Paired cycle‐GAN‐based image correction for quantitative cone‐beam computed tomography

Abstract

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‐cycle GAN, to learn a mapping between CBCT images and paired planning CT images. Compared with a GAN, a cycle‐GAN includes an inverse transformation from CBCT to CT images, which constrains the model by forcing calculation of both a CCBCT and a synthetic CBCT. A fully convolution neural network with residual blocks is used in the generator to enable end‐to‐end CBCT‐to‐CT transformations. The proposed algorithm was evaluated using 24 sets of patient data in the brain and 20 sets of patient data in the pelvis. The mean absolute error (MAE), peak signal‐to‐noise ratio (PSNR), normalized cross‐correlation (NCC) indices, and spatial non‐uniformity (SNU) were used to quantify the correction accuracy of the proposed algorithm. The proposed method is compared to both a conventional scatter correction and another machine learning‐based CBCT correction method. Results Overall, the MAE, PSNR, NCC, and SNU were 13.0 HU, 37.5 dB, 0.99, and 0.05 in the brain, 16.1 HU, 30.7 dB, 0.98, and 0.09 in the pelvis for the proposed method, improvements of 45%, 16%, 1%, and 93% in the brain, and 71%, 38%, 2%, and 65% in the pelvis, over the CBCT image. The proposed method showed superior image quality as compared to the scatter correction method, reducing noise and artifact severity. The proposed method produced images with less noise and artifacts than the comparison machine learning‐based method. Conclusions The authors have developed a novel deep learning‐based method to generate high‐quality corrected CBCT images. The proposed method increases onboard CBCT image quality, making it comparable to that of the planning CT. With further evaluation and clinical implementation, this method could lead to quantitative adaptive radiation therapy.

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

“…In fact, pCT generated from CBCT are used to monitor delivered doses or to estimate the cumulative delivered dose during the treatment course in the context of dose-guided adaptive radiotherapy. Studies using DL for pCT generation from CBCT are scarce for brain, 55 H&N, [13][14][15] pancreas, 38 or prostate cancer. [35][36][37]55,56 The studies involved an imaging analysis (pCT vs. reference CT), but only half of them evaluated the dose accuracy.…”

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

“…The contrast between bone and soft tissue is significantly larger than the contrast between fat and muscle. In previous cycle-GAN methods proposed for medical imaging, where the loss function has been intensity based, for example, mean absolute error (MAE) used by Armanious et al, 39 or structure based, for example, gradient difference (GD) used in our previous work, 36 it cannot be guaranteed that the trained cycle-GAN model will be able to differentiate tissues with such similar intensities. Thus, in this study, a statistical matching-based loss was integrated into generator's loss function.…”

“…38 The proposed network has been described in detail in a previous publication from our group. 36 In this previous work, the network was used for artifact correction, whereas in this work, the network is used for both artifact correction and domain mapping from CBCT to RSP. Here, we briefly summarize the workflow, which can be seen in Fig.…”

Head and neck multi‐organ auto‐segmentation on CT images aided by synthetic MRI

“…Recently, some studies on CBCT-to-synthetic CT (sCT) imaging transfer using cycleGAN have been published. [40][41][42] sCT + sMR based DPN has the potential to be used for on-line adaptive segmentation.…”