Searcharxiv⌕ Search

arXiv · 2611.18766

Angel: Anchor-Structured Neural Gaussians for Explicit Fetal Diffusion MRI Slice-to-Volume Reconstruction with Cross-Subject Learning

Abstract

Diffusion magnetic resonance imaging (dMRI) provides a noninvasive view of developing tissue microstructure and white matter tracts in the fetal brain. During acquisition, however, fetal motion disrupts spatial correspondence between slices, diffusion encoding directions and signal intensities. We introduce ANGEL, an anchor-structured neural Gaussian framework for accurate slice-to-volume reconstruction (SVR) of fetal dMRI, that simultaneously estimates diffusion signal, fetal motion and intensity artefacts. Unlike standard explicit Gaussian representation, ANGEL is designed to prevent absorption of noise and artefacts into the reconstructed dMRI by learning a compact representation of spatio-angular signal via a shared decoder that predicts 3D Gaussians across anchors and subjects. The individual anatomical variability is compactly represented through subject-specific anchors. Experiments on simulated neonatal and in vivo fetal data show significant improvement of dMRI reconstruction in both spatial and angular domains, in comparison to explicit Gaussian representation, implicit neural representation as well as classical methods.

Explore related subjects

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Wenxuan Wu, Lucy Hall, Jonathan O'Muircheartaigh, Joseph V. Hajnal, J-Donald Tournier, Maria Deprez. 2026-11-12. Angel: Anchor-Structured Neural Gaussians for Explicit Fetal Diffusion MRI Slice-to-Volume Reconstruction with Cross-Subject Learning. https://searcharxiv.com/abs/2611.18766

Citations retain the original authors and DOI and include this site’s record URL. If this discovery service contributed to your research, you can also cite Searcharxiv by Indiciae: BibTeX · RIS.

KEEP EXPLORING

Related papers

Physics-Aware Zero-Shot Learning for Low-Dose CT from a Single Sinogram

Low-dose computed tomography (CT) imaging remains a persistent challenge, especially under heterogeneous acquisition conditions. Zero-shot learning, which enables models to learn directly from individual images without external supervision, represents a compelling approach to mitigate distribution mismatches arising from variations in scanner hardware, imaging protocols, and patient demographics. However, most existing zero-shot approaches employ down-sampling operations that reduce spatial resolution and cause domain shifts, which become particularly evident when applied to raw sinogram data. Here, we propose a physics-aware zero-shot CT denoising framework to exploit the conjugate symmetry property intrinsic to CT sinograms. Conjugate X-ray pencil beams traverse identical anatomical paths and ideally yield the same expected attenuation values but with independent noise realizations. By randomly mapping sinogram values between conjugate-ray pairs, we synthesize structurally consistent yet noise-diverse sinogram pairs, enabling self-supervised network training without external data. The resulting lightweight denoising network demonstrates strong generalization across low-dose simulation, clinical helical CT, and preclinical photon-counting CT datasets. Quantitative and qualitative results demonstrate competitive denoising performance, with favorable noise suppression and structural preservation across diverse acquisition settings. This work establishes a generalizable and data-efficient paradigm for CT denoising that integrates physical redundancy into learning-based image reconstruction. The code for the proposed method is openly available at: https://github.com/shiyongyi/Physics-Aware-Zero-Shot-CT-Sinogram-Denoising

eess.IV↗

QUTCC: Quantile Uncertainty Training and Conformal Calibration for Imaging Inverse Problems

While deep learning offers tremendous promise for scientific and medical imaging, any failures and hallucinations (predictions that do not coincide with reality) are hard to pinpoint and can have serious downstream consequences. Uncertainty estimation techniques, such as conformal prediction, can help by predicting statistically valid error bars for a model's prediction. However, popular conformal prediction methods were not designed for high-dimensional image-valued problems and do not take into account spatial correlations within an image during conformal calibration, resulting in larger-than-necessary uncertainty intervals. We propose a practical simultaneous quantile regression method that enables non-linear, spatially-adaptive scaling during conformal calibration. Our method, QUTCC uses a U-Net architecture with a quantile embedding to learn a full conditional quantile distribution during training, and then leverages this non-linear, learned function for spatially-adaptive conformal calibration. At test time, our method can efficiently estimate uncertainty intervals with pixel-marginal coverage guarantees. In addition, QUTCC can also predict pixel-wise conditional probability density estimates without any built-in distributional assumptions. We evaluate our method on several denoising problems, accelerated magnetic resonance imaging, and quantitative phase microscopy. Our method consistently produces tighter uncertainty intervals than prior conformal methods at the same coverage level, can predict plausible conditional distributions for different tasks, and in some cases, high-uncertainty regions can help us locate hallucinations in a model's prediction.

eess.IV↗

GroundGazer: An optical reference system for planar localization with millimeter accuracy at low cost

Highly accurate indoor localization systems with absolute mm positioning accuracy are currently expensive. They include laser trackers, total stations, and motion capture systems relying on multiple high-end cameras. In this work, we introduce a high-accuracy, planar indoor localization system named GroundGazer (GG) for autonomous mobile robots (AMRs). GG estimates the AMR's planar position with mm and its heading with sub-degree accuracy. The system requires only a monocular (fisheye) camera, a chessboard floor, and an optional laser diode. Our system is simple and low-cost due to the chessboard floor, robust, scalable to multiple robots, and extendable to 3D position and orientation estimation.

eess.IV↗