Robust Body Composition Analysis by Generating 3D CT Volumes from Limited 2D Slices

TL;DR

This paper introduces a novel method that generates 3D CT volumes from limited 2D slices using a latent diffusion model, significantly improving body composition analysis accuracy while reducing radiation exposure concerns.

Contribution

The paper presents a new approach combining variational autoencoders and latent diffusion models to reconstruct 3D CT volumes from few 2D slices, enhancing robustness and accuracy.

Findings

Reduced error rate from 23.3% to 15.2% in body composition analysis.

Effective interpolation of intermediate slices using body part regression.

Improved robustness over traditional 2D analysis methods.

Abstract

Body composition analysis provides valuable insights into aging, disease progression, and overall health conditions. Due to concerns of radiation exposure, two-dimensional (2D) single-slice computed tomography (CT) imaging has been used repeatedly for body composition analysis. However, this approach introduces significant spatial variability that can impact the accuracy and robustness of the analysis. To mitigate this issue and facilitate body composition analysis, this paper presents a novel method to generate 3D CT volumes from limited number of 2D slices using a latent diffusion model (LDM). Our approach first maps 2D slices into a latent representation space using a variational autoencoder. An LDM is then trained to capture the 3D context of a stack of these latent representations. To accurately interpolate intermediateslices and construct a full 3D volume, we utilize body part regression to determine the spatial location and distance between the acquired slices. Experiments on both in-house and public 3D abdominal CT datasets demonstrate that the proposed method significantly enhances body composition analysis compared to traditional 2D-based analysis, with a reduced error rate from 23.3% to 15.2%.