A novel method for automatic evaluation of the accuracy in distance measurement performed by medical ultrasound scanners

In the last years the number of ultrasound (US) scanners utilized for medical diagnosis rapidly increased. As a consequence, specific methods for the evaluation of the quality of the US images have been developed, with the aim of verifying the performances of these high-technology apparatuses. However, most of the proposed methods seem to depend on the subjective evaluation of US images, produced by means of test objects (US phantoms), performed by expert operators. Among the examined performances, distance accuracy was usually evaluated on the screen of the US scanner by means of (a) the adjustment of the device settings and (b) the use of the electronic caliper implemented in the scanner, both directly set by the operator. Moreover, also the method for distance accuracy evaluation is not standardized and, therefore, the result is definitely dependent on the adopted procedure. The aim of the present study is the development of a novel method for the automatic evaluation of accuracy in distance measurement performed by medical US scanners. The proposed set-up adopts common US test phantoms as a reference and allows the evaluation of accuracy in vertical and horizontal distance measurements by a least squares estimation on the results of the analysis of the uncompressed bi-dimensional US images directly produced by the scanner. Finally, a specific index for global distance measurement accuracy is proposed.