Abstract
Software beamforming allows flexible adaptive beamformers with great popularity and exponential growth in the number of papers published. Software beamforming can be a blessing to the image quality since access to raw channel data allows algorithms that better exploit the data. However, software beamforming can also be considered a curse, since the flexibility leads to non-linear algorithms that invalidate the conventional metrics used to evaluate the image quality. A review of the contrast metrics used in ultrasound imaging shows no consensus on the metrics used in the research literature. Multiple metrics exist and different types of data, in different scales, from different stages of the processing chain, are used as input to the metrics.We have demonstrated that many adaptive algorithms alter the dynamic range of the ultrasound images, effectively breaking the conventional metrics both for contrast and resolution. Therefore, we have introduced a new improved contrast metric, the generalized contrast-to-noise ratio (gCNR) immune to dynamic range alternations. The gCNR can be estimated on all kinds of images, regardless of compression, scale, or output units. A major drawback of the conventional contrast metrics, as well as the gCNR, is that they require localization of specific regions and targets in the image. Such objects can be difficult to identify in in-vivo images. We recently introduced Global Image Coherence (GIC), as an in-vivo image quality metric that does not require any identified regions.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call