The new ACR Digital Mammography Quality Control Manual for Tomosynthesis Has Arrived!

On November 19, 2018, the ACR released the 2018 Digital Mammography Quality Control manual that includes testing procedures for Digital Breast Tomosynthesis (DBT). The previous manual, released in 2016, was not applicable to DBT, but after significant effort by the ACR Committee on Mammography Accreditation, the new manual has been approved by the FDA as an alternative standard under MQSA. A link to download the new manual was sent to each accredited facility and can be shared with the supporting physicist.

There are a number of important points to note about the new quality control manual before we address some of the specifics:

• Implementation of the new manual is NOT REQUIRED. It was approved as an alternative standard. A facility may choose to continue using the manufacturer's QC program.
• The new manual REQUIRES the use of the new ACR FFDM phantom. The FFDM phantom is 31 x 19 cm and allows artifact evaluation with the phantom score. The phantom includes more fibers, speck groups, and masses than the old phantom and these objects are manufactured to a higher tolerance than the previous phantom. This new phantom is available from a number of retailers, listed on the ACR website.
• If your facility chooses to implement the new manual, the entire manual must be implemented. You are not allowed to use some parts of the manufacturer's QC program and parts of the ACR program. However, facilities must maintain any manufacturer-specified calibrations at their specified intervals (i.e., flat field calibration, geometry calibration, etc.)
• Prior to a facility implementing the new QC program, the medical physicist MUST PERFORM a full annual survey using the new ACR QC manual. This will allow the physicist to establish baselines and review the technologist QC program prior to implementation.

The ACR has posted a very helpful FAQ regarding the new QC manual, as well as Excel versions of the medical physicist forms. The Excel forms include functional graphs, testing instructions, and basic pictures to help demonstrate the testing methodology. The use of these forms is highly encouraged.

While many of the tests in the new QC manual are familiar, there are some notable changes and additions to the standard manufacturer QC programs.

• The minimum passing score for the new phantom is 2 fibers, 3 speck groups, and 2 masses. This score represents the same level of visualization from the previous phantom, but due to the new phantom specifications, the largest objects from the previous phantom were omitted. The ACR also provides new guidance for what must be visualized to count a test object. For example, to score a full fiber, at least 8 mm of the fiber must be visible, it must be in the correct location, correct orientation and 1 break is allowed as long as it is less than or equal to the width of the fiber. A half fiber must be greater than or equal to 5 mm in length.
• Artifacts should be inspected using the ACR FFDM phantom. Clinically relevant artifacts that can obscure clinical information will cause a facility to fail accreditation. With the old phantom, reviewers were not permitted to fail a facility due to artifacts that were visible on the submitted phantom.
• There is a test for Z-resolution. This test requires placing ROIs over the specks in the largest speck group and recording the maximum signal for the best slice, as well as the three slices before and after. The provided Excel forms will calculate the FWHM resolution and this value will be compared to the initial baseline annually.
• Spatial resolution Pass/Fail limits are 4 lp/mm in 2D mode, 6 lp/mm in magnification mode, and 2 lp/mm in 3D mode.
• AEC testing requires the calculation of SNR for 2, 4, 6, and 8 cm phantoms. The annual results must be within 15% of the MEE. Density step testing has been removed.
• For systems that display a computed Average-Glandular Dose, the ACR recommends that the displayed value be within 25% of the physicist measured value.
• There are QC requirements for the acquisition display monitor. This includes maximum and minimum luminance, luminance uniformity, and contrast response.
• Beam Quality (HVL) and kVp accuracy and reproducibility are now required at MEE and for troubleshooting only. They are not required for annual physics testing.

The ACR will be scheduling webinars for facilities, technologists, and physicists in the coming months to help facilities implement the new QC manual. There are new requirements as well for technologist QC, as well as a section helping the lead interpreting physician to better understand and oversee the full quality control program. For more information, click here.

Finally, if you have implemented the new manual or intend to do so in the future, please reach out to me. There may be interest in having early adopters share their experiences at future AAPM meetings.