Uncertainty-aware MR-based CT synthesis for robust proton therapy planning of brain tumour.

Li, Xia; Bellotti, Renato; Meier, Gabriel; Bachtiary, Barbara; Weber, Damien; Lomax, Antony; Buhmann, Joachim; Zhang, Ye (2024). Uncertainty-aware MR-based CT synthesis for robust proton therapy planning of brain tumour. Radiotherapy and oncology : journal of the European Society for Therapeutic Radiology and Oncology, 191(110056), p. 110056. Elsevier Scientific Publ. Ireland 10.1016/j.radonc.2023.110056

Preview

Text 1-s2.0-S0167814023093635-main.pdf - Accepted Version Available under License Creative Commons: Attribution (CC-BY). Download (4MB) | Preview

BACKGROUND AND PURPOSE

Deep learning techniques excel in MR-based CT synthesis, but missing uncertainty prediction limits its clinical use in proton therapy. We developed an uncertainty-aware framework and evaluated its efficiency in robust proton planning.

MATERIALS AND METHODS

A conditional generative-adversarial network was trained on 64 brain tumour patients with paired MR-CT images to generate synthetic CTs (sCT) from combined T1-T2 MRs of three orthogonal planes. A Bayesian neural network predicts Laplacian distributions for all voxels with parameters (μ, b). A robust proton plan was optimized using three sCTs of μ and μ±b. The dosimetric differences between the plan from sCT (sPlan) and the recalculated plan (rPlan) on planning CT (pCT) were quantified for each patient. The uncertainty-aware robust plan was compared to conventional robust (global ±3%) and non-robust plans.

RESULTS

In 8-fold cross-validation, sCT-pCT image differences (Mean-Absolute-Error) were 80.84±9.84HU (body), 35.78±6.07HU (soft tissues) and 221.88±31.69HU (bones), with Dice scores of 90.33±2.43%, 95.13±0.80%, and 85.53±4.16%, respectively. The uncertainty distribution positively correlated with absolute prediction error (Correlation Coefficient: 0.62±0.01). The uncertainty-conditioned robust optimisation improved the rPlan-sPlan agreement, e.g., D95 absolute difference (CTV) was 1.10±1.24% compared to conventional (1.64±2.71%) and non-robust (2.08±2.96 %) optimisation. This trend was consistent across all target and organs-at-risk indexes.

CONCLUSION

The enhanced framework incorporates 3D uncertainty prediction and generates high-quality sCTs from MR images. The framework also facilitates conditioned robust optimisation, bolstering proton plan robustness against network prediction errors. The innovative feature of uncertainty visualisation and robust analyses contribute to evaluating sCT clinical utility for individual patients.

Interest & Impact

Downloads

21 since deposited on 18 Dec 2023
21 in the past 12 months

Citations

5 Citations in Web of Science ®
6 Citations in Scopus

Search

in Google Scholar™

Services

Actions (login required)

Edit item

Actions (login required)

Edit item