DeepSparse: A Foundation Model for Sparse-View CBCT Reconstruction

TL;DR

DeepSparse is a novel foundation model for sparse-view CBCT reconstruction that combines multi-view and multi-scale features, pretraining on large datasets, and finetuning strategies to improve image quality and reduce radiation exposure.

Contribution

It introduces DeepSparse, the first foundation model for sparse-view CBCT, with a new network architecture and pretraining framework for better generalization and reconstruction quality.

Findings

Outperforms existing methods in reconstruction quality

Effective pretraining on large datasets improves generalization

Two-step finetuning enhances adaptation to new datasets

Abstract

Cone-beam computed tomography (CBCT) is a critical 3D imaging technology in the medical field, while the high radiation exposure required for high-quality imaging raises significant concerns, particularly for vulnerable populations. Sparse-view reconstruction reduces radiation by using fewer X-ray projections while maintaining image quality, yet existing methods face challenges such as high computational demands and poor generalizability to different datasets. To overcome these limitations, we propose DeepSparse, the first foundation model for sparse-view CBCT reconstruction, featuring DiCE (Dual-Dimensional Cross-Scale Embedding), a novel network that integrates multi-view 2D features and multi-scale 3D features. Additionally, we introduce the HyViP (Hybrid View Sampling Pretraining) framework, which pretrains the model on large datasets with both sparse-view and dense-view projections, and a two-step finetuning strategy to adapt and refine the model for new datasets. Extensive experiments and ablation studies demonstrate that our proposed DeepSparse achieves superior reconstruction quality compared to state-of-the-art methods, paving the way for safer and more efficient CBCT imaging.