Comparison of CBCT‐based dose calculation methods in head and neck cancer radiotherapy: from Hounsfield unit to density calibration curve to deep learning

Published on Oct 1, 2020in Medical Physics3.317
· DOI :10.1002/MP.14387
A. Barateau5
Estimated H-index: 5
(French Institute of Health and Medical Research),
Renaud de Crevoisier28
Estimated H-index: 28
(French Institute of Health and Medical Research)
+ 8 AuthorsCaroline Lafond15
Estimated H-index: 15
Sources
Abstract
PURPOSE Anatomical variations occur during head and neck (H&N) radiotherapy treatment. kV cone-beam computed tomography (CBCT) images can be used for daily dose monitoring to assess dose variations owing to anatomic changes. Deep learning methods (DLMs) have recently been proposed to generate pseudo-CT (pCT) from CBCT to perform dose calculation. This study aims to evaluate the accuracy of a DLM and to compare this method with three existing methods of dose calculation from CBCT in H&N cancer radiotherapy. METHODS Forty-four patients received VMAT for H&N cancer (70-63-56 Gy). For each patient, reference CT (Bigbore, Philips) and CBCT images (XVI, Elekta) were acquired. The DLM was based on a generative adversarial network. The three compared methods were: i) a method using a density to Hounsfield Unit (HU) relation from phantom CBCT image (HU-D curve method), ii) a water-air-bone density assignment method (DAM), and iii) a method using deformable image registration (DIR). The imaging endpoints were the mean absolute error (MAE) and mean error (ME) of HU from pCT and reference CT (CTref ). The dosimetric endpoints were dose discrepancies and 3D gamma analyses (local, 2%/2mm, 30% dose threshold). Dose discrepancies were defined as the mean absolute differences between DVHs calculated from the CTref and pCT of each method. RESULTS In the entire body, the MAEs and MEs of the DLM, HU-D curve method, DAM, and DIR method were 82.4 and 17.1 HU, 266.6 and 208.9 HU, 113.2 and 14.2 HU, and 95.5 and -36.6 HU, respectively. The MAE obtained using the DLM differed significantly from those of other methods (Wilcoxon, p < 0.05). The DLM dose discrepancies were 7 ± 8 cGy (maximum = 44 cGy) for the ipsilateral parotid gland Dmean and 5 ± 6 cGy (max = 26 cGy) for the contralateral parotid gland mean dose (Dmean ). For the parotid gland Dmean , no significant dose difference was observed between the DLM and other methods. The mean 3D gamma pass-rate ± standard deviation was 98.1 ± 1.2%, 91.0 ± 5.3%, 97.9 ± 1.6%, and 98.8 ± 0.7% for the DLM, HU-D method, DAM, and DIR method, respectively. The gamma pass-rates and mean gamma results of the HU-D curve method, DAM, and DIR method differed significantly from those of the DLM. The mean calculation time to generate one pCT was 30 s for the deep learning and DIR methods. CONCLUSIONS For H&N radiotherapy, DIR method and DLM appears as the most appealing CBCT-based dose calculation methods among the four methods in terms of dose accuracy as well as calculation time. Using the DIR method or DLM with CBCT images enables dose monitoring in the parotid glands during the treatment course and may be used to trigger replanning.
📖 Papers frequently viewed together
9 Authors (A. Barateau, ..., Cedrik Lafond)
References57
Newest
#1Yingzi Liu (Emory University)H-Index: 12
#1Yingzi Liu (Emory University)H-Index: 9
Last. Xiaofeng Yang (Emory University)H-Index: 86
view all 9 authors...
PURPOSE: Current clinical application of CBCT is limited to patient setup. Imaging artifacts and Hounsfield unit (HU) inaccuracy make the process of CBCT-based adaptive planning presently impractical. In this study, we developed a deep-learning-based approach to improve CBCT image quality and HU accuracy for potential extended clinical use in CBCT-guided pancreatic adaptive radiotherapy. METHODS: Thirty patients previously treated with pancreas SBRT were included. The CBCT acquired prior to the ...
15 CitationsSource
#1Mengke Qi (Southern Medical University)H-Index: 3
#2Yongbao Li (SYSU: Sun Yat-sen University)H-Index: 4
Last. Ting Song (Southern Medical University)H-Index: 5
view all 11 authors...
PURPOSE: The purpose of this study is to investigate the effect of different magnetic resonance (MR) sequences on the accuracy of deep learning-based synthetic computed tomography (sCT) generation in the complex head and neck region. METHODS: Four sequences of MR images (T1, T2, T1C, and T1DixonC-water) were collected from 45 patients with nasopharyngeal carcinoma. Seven conditional generative adversarial network (cGAN) models were trained with different sequences (single channel) and different ...
12 CitationsSource
#1Lukas Schröder (NKI-AVL: Netherlands Cancer Institute)H-Index: 1
#2Uros Stankovic (NKI-AVL: Netherlands Cancer Institute)H-Index: 6
Last. Jan-Jakob Sonke (NKI-AVL: Netherlands Cancer Institute)H-Index: 55
view all 3 authors...
Purpose: The goal of this study was to investigate the stability of a phantom-based Hounsfield unit (HU) calibration for cone beam CT (CBCT). Methods: Three consecutive scans of a large phantom configuration and a small phantom configuration were acquired and reconstructed with a uniform scatter correction method. The CBCT gray values of the phantom inserts were measured and the three values of each insert averaged. The linear calibration curve was determined and its slope and intercept were eva...
1 CitationsSource
#1Liyuan Chen (UTSW: University of Texas Southwestern Medical Center)H-Index: 11
#2Xiao Liang (UTSW: University of Texas Southwestern Medical Center)H-Index: 4
Last. Jing Wang (UTSW: University of Texas Southwestern Medical Center)H-Index: 106
view all 5 authors...
PURPOSE: Cone-beam computed tomography (CBCT) scanning is used daily or weekly (i.e., on-treatment CBCT) for accurate patient setup in image-guided radiotherapy. However, inaccuracy of CT numbers prevents CBCT from performing advanced tasks such as dose calculation and treatment planning. Motivated by the promising performance of deep learning in medical imaging, we propose a deep U-net-based approach that synthesizes CT-like images with accurate numbers from planning CT, while keeping the same ...
14 CitationsSource
#1Satoshi Kida (UTokyo: University of Tokyo)H-Index: 32
#2Shizuo Kaji (Kyushu University)H-Index: 8
Last. Keiichi Nakagawa (UTokyo: University of Tokyo)H-Index: 46
view all 9 authors...
PURPOSE: Cone-beam computed tomography (CBCT) offers advantages over conventional fan-beam CT in that it requires a shorter time and less exposure to obtain images. However, CBCT images suffer from low soft-tissue contrast, noise, and artifacts compared to conventional fan-beam CT images. Therefore, it is essential to improve the image quality of CBCT. METHODS: In this paper, we propose a synthetic approach to translate CBCT images with deep neural networks. Our method requires only unpaired and...
14 CitationsSource
#1Xin Yi (U of S: University of Saskatchewan)H-Index: 6
#2Ekta Walia (U of S: University of Saskatchewan)H-Index: 20
Last. Paul Babyn (U of S: University of Saskatchewan)H-Index: 11
view all 3 authors...
Abstract Generative adversarial networks have gained a lot of attention in the computer vision community due to their capability of data generation without explicitly modelling the probability density function. The adversarial loss brought by the discriminator provides a clever way of incorporating unlabeled samples into training and imposing higher order consistency. This has proven to be useful in many cases, such as domain adaptation, data augmentation, and image-to-image translation. These p...
338 CitationsSource
In presence of inter-fractional anatomical changes, clinical benefits are anticipated from image-guided adaptive radiotherapy. Nowadays, cone-beam CT (CBCT) imaging is mostly utilized during pre-treatment imaging for position verification. Due to various artifacts, image quality is typically not sufficient for photon or proton dose calculation, thus demanding accurate CBCT correction, as potentially provided by deep learning techniques. This work aimed at investigating the feasibility of utilizi...
22 CitationsSource
#1Ruba Alkadi (Khalifa University)H-Index: 1
#2Fatma Taher (Khalifa University)H-Index: 11
Last. Naoufel Werghi (Khalifa University)H-Index: 20
view all 4 authors...
We address the problem of prostate lesion detection, localization, and segmentation in T2W magnetic resonance (MR) images. We train a deep convolutional encoder-decoder architecture to simultaneously segment the prostate, its anatomical structure, and the malignant lesions. To incorporate the 3D contextual spatial information provided by the MRI series, we propose a novel 3D sliding window approach, which preserves the 2D domain complexity while exploiting 3D information. Experiments on data fro...
30 CitationsSource
#1Anna M. Dinkla (UU: Utrecht University)H-Index: 4
#2Mateusz C. Florkow (UU: Utrecht University)H-Index: 3
Last. Peter R. Seevinck (UU: Utrecht University)H-Index: 22
view all 10 authors...
Purpose: To develop and evaluate a patch-based convolutional neural network (CNN) to generate synthetic computed tomography (sCT) images for magnetic resonance (MR)-only workflow for radiotherapy of head and neck tumors. A patch-based deep learning method was chosen to improve robustness to abnormal anatomies caused by large tumors, surgical excisions, or dental artifacts. In this study, we evaluate whether the generated sCT images enable accurate MR-based dose calculations in the head and neck ...
20 CitationsSource
#1Joseph Harms (Emory University)H-Index: 7
#2Yang Lei (Emory University)H-Index: 22
Last. Xiaofeng Yang (Emory University)H-Index: 86
view all 9 authors...
PURPOSE: The incorporation of cone-beam computed tomography (CBCT) has allowed for enhanced image-guided radiation therapy. While CBCT allows for daily 3D imaging, images suffer from severe artifacts, limiting the clinical potential of CBCT. In this work, a deep learning-based method for generating high quality corrected CBCT (CCBCT) images is proposed. METHODS: The proposed method integrates a residual block concept into a cycle-consistent adversarial network (cycle-GAN) framework, called res-c...
51 CitationsSource
Cited By2
Newest
#2Matteo MasperoH-Index: 11
Last. Joao Seco (DKFZ: German Cancer Research Center)H-Index: 29
view all 4 authors...
Recently, deep learning (DL)-based methods for the generation of synthetic computed tomography (sCT) have received significant research attention as an alternative to classical ones. We present here a systematic review of these methods by grouping them into three categories, according to their clinical applications: I) to replace CT in magnetic resonance (MR)-based treatment planning, II) facilitate cone-beam computed tomography (CBCT)-based image-guided adaptive radiotherapy, and III) derive at...
#1Miriam Eckl (Heidelberg University)H-Index: 2
#2Lea Hoppen (Heidelberg University)H-Index: 1
Last. Jens Fleckenstein (Heidelberg University)H-Index: 13
view all 8 authors...
Abstract Purpose Image-guided radiation therapy could benefit from implementing adaptive radiation therapy (ART) techniques. A cycle-generative adversarial network (cycle-GAN)-based cone-beam computed tomography (CBCT)-to-synthetic CT (sCT) conversion algorithm was evaluated regarding image quality, image segmentation and dosimetric accuracy for head and neck (H&N), thoracic and pelvic body regions. Methods Using a cycle-GAN, three body site-specific models were priorly trained with independent ...
Source