NAF: Neural Attenuation Fields for Sparse-View CBCT Reconstruction

TL;DR

This paper introduces a fast, self-supervised neural method for sparse-view CBCT reconstruction that does not require external data and achieves state-of-the-art accuracy by representing attenuation as a neural function.

Contribution

It presents a novel neural attenuation field approach with an efficient hash coding encoder for high-quality, sparse-view CBCT reconstruction without external training data.

Findings

Achieves state-of-the-art accuracy in CBCT reconstruction.

Operates efficiently with short computation time.

Outperforms traditional frequency-domain encoders.

Abstract

This paper proposes a novel and fast self-supervised solution for sparse-view CBCT reconstruction (Cone Beam Computed Tomography) that requires no external training data. Specifically, the desired attenuation coefficients are represented as a continuous function of 3D spatial coordinates, parameterized by a fully-connected deep neural network. We synthesize projections discretely and train the network by minimizing the error between real and synthesized projections. A learning-based encoder entailing hash coding is adopted to help the network capture high-frequency details. This encoder outperforms the commonly used frequency-domain encoder in terms of having higher performance and efficiency, because it exploits the smoothness and sparsity of human organs. Experiments have been conducted on both human organ and phantom datasets. The proposed method achieves state-of-the-art accuracy and spends reasonably short computation time.