[Intl_DxMedPhys] ACR NM Image Acquisition - Discrepancy

[Mike Nimmo]

No. And maybe that's the point. Since ACR evaluates the images
qualitatively, maybe 10 million counts is enough to evaluate artifacts. But
due to counting statistics per pixel, you have a decent chance to fail a
UFOV or CFOV differential or integral uniformity test against manufacturer
specifications at 10 million that you would otherwise pass at 17 or 30
million.

In my old life as residency faculty, these counting statistics questions
were a favorite to plague my residents with! Good question, Gretchen!

Mike Nimmo

On Wed, Nov 5, 2025, 2:53 PM Gretchen Raterman Bell <
gretchen.raterman at gmail.com> wrote:

> Thank you for the explanation, Mike.  Official nonuniformity numbers for
> "peak performance" should be done with 17-30 million.  The nonuniformity
> numbers will be better (or the same if there truly is a problem) when going
> from 10 to 30 million count statistics.  I would surmise that if you get
> acceptable numbers with 10 million, then the 30 million count flood may not
> be necessary.  However, from your experience, has there ever been an
> artifact that was seen at 30 million counts that would be possibly covered
> up by the noise in a 10 million flood?  I ask because we have run higher
> counts as a sanity check when we have gotten poor numbers, and the number
> typically stays the same.
>
> Gretchen R. Bell, M.S., DABR
> Diagnostic Imaging Physicist
> Ochsner Medical Center
> (504)842-8506
>
>
>
> On Wed, Nov 5, 2025 at 2:16 PM Mike Nimmo <mike.t.nimmo at gmail.com> wrote:
>
>> Hi Gretchen
>>
>> I think the disparity is between acr flood images for submission and
>> annual QC. TG177 lists a table of count statistics outlining why at least
>> 17 million counts should be acquired for annual testing, but it recommends
>> 30 million.
>>
>> Honestly I've always wondered why the ACR is so low. I always do at least
>> 17 million counts for my annual testing.
>>
>> Mike Nimmo
>>
>> On Wed, Nov 5, 2025, 2:06 PM Gretchen Raterman Bell via
>> Intl_dxmedphys_wd_osu_list <intl_dxmedphys_wd_osu_list at lists.osu.edu>
>> wrote:
>>
>>> Hey everyone, My team and I noticed a discrepancy between the ACR NM QC
>>> manual and the "Phantom Testing" guidance page on their website: We've of
>>> course always done 10 million (and previously 5 million).   Is this a typo
>>> in the
>>> Hey everyone,
>>>
>>> My team and I noticed a discrepancy between the ACR NM QC manual and the
>>> "Phantom Testing" guidance page on their website:
>>>
>>> [image: image.png]
>>>
>>> [image: image.png]
>>>
>>> We've of course always done 10 million (and previously 5 million).  Is
>>> this a typo in the ACR NM QC Manual?
>>>
>>> Gretchen R. Bell, M.S., DABR
>>> Diagnostic Imaging Physicist
>>> Ochsner Medical Center
>>> (504)842-8506
>>>
>>>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.osu.edu/mailman/private/intl_dxmedphys_wd_osu_list/attachments/20251105/01ad9d52/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image.png
Type: image/png
Size: 31052 bytes
Desc: not available
URL: <https://lists.osu.edu/mailman/private/intl_dxmedphys_wd_osu_list/attachments/20251105/01ad9d52/attachment.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image.png
Type: image/png
Size: 34639 bytes
Desc: not available
URL: <https://lists.osu.edu/mailman/private/intl_dxmedphys_wd_osu_list/attachments/20251105/01ad9d52/attachment-0001.png>