Geoff,
OK; but what (which) test or QA method are we talking about?
The BMUS publication you linked to shows a cross filament phantom which provides a 3D qualitative presentation of the beam shape throughout depth as described in the text. This test object is a useful tool to give a more in depth understanding of the physics behind what we see clinically but it cannot provide quantitative data. The set up time is longer due to alignment of the threads. This test object could be used to establish a baseline to be referred to over time and for educational purposes but it would not be the first choice of tool to use for routine QA.
On the other hand a random void phantom (RVP) can provide quantitative data in terms of "Sonic Contrast" ( SC) which is defined in IEC TS 61390. SC has clinical significance and can be used to determine whether or not a probe should be used for a given clinical application taking into account the amount of overlying fat of a given patient. Further research into the clinical uses of SC is needed. The RVP also provides a way of viewing image data in 3D which makes subtle defects evident to the eye. The RVP in essence is a recticulated foam sponge of known attenuation, immersed in salt water, again described in the TS 61390. Image capture is quick and alignment issues are overcome.
The tests can be carried out in house as part of a routine QA program. Once you are up to speed little additional test time is required. Analysis can be done back at the office. The stored information would then be used as a reference should a probe be repaired or the image quality is suspect.
It would not be too much of a stretch of the imagination to envisage manufacturers building a propriety version of the software into their scanner and request a daily test scan by the sonographer for the probe selected with a on screen message appearing should a probe go out of tolerance. It is important to retain an independent version of the test method to ensure transparency of evaluation of the manufacturers results. It would be unlikely "anomalies" in their images would be flagged up as out of tolerance. Nor are propriety versions likely to provide comparative image quality information.
