A review of two alternative ultrasound quality assurance programmes.

OBJECTIVE: This study reviewed the results of B-mode Quality Assurance (QA) performance tests on 17 real-time ultrasound scanners, performed over a period of 3 years, in order to assess their value. Following this review we revised and simplified our testing schedules to include two tests for noise and sensitivity. The value of the new schedules was assessed.
METHODS: Initially, testing schedules were similar to those recommended by two professional bodies. Results were reviewed to determine whether the tests predicted or confirmed faults. We then introduced a simplified testing programme using alternative measurements, attempting to demonstrate or predict noise related faults that affect the image, but were not demonstrated by current tests. These new tests have been performed on 24 ultrasound machines for up to 18 months.
RESULTS: A review of results has shown that measurements occasionally fall outside tolerance due to chance, and that faults that significantly affect the image, e.g. probe faults and noise, are reported by the users without predictive or concomitant changes in test results using our original schedules. Faults occur that do not immediately affect image quality and are not reported by the users. Inappropriate settings, e.g. monitors, are frequently reset at QA, particularly where there are potentially untrained users. The additional tests showed consistent changes in noise (four) or sensitivity (one) on five machines.
CONCLUSION: Our earlier tests were inadequate in demonstrating deterioration in the clinical performance of ultrasound imaging equipment. Introduction of a revised testing schedule has resulted in changes in equipment performance being detected and rectified.