Energy-Based Prior Latent Space Diffusion model for Reconstruction of Lumbar Vertebrae from Thick Slice MRI

TL;DR

This paper introduces a novel energy-based prior latent space diffusion model that improves 3D reconstruction of lumbar vertebrae from thick slice MRI, outperforming existing methods in quality and feature capture.

Contribution

It proposes a new diffusion-based approach with an energy-based latent prior to enhance lumbar vertebrae reconstruction from thick slice MRI, reducing computational costs and improving accuracy.

Findings

Outperforms existing methods in Dice and VS scores.

More accurately captures 3D vertebral features.

Reduces computational cost of diffusion models.

Abstract

Lumbar spine problems are ubiquitous, motivating research into targeted imaging for treatment planning and guided interventions. While high resolution and high contrast CT has been the modality of choice, MRI can capture both bone and soft tissue without the ionizing radiation of CT albeit longer acquisition time. The critical trade-off between contrast quality and acquisition time has motivated 'thick slice MRI', which prioritises faster imaging with high in-plane resolution but variable contrast and low through-plane resolution. We investigate a recently developed post-acquisition pipeline which segments vertebrae from thick-slice acquisitions and uses a variational autoencoder to enhance quality after an initial 3D reconstruction. We instead propose a latent space diffusion energy-based prior to leverage diffusion models, which exhibit high-quality image generation. Crucially, we mitigate their high computational cost and low sample efficiency by learning an energy-based latent representation to perform the diffusion processes. Our resulting method outperforms existing approaches across metrics including Dice and VS scores, and more faithfully captures 3D features.