Purpose: To explore the potential of using a relatively high-density random spherical void (RSV) phantom along with statistically-based, quantitative metrics as an approach for assessing the performance of ultrasound imaging systems.
Methods: Ten computer-simulated RSV phantoms with a RSV volume fraction of 0.15 were generated. To determine the number of image planes necessary to provide robust measurements, a series of 1 to 150 “ideal image planes” were analyzed and the observed circle cross-section radii histograms were compared with the expected theoretical histogram. Simulated ultrasound image planes were produced by adding speckle and degradation of imaging system performance was modeled by averaging 1 to 9 neighboring planes to represent increasing elevation plane thicknesses. Quantification of the performance of the imaging system was determined by measuring the: 1) mean number of observed circle cross sections per frame; 2) mean fractional area of circle cross-sections per frame; and 3) agreement of observed circle cross section radii histograms with the theoretical distribution (Chi-square statistic).
Results: Results suggest that analyses of over 100 “ideal image planes” (>3000 circle cross-sections) provide excellent agreement between the observed and expected theoretical histogram distributions (mean chi-square < 0.004). For 150 image plane analyses, the degraded RSV phantom images show decreasing: mean number of circle cross sections detected per frame (31.5±0.3, 28.4±0.3, 28.2±0.3, 26.3±0.3 and 25.3±0.3); mean fractional area of circle cross-section per frame (0.157±0.002, 0.133±0.001, 0.133±0.001, 0.111±0.001 and 0.108±0.001); and agreement with theoretical histogram distribution (chi-square: 0.070±0.004, 0.140±0.005, 0.149±0.007, 0.379±0.011 and 0.518±0.010) for 1, 3, 5, 7 and 9 plane averages, respectively.
Conclusion: This simulation suggests that using a relatively high-density a RSV phantom, with an appropriate number of image frames analyzed, may represent an approach for providing quantitative, statistically-based assessment of ultrasound system performance in a reproducible fashion.