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

2025-01-22

AI Summary

Conventional chest CT scans cover only the thoracic region, limiting assessment of secondary abdominal organ involvement (e.g., liver, kidneys) in pulmonary diseases. To address this, we propose SCOPE—a novel framework that extends the field of view (FOV) from chest to abdomen without paired training data. SCOPE leverages a variational autoencoder (VAE) to encode 2D axial slices into a latent space, then constructs a 3D latent representation and enables zero-shot slice generation along the z-axis via implicit 3D contextual modeling through latent stacking. Its core innovations are (1) latent-space-based 3D contextual modeling and (2) zero-shot cross-domain generation. Evaluated on the NLST dataset, SCOPE successfully reconstructs anatomically consistent abdominal regions—including liver and kidneys. On a whole-body CT dataset, generated slices achieve an SSIM of 0.81 with high visual fidelity, enabling cross-organ pathological correlation analysis.

Technology Category

• Computer Vision: Medical and Biological Imaging
• Machine Learning: Deep Generative Models & Autoencoders
• Search and Optimization: Mixed Discrete/Continuous Search

Application Category

• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Security and Privacy: Data transparency and provenance
• User Modeling, Personalization and Recommendation: Accountability, Transparency, and Ethics for personalization

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.

Problem

• comprehensive disease understanding
• extrapulmonary effects
• chest CT

Innovation

• SCOPE
• Latent Diffusion Models
• Variational Autoencoders