Multivariate Gaussian NeRF for Wide Field-of-View Ultrasound Reconstruction

TL;DR

Ultra-Wide-NeRF introduces a multivariate Gaussian NeRF-based method for wide field-of-view ultrasound reconstruction, reducing artifacts and enabling high-fidelity view synthesis for improved intraoperative navigation.

Contribution

The paper presents a novel MVG NeRF approach that explicitly models beam geometry, mitigating artifacts and providing continuous tissue representations for ultrasound imaging.

Findings

Reduces compounding artifacts and aliasing in WFoV ultrasound reconstruction.

Enables synthesis of high-fidelity views from arbitrary trajectories.

Validated on phantom and porcine datasets for intracardiac echocardiography.

Abstract

Wide Field-of-View (WFoV) reconstruction enhances 3D ultrasound imaging by providing valuable anatomical context for segmentation models and visualization. Clinical ultrasound volumes are predominantly acquired using convex probes, which generate expanding, diverging acoustic beams to maximize anatomical coverage. Stitching these sweeps together traditionally introduces significant compounding artifacts and aliasing due to depth-dependent resolution changes. Here, we introduce Ultra-Wide-NeRF, a Multivariate 3D Gaussian (MVG) NeRF-based method for WFoV ultrasound reconstruction. By explicitly modeling the complex beam geometry using distance-dependent convex volumetric sampling and anisotropic 3D Gaussians, our method inherently mitigates these compounding artifacts and provides anti-aliasing. Beyond simply reconstructing a static 3D grid, our NeRF-based approach yields a continuous neural representation of the tissue, enabling the synthesis of high-fidelity novel views from arbitrary virtual trajectories. We validate Ultra-Wide-NeRF for intracardiac echocardiography on phantom and porcine datasets, demonstrating that our method expands the spatial context important in intraoperative navigation. Code will be open-sourced upon publication.