 kV) or 10% (above 25 kV). For a system with an HVL ~ 0.5 mm Al at 28 kV, then the measurement error is well above 0.01 mm Al.

So how do we (QMPs) address the situation where a manufacturer cites an issue based on a problematic performance criterion? For any other modality I would make a note of why the manufacturer’s criteria are problematic, but with the added regulatory requirements of mammography and the authority that MQSA wields over us all, what is a reasonable approach?

Should we start reporting measured values as a range of values to represent errors in the measurement equipment? There are regulations saying that manufacturers need to provide the facility with a service report, but how much information are they required to provide? It seems reasonable that they would need to provide a description of the test that failed and what their measurements were, but does anyone get that in their service reports?

Rebecca.



Rebecca Milman, Ph.D., DABR
Division Chief, Radiological Sciences | Associate Professor
Department of Radiology
University of Colorado Anschutz
720.848.6625

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 Tao, Ashley T via Intl_dxmedphys_wd_osu_list <intl_dxmedphys_wd_osu_list at lists.osu.edu<mailto:intl_dxmedphys_wd_osu_list at lists.osu.edu>>
Date: Tuesday, May 19, 2026 at 2:27 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] MQSA - Manufacturer PM vs ACR DM QC manual

[External Email - Use Caution]
Dear DxMedPhys Listerv,

 I am wondering if anyone who is using the ACR DM QC manual has had any experience with MQSA inspectors giving citations for not following up with PM specifications.

Let me set the scene:

  1.
Our biomed engineer performed service on our unit and determined HVL to be outside of Hologic specs of +/-0.01 mm Al from their baseline/reference value. At 28kV, hologic baseline was 0.513 mm Al, and biomed measured 0.517. We were due for annual testing about a month and a half from the discovery. biomed asked if they could schedule PM for the day before physics testing. Since we were well above the minimum HVL requirement of 0.28 mm Al, we agreed.
MQSA inspectors said that ACR DM QC Test summary (mee report), Beam quality (HVL) assessment failures must be corrected before clinical use. Our HVL passes the ACR DM QC manual, but because biomed failed it, they said we needed to correct before clinical use and have an MEE.
  2.
Our biomed engineer performed service on our unit and determined AEC performance to be outside Hologic's specifications @ 2cm. Hologic references a range of EI values, whereas ACR DM QC manual specifies the SNR needs to be within +/-15% of the MEE. These two metrics are not comparable. We did perform a partial MEE because we were not certain whether or not we needed to. It turns out the MEE SNR was 22.34 and after AEC calibration, the new MEE SNR was 22.21. The MEE was completed 34 days after service but we were cited because it needed to be within 30 days. How do you reconcile the metrics used by Hologic vs ACR DM QC manual if they are not comparable?
  3.
Our biomed engineer performed service on our unit and determined target dose to be outside Hologic's specifications of +/-0.1 mGy. The ACR DM QC manual specifies the target dose for a CC view of the ACR DM QC phantom to be <=3 mGy and that the unit AGD must be within +/-25% of the calculated AGD. Again, Hologics metrics are different than the ACR DM QC manual. We are well within specifications, but because biomed failed it and we did not service it within 30 days, we were cited.

We showed the MQSA inspectors the service report and explained what the results were but they didn't seem to care and said that we would need to provide a corrective action letter before escalating to their next level inspector. It is my understanding that Hologic PM specs are more stringent for optimal performance. The ACR DM QC manual does not specify a timeline to address deficiencies in PM except for if there is relevant service and a finding were to be out of compliance with the ACR DM QC manual then you would need appropriate follow-up (e.g. immediately before clinical use versus 30 days). We are writing a response letter addressing these citations, but I would appreciate any input this group may have regarding their experience w/ inspectors and PMs being out of spec.

On a side note, if HVL is out of Hologic spec (but within MQSA requirements), and my biomed engineer needs to recalibrate but it also requires them to recalibrate AEC as part of the process, it seems that there would need to be on-site physicist testing of the AEC because according to  ACR DM QC manual - Table 3 Medical Physicist Involvement in Equipment Adjustments, Changes or Repairs(pg 138) - are any of you basically having to do MEEs every time your PMs are just a little bit out of spec?

Looking forward to hearing about your experiences.

Thanks
Ashley
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osu.edu/pipermail/intl_dxmedphys_wd_osu_list/attachments/20260602/b348b1c5/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Outlook-so0maln5.png
Type: image/png
Size: 90825 bytes
Desc: Outlook-so0maln5.png
URL: <http://lists.osu.edu/pipermail/intl_dxmedphys_wd_osu_list/attachments/20260602/b348b1c5/attachment.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Outlook-qqydiuas.png
Type: image/png
Size: 327487 bytes
Desc: Outlook-qqydiuas.png
URL: <http://lists.osu.edu/pipermail/intl_dxmedphys_wd_osu_list/attachments/20260602/b348b1c5/attachment-0001.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Outlook-IMG_202606.jpg
Type: image/jpeg
Size: 1577668 bytes
Desc: Outlook-IMG_202606.jpg
URL: <http://lists.osu.edu/pipermail/intl_dxmedphys_wd_osu_list/attachments/20260602/b348b1c5/attachment.jpg>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Outlook-IMG_1368.j.jpg
Type: image/jpeg
Size: 682865 bytes
Desc: Outlook-IMG_1368.j.jpg
URL: <http://lists.osu.edu/pipermail/intl_dxmedphys_wd_osu_list/attachments/20260602/b348b1c5/attachment-0001.jpg>

More information about the Intl_dxmedphys_wd_osu_list mailing list