Beyond the Lungs: Extending the Field of View in Chest CT with Latent Diffusion Models

TL;DR

This paper introduces SCOPE, a novel method using latent diffusion models to extend the field of view in chest CT scans, enabling the inclusion of additional organs like the liver and kidneys for comprehensive analysis.

Contribution

The paper presents a new approach combining VAE and latent diffusion models to extend CT scan FOVs in a zero-shot manner, capturing inter-organ relationships.

Findings

Successfully extended FOV to include liver and kidneys

Achieved high fidelity with SSIM of 0.81 on held-out data

Demonstrated potential for comprehensive organ analysis in CT scans

Abstract

The interconnection between the human lungs and other organs, such as the liver and kidneys, is crucial for understanding the underlying risks and effects of lung diseases and improving patient care. However, most research chest CT imaging is focused solely on the lungs due to considerations of cost and radiation dose. This restricted field of view (FOV) in the acquired images poses challenges to comprehensive analysis and hinders the ability to gain insights into the impact of lung diseases on other organs. To address this, we propose SCOPE (Spatial Coverage Optimization with Prior Encoding), a novel approach to capture the inter-organ relationships from CT images and extend the FOV of chest CT images. Our approach first trains a variational autoencoder (VAE) to encode 2D axial CT slices individually, then stacks the latent representations of the VAE to form a 3D context for training a latent diffusion model. Once trained, our approach extends the FOV of CT images in the z-direction by generating new axial slices in a zero-shot manner. We evaluated our approach on the National Lung Screening Trial (NLST) dataset, and results suggest that it effectively extends the FOV to include the liver and kidneys, which are not completely covered in the original NLST data acquisition. Quantitative results on a held-out whole-body dataset demonstrate that the generated slices exhibit high fidelity with acquired data, achieving an SSIM of 0.81.