Framework for Fusion of Data- and Model-Based Approaches for Ultrasound Simulation

Tanner, Christine; Starkov, Rastislav; Bajka, Michael; Göksel, Orçun

doi:10.1007/978-3-030-00937-3_39

Cited by 2 publications

(12 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition to probe manipulation by an examiner, in-vivo US examination often involves a wide range of anatomical motions and displacement, for instance due to haphazard fetal muscle activity or periodic movements, such as the beating heart or breathing. Omitting these dynamic effects (e.g., as in [2,5] and even when they are user-interactive as in [7,14]) results in lifeless static anatomy, impairing the realism of the Fig. 7: Once Ray 1 encounters the initial intersection with the coarse acceleration structure, it examines the two underlying bounding boxes to find the next intersection and then proceeds to traverse the fine acceleration structure referenced by Bounding Box 2.…”

Section: B Animation Pipelinementioning

confidence: 99%

“…Firstly, if desired, in an offline step [14] the original image volume may be deformed to allow for inserting synthetic anatomy of a different size. Subsequently, the region of the prerecorded background data that is redundant and to be replaced by the simulated content is marked by a mask as shown in Fig.…”

Section: On-the-fly Fusionmentioning

confidence: 99%

“…We then use this mask during the onthe-fly fusion step where the values within the foreground are taken from the simulated ray-tracing output, whereas the remaining values are extracted from the US volume. This is akin to removing this masked content from the volume (as in [14]), however the fusion is achieved during simulation runtime, therefore obviating the need for an actual imagecontent deletion.…”

Section: On-the-fly Fusionmentioning

confidence: 99%

“…In earlier work [14] we introduced a combination of dataand model-based approaches as a compromise between the two, where the region of clinical interest is modeled in detail and simulated by means of Monte-Carlo ray-tracing, and then fused in a preprocessing step with lower-interest background images collected in-vivo, as presented in Fig. 1.…”

Section: Introductionmentioning

confidence: 99%

“…It can be seen that if the simulated image (a) is simply rotated, e.g. by 90 • , to achieve the orientation in (b), using image-based techniques such as in [7,14], then the acoustic shadowing artifact would be pointing in a completely wrong and unrealistic direction (c). Furthermore, orientations of speckles would similarly be incorrect.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Ultrasound simulation with animated anatomical models and on-the-fly fusion with real images via path-tracing

Starkov

Tanner

Bajka

et al. 2019

Computers & Graphics

Self Cite

View full text Add to dashboard Cite

Ultrasound is an essential imaging modality in clinical screening and diagnosis, for reducing morbidity and improving quality of life. Successfully performing ultrasound imaging, however, requires extensive training and expertise in navigating a hand-held probe to a correct anatomical location as well as subsequently interpreting the acquired image. Computergenerated simulations can offer a safe, flexible, and standardized environment to train such skills. Data-based simulations display interpolated slices from a-priori-acquired real ultrasound volumes, whereas generative simulations aim to reproduce the complex ultrasound interactions with comprehensive, geometric anatomical models, such as using ray-tracing to mimic acoustic propagation. Although sonographers typically focus on relatively smaller structures of interest in ultrasound images, the fidelity of the background anatomy may still play a role in contributing to the realism of a generated US image; e.g. when imaging a relatively smaller fetus within large abdominal background. It was proposed earlier to compose ray-traced images with acquired volumes in a preprocessing step. Despite its simplicity, this prevents any view-dependent artifacts and interactive model changes, such as those induced by animations, which can, for instance, model fetal motion. To fully leverage the flexibility of the model-based generative approach, we propose herein an onthe-fly image fusion, based on the two techniques, by moving the interpolation stage within the ray-tracer, such that the preacquired image data can be referred to in the background, while the acoustic interactions with the model can be resolved in the foreground. This allows for animated anatomical models, which we realize during simulation runtime via scene-hierarchy subtree switching between precomputed acceleration structure graphs. We demonstrate our proposed techniques on ultrasound sequences of fetal and heart motion, where only animated models can afford to meet realism requirements entailed by the temporal domain.

show abstract