Background: Colour Doppler has become an integral part of many clinical ultrasound investigations, but evaluation of the spatial resolution of colour Doppler scanners remains problematic due to a lack of suitable test phantoms.

Methods: The aims of this study were to evaluate two designs of colour Doppler spatial resolution test-phantom to compare their suitability for determining the spatial resolution of five commercially available ultrasound scanners. For the purpose of this study, spatial resolution was defined as the minimum resolvable distance between two channels of flow. Phantom 1 consisted of two diverging flow-channels. Phantom 2 consisted of four pairs of parallel channels with different separations.

Results and Discussion: Both designs of phantom highlighted a number of advantages and disadvantages. In principle, phantom 1 allowed the minimal resolvable distance to be identified based on imaging a continuous range of separations, but in practice, the average separation being probed was uncertain, and the minimum separation between vessels of 1 mm was too large to test the resolutions of some of the scanners we studied. Phantom 2 was easier to set up and had a narrower minimum separation of 0.6 mm, but only four nominal separations were tested. Evaluation of five scanners using each prototype phantom suggested typical resolutions of 1·2 mm or lower. Many of the scanners possessed resolutions of approximately 0.6 mm or less, which was the minimum separation we were able to test. The poorest lateral resolution of approximately 1·2 mm was observed for a 13 year old scanner with curvilinear 4 MHz phased array transducer.

Conclusion: Construction of the 'perfect' test-object for measuring colour Doppler resolution remains a considerable challenge. Further work will be required to produce acoustically realistic flow phantoms capable of testing the ability of colour Doppler to resolve increasingly narrow flow separations.