[Intl_DxMedPhys] EXT MSG Fwd: The New CMS Measure: Methods for Size-Adjusted Dose and Their Variabilities

Tue Mar 11 14:57:40 EDT 2025

Further efforts on identifying optimal GN parameters to match with
subjective observers (ROI placement) are here:

https://urldefense.com/v3/__https://aapm.onlinelibrary.wiley.com/doi/full/10.1002/mp.14635__;!!KGKeukY!xb4PCI8L1oc7PteXhhiJFaDYWopNEYSG3GHW9jbZwWowPSpXp02zVcwpDb_KYnUUNVIG3LBSvH0lq7v6ZjuoHlYAtNY8C1rq0FRG9R9hK0Eo$ 

https://urldefense.com/v3/__https://aapm.onlinelibrary.wiley.com/doi/full/10.1002/mp.15133__;!!KGKeukY!xb4PCI8L1oc7PteXhhiJFaDYWopNEYSG3GHW9jbZwWowPSpXp02zVcwpDb_KYnUUNVIG3LBSvH0lq7v6ZjuoHlYAtNY8C1rq0FRG9eNLouD6$ 

Of note: we found that when using the optimal parameters derived from adult
CT on pediatric CT, in particular the square kernel size (ROI size) for
automated image noise assessment, they did not work as well as for adult
CT. The variations in reconstructed FOV and pixel size made the kernel size
choice poor and there were biases in the eventual GN data produced. This
can be addressed by scaling the kernel size to the reconstructed FOV and
pixel size, if desired.

I'm not sure how, if at all, this is handled by software related to the CMS
measure and trying to measure GN across such a wide range of clinical CT
data with likely very wide ranges of reconstructed FOV and pixel size.

----------------------------

*Allan Thomas, PhD, DABR*

Assistant Professor of Radiology

Mallinckrodt Institute of Radiology   *|  * Washington University School of
Medicine

510 S. Kingshighway  |  St. Louis, MO  63110

(*:*501.529.8496  |  **:* allant at wustl.edu

[image: A close-up of a logo Description automatically generated]

On Tue, Mar 11, 2025 at 1:43 PM Justin Solomon via
Intl_dxmedphys_wd_osu_list <intl_dxmedphys_wd_osu_list at lists.osu.edu> wrote:

> I believe most commercial offerings that do CT global noise measurements
> follow some variant of this method: https: //pubmed. ncbi. nlm. nih.
> gov/26102424/ There are some details/parameters that could vary between
> implementations. Would be an interesting
> I believe most commercial offerings that do CT global noise measurements
> follow some variant of this method:
> https://urldefense.com/v3/__https://pubmed.ncbi.nlm.nih.gov/26102424/__;!!KGKeukY!xb4PCI8L1oc7PteXhhiJFaDYWopNEYSG3GHW9jbZwWowPSpXp02zVcwpDb_KYnUUNVIG3LBSvH0lq7v6ZjuoHlYAtNY8C1rq0FRG9e4cf1yN$ 
> <https://urldefense.com/v3/__https://pubmed.ncbi.nlm.nih.gov/26102424/__;!!KGKeukY!wo8RkvE8k7a3feJqhXalDEECDyEqcOCuOM5YS3bSVdbt6V2c0_wMGKXkI8zHDRIfHlhWeWhQghkXUqkFtYXfDNv2RvQQ8WUZ5K9mSIUFlXtfu4CI$>
>
> There are some details/parameters that could vary between implementations.
> Would be an interesting study to compare vendors with a common dataset.
>
> That’s said we’ve had an in house built system making these measurements
> for many years now. In my experience the challenging part of using those
> data is sorting through meta data / protocol / series naming
> inconsistencies when doing aggregated analysis. I’m guessing those sources
> of variability and uncertainty dwarf the variability due to global noise
> measurement implementation.
>
> Cheers,
> Justin
> ————————
> Justin Solomon, PhD DABR
> Radiation Physicist
> Duke Health
> 919-684-1441
>
> On Mar 11, 2025, at 9:51 AM, Jie Zhang via Intl_dxmedphys_wd_osu_list <
> intl_dxmedphys_wd_osu_list at lists.osu.edu> wrote:
>
> 
> Yes. We have also investigated global noise metrics and their potential
> limitations in clinical applications. We are wrapping up our writing and
> hope to share our results soon. Thanks! Jie On Tue, Mar 11, 2025 at 11: 32
> AM Gauntt, David <davidgauntt@ uabmc. edu>
> Yes. We have also investigated global noise metrics and their potential
> limitations in clinical applications. We are wrapping up our writing and
> hope to share our results soon.
>
> Thanks!
>
> Jie
>
> On Tue, Mar 11, 2025 at 11:32 AM Gauntt, David <davidgauntt at uabmc.edu>
> wrote:
>
>> Jie,
>>
>>
>> Does your paper  address the issue of image quality?
>>
>>
>> The new CMS measure addresses both dose and image quality, using "global
>> noise" as a measure of image quality.  However, as near as I can tell the
>> measure does not address how you measure "global noise" in a clinical
>> image, and have not found any papers from the measure's stewards addressing
>> this issue.
>>
>>
>> Does anyone have any references to anything addressing how you calculate
>> global noise?
>>
>>
>> ---
>> David M. Gauntt, Ph.D.
>> Associate Professor
>> Division of Physics and Engineering
>> Department of Radiology
>> (205)975-3777
>>
>> <OutlookEmoji-1647458080270d7f02ca5-59f0-4d92-8fab-6fcda46b3f08.png>
>>
>>
>>
>>
>>
>> ------------------------------
>> *From:* Intl_dxmedphys_wd_osu_list
>> <intl_dxmedphys_wd_osu_list-bounces+davidgauntt=uabmc.edu at lists.osu.edu>
>> on behalf of Jie Zhang via Intl_dxmedphys_wd_osu_list <
>> intl_dxmedphys_wd_osu_list at lists.osu.edu>
>> *Sent:* Tuesday, March 11, 2025 10:04 AM
>> *To:* intl_dxmedphys_wd_osu_list at lists.osu.edu
>> *Subject:* EXT MSG [Intl_DxMedPhys] Fwd: The New CMS Measure: Methods
>> for Size-Adjusted Dose and Their Variabilities
>>
>> Thanks to Dr. Wunderle! I originally posted this to the old listserver
>> and am now forwarding it to the new Dx list. I am sorry if you receive it
>> twice. Jie ---------- Forwarded message --------- From: Jie Zhang <jzctjmn@
>> gmail. com> Date:
>> Thanks to Dr. Wunderle! I originally posted this to the old listserver
>> and am now forwarding it to the new Dx list.  I am sorry if you receive it
>> twice.
>>
>> Jie
>>
>> ---------- Forwarded message ---------
>> From: Jie Zhang <jzctjmn at gmail.com>
>> Date: Tue, Mar 11, 2025 at 9:15 AM
>> Subject: The New CMS Measure: Methods for Size-Adjusted Dose and Their
>> Variabilities
>> To: DXIMGMEDPHYS at HERMES.GWU.EDU <dximgmedphys at hermes.gwu.edu>
>>
>>
>> Dear Colleagues,
>>
>> The new CMS guidance on excessive radiation measurement allows
>> flexibility in software selection, but this may lead to inconsistencies and
>> undermine its goal.
>>
>> We recently evaluated five methods for estimating effective diameter and
>> their impact on Size-Specific Dose Estimate (SAD) in thoracic and abdominal
>> CT. A retrospective analysis of 719 exams revealed inconsistencies in SAD
>> methods, dose thresholds, and application across patient populations and
>> institutions. Addressing these issues is critical for accurate dose
>> reporting and diagnostic integrity.
>>
>> Our manuscript is available on arXiv: https://urldefense.com/v3/__https://arxiv.org/abs/2503.06644__;!!KGKeukY!xb4PCI8L1oc7PteXhhiJFaDYWopNEYSG3GHW9jbZwWowPSpXp02zVcwpDb_KYnUUNVIG3LBSvH0lq7v6ZjuoHlYAtNY8C1rq0FRG9fA11rCS$ 
>> <https://urldefense.com/v3/__https://arxiv.org/abs/2503.06644__;!!KGKeukY!wkvI7i3VRXaaElI73WdyPRf_Zk6qQqZ2CCRW0BoG-aYbTN1bVG_WIFbTeFOBYzuVEpUvax1OJJGkFDM4g68KPmxld3yXDnHKjwB6aQ$>
>> .
>>
>> Our findings reinforce concerns raised by the AAPM Commissioned Panel,
>> recently published in *AJR*, and we hope to provide a technical
>> perspective on the CMS measure.
>>
>> Best Regards,
>>
>> Jie
>>
>> *Jie Zhang, PhD, DABR (D, N), FAAPM*
>>
>> Professor of Radiology and Biomedical Engineering
>>
>> Chief, Division of Diagnostic & Nuclear Medical Physics
>>
>> Program Director, Diagnostic Imaging Physics Residency
>>
>> Department of Radiology
>>
>> University of Kentucky College of Medicine
>>
>> UK HealthCare
>>
>> 800 Rose Street, Room HX-307
>>
>> Lexington, KY 40536-0293
>>
>> Phone: (859) 323-2954
>>
>> Email: jnzh222 at uky.edu
>>
>>>
>>>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.osu.edu/mailman/private/intl_dxmedphys_wd_osu_list/attachments/20250311/69862cab/attachment.html>