Deep learning based synthetic CT from cone beam CT generation for abdominal paediatric radiotherapy.

Researchers

Journal

Modalities

Models

Abstract

Adaptive radiotherapy workflows require images with the quality of computed tomography
(CT) for re-calculation and re-optimisation of radiation doses. In this work we aim to improve quality of
cone beam CT (CBCT) images for dose calculation using deep learning.
Approach: We propose a novel framework for CBCT-to-CT synthesis using cycle-consistent Generative Adversarial Networks (cycleGANs). The framework was tailored for paediatric abdominal patients, a
challenging application due to the inter-fractional variability in bowel filling and smaller patient numbers.
We introduced the concept of global residuals only learning to the networks and modified the cycleGAN
loss function to explicitly promote structural consistency between source and synthetic images. Finally, to
compensate for the anatomical variability and address the difficulties in collecting large datasets in the paediatric population, we applied a smart 2D slice selection based on the common field-of-view across
the dataset (abdomen). This acted as a weakly paired data approach that allowed us to take advantage of
scans from patients treated for a variety of malignancies (thoracic-abdominal-pelvic) for training
purposes. We first optimised the proposed framework and benchmarked its performance on a
development dataset. Later, a comprehensive quantitative evaluation was performed on an unseen dataset, which included calculating global image similarity metrics, segmentation-based measures and
proton therapy-specific metrics.
Main results: We found improved performance, compared to a baseline implementation, on imagesimilarity
metrics such as Mean Absolute Error calculated for a matched virtual CT (55.0±16.6 proposed, 58.9±16.8 baseline). There was also a higher level of structural agreement for gastrointestinal gas between source and synthetic images measured through dice similarity overlap (0.872±0.053 proposed, 0.846±0.052 baseline). Differences found in water-equivalent thickness metrics were also smaller for
our method (3.3±2.4% proposed, 3.7±2.8% baseline).
Significance: Our findings indicate that our innovations to the cycleGAN framework improved the quality
and structure consistency of the synthetic CTs generated.Creative Commons Attribution license.