Equivariant Multiscale Learned Invertible Reconstruction for Cone Beam CT: From Simulated to Real Data
This work addresses image quality limitations in CBCT for medical imaging applications, offering incremental improvements over existing deep learning methods.
The authors tackled the problem of low image quality in Cone Beam CT (CBCT) by proposing LIRE++, an equivariant multiscale learned invertible reconstruction method, which improved Peak Signal-to-Noise Ratio by 1 dB on synthetic data and reduced Mean Absolute Error by 10 Hounsfield Units on real clinical data compared to baselines.
Cone Beam CT (CBCT) is an important imaging modality nowadays, however lower image quality of CBCT compared to more conventional Computed Tomography (CT) remains a limiting factor in CBCT applications. Deep learning reconstruction methods are a promising alternative to classical analytical and iterative reconstruction methods, but applying such methods to CBCT is often difficult due to the lack of ground truth data, memory limitations and the need for fast inference at clinically-relevant resolutions. In this work we propose LIRE++, an end-to-end rotationally-equivariant multiscale learned invertible primal-dual scheme for fast and memory-efficient CBCT reconstruction. Memory optimizations and multiscale reconstruction allow for fast training and inference, while rotational equivariance improves parameter efficiency. LIRE++ was trained on simulated projection data from a fast quasi-Monte Carlo CBCT projection simulator that we developed as well. Evaluated on synthetic data, LIRE++ gave an average improvement of 1 dB in Peak Signal-to-Noise Ratio over alternative deep learning baselines. On real clinical data, LIRE++ improved the average Mean Absolute Error between the reconstruction and the corresponding planning CT by 10 Hounsfield Units with respect to current proprietary state-of-the-art hybrid deep-learning/iterative method.