Room: Karl Dean Ballroom B1
Purpose: Real-time target localization using fast volumetric ultrasound (4D-US) is being increasingly investigated for motion compensation. New 4D-US systems with matrix arrays can achieve much higher framerates than mechanical 3D probes. In this study, three current systems with streaming capabilities were assessed regarding their suitability for real-time tracking tasks.
Methods: Maximum volume sizes, volumetric framerates and image quality metrics were compared for a GE Vivid7 dimension (3V probe), a GE Vivid E95 (4V) and a Philips Epiq7 (X6-1) at similar settings. As an indicator for spatial resolution, the mean full width at half maximum (FWHM) was measured in a commercial wire phantom in each direction (0.2mm wire diameter). The speckle signal-to-noise ratio (SSNR) and the contrast-to-noise ratio (CNR) were measured in a calibration phantom with a spherical target and in an abdominal phantom.
Results: At 150mm depth, the Vivid7 covered a maximum area of 122x84mmÂ² (azimuth, elevation) at 13.8Hz. For a similar size, E95 acquired at 20.9Hz and Epiq7 at 8Hz. Set to maximum, the area increased to 266x266mmÂ² at 3Hz for E95 and to 294x290mmÂ² at 2Hz for Epiq7. The mean FWHM in the smaller volumes was 2.8mm (E95), 3.0mm (Epiq7) and 3.1mm (Vivid7), which increased to 3.1mm (E95) and 3.4mm (Epiq7) for maximum volume sizes. The Epiq7 had superior SSNR values throughout all experiments, leading to smoother homogenous regions compared to E95 or Vivid7 volumes. In contrast, CNR was target-dependent. In the kidney tumor, CNR was higher for E95 and Vivid7 while Epiq7 had higher CNR in the spherical target.
Conclusion: All three 4D-US systems achieved higher framerates than the currently available commercial solution, with newer systems (E95, Epiq7) covering larger volumes. While the E95 showed the highest spatiotemporal resolution, the Epiq7 had superior SSNR. Their individual strengths will be further investigated in tracking experiments and in-vivo.
Funding Support, Disclosures, and Conflict of Interest: Research supported by the German Federal Ministry of Education and Research (grant no. 13GW0228B), the German Federal Ministry of Economy (grant no. ZF4109402BZ6), the German Research Foundation (grant no. ER 817/1-1) and the Ministry of Economic Affairs, Employment, Transport and Technology of Schleswig-Holstein.