Neural Modulation Fields for Conditional Cone Beam Neural Tomography

TL;DR

This paper introduces a conditional neural field approach for cone beam CT reconstruction that leverages anatomical consistency across scans, enabling improved density imaging with fewer projections and robustness to noise.

Contribution

It proposes a novel neural modulation field for conditioning neural fields on patient data, reducing the need for training from scratch for each scan.

Findings

Enhanced reconstruction quality with fewer projections.

Robust performance on noisy data.

Effective leveraging of anatomical consistency.

Abstract

Conventional Computed Tomography (CT) methods require large numbers of noise-free projections for accurate density reconstructions, limiting their applicability to the more complex class of Cone Beam Geometry CT (CBCT) reconstruction. Recently, deep learning methods have been proposed to overcome these limitations, with methods based on neural fields (NF) showing strong performance, by approximating the reconstructed density through a continuous-in-space coordinate based neural network. Our focus is on improving such methods, however, unlike previous work, which requires training an NF from scratch for each new set of projections, we instead propose to leverage anatomical consistencies over different scans by training a single conditional NF on a dataset of projections. We propose a novel conditioning method where local modulations are modeled per patient as a field over the input domain through a Neural Modulation Field (NMF). The resulting Conditional Cone Beam Neural Tomography (CondCBNT) shows improved performance for both high and low numbers of available projections on noise-free and noisy data.