Neural Modulation Fields for Conditional Cone Beam Neural Tomography
This work addresses the limitation of requiring per-scan training in neural field-based CT reconstruction, offering a more efficient approach for medical imaging applications.
The paper tackles the problem of reconstructing density from cone beam CT projections by proposing a conditional neural field that leverages anatomical consistencies across scans, resulting in improved performance for both high and low numbers of projections on noise-free and noisy data.
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.