Deep learning-based relative stopping power mapping generation with cone-beam CT in proton radiation therapy

2020 ◽  
Author(s):  
Tonghe Wang ◽  
Joseph Harms ◽  
Yang Lei ◽  
Beth Ghavidel ◽  
William Stokes ◽  
...  
Keyword(s):  
Radiation Therapy ◽  
Deep Learning ◽  
Cone Beam Ct ◽  
Stopping Power ◽  
Cone Beam ◽  
Proton Radiation ◽  
Proton Radiation Therapy ◽  
Power Mapping

High quality proton portal imaging using deep learning for proton radiation therapy: a phantom study

2020 ◽  
Vol 6 (3) ◽  
pp. 035029
Author(s):  
Serdar Charyyev ◽  
Yang Lei ◽  
Joseph Harms ◽  
Bree Eaton ◽  
Mark McDonald ◽  
...  
Keyword(s):  
Radiation Therapy ◽  
Deep Learning ◽  
Phantom Study ◽  
Proton Radiation ◽  
High Quality ◽  
Portal Imaging ◽  
Proton Radiation Therapy

Cone‐beam CT‐derived relative stopping power map generation via deep learning for proton radiotherapy

2020 ◽  
Vol 47 (9) ◽  
pp. 4416-4427
Author(s):  
Joseph Harms ◽  
Yang Lei ◽  
Tonghe Wang ◽  
Mark McDonald ◽  
Beth Ghavidel ◽  
...  
Keyword(s):  
Deep Learning ◽  
Cone Beam Ct ◽  
Stopping Power ◽  
Cone Beam ◽  
Proton Radiotherapy ◽  
Map Generation

scholarly journals Learning-Based Stopping Power Mapping on Dual-Energy CT for Proton Radiation Therapy

2021 ◽  
Vol 7 (3) ◽  
pp. 46-60
Author(s):  
Tonghe Wang ◽  
Yang Lei ◽  
Joseph Harms ◽  
Beth Ghavidel ◽  
Liyong Lin ◽  
...  
Keyword(s):  
Radiation Therapy ◽  
Ground Truth ◽  
Dual Energy ◽  
Stopping Power ◽  
Whole Body ◽  
Chemical Compositions ◽  
Proton Radiation ◽  
Dose Volume Histogram ◽  
Proton Radiation Therapy ◽  
Dose Volume

Abstract Purpose Dual-energy computed tomography (DECT) has been used to derive relative stopping power (RSP) maps by obtaining the energy dependence of photon interactions. The DECT-derived RSP maps could potentially be compromised by image noise levels and the severity of artifacts when using physics-based mapping techniques. This work presents a noise-robust learning-based method to predict RSP maps from DECT for proton radiation therapy. Materials and Methods The proposed method uses a residual attention cycle-consistent generative adversarial network to bring DECT-to-RSP mapping close to a 1-to-1 mapping by introducing an inverse RSP-to-DECT mapping. To evaluate the proposed method, we retrospectively investigated 20 head-and-neck cancer patients with DECT scans in proton radiation therapy simulation. Ground truth RSP values were assigned by calculation based on chemical compositions and acted as learning targets in the training process for DECT datasets; they were evaluated against results from the proposed method using a leave-one-out cross-validation strategy. Results The predicted RSP maps showed an average normalized mean square error of 2.83% across the whole body volume and an average mean error less than 3% in all volumes of interest. With additional simulated noise added in DECT datasets, the proposed method still maintained a comparable performance, while the physics-based stoichiometric method suffered degraded inaccuracy from increased noise level. The average differences from ground truth in dose volume histogram metrics for clinical target volumes were less than 0.2 Gy for D95% and Dmax with no statistical significance. Maximum difference in dose volume histogram metrics of organs at risk was around 1 Gy on average. Conclusion These results strongly indicate the high accuracy of RSP maps predicted by our machine-learning–based method and show its potential feasibility for proton treatment planning and dose calculation.

Single Room Proton Radiation Therapy

2012 ◽  
Vol 84 (3) ◽  
pp. S835-S836
Author(s):  
E.E. Klein ◽  
C. Block ◽  
B. Pierburg ◽  
J. Bradley
Keyword(s):  
Radiation Therapy ◽  
Proton Radiation ◽  
Proton Radiation Therapy ◽  
Single Room
Sign in / Sign up

Export Citation Format

Share Document