Advanced 4-dimensional cone-beam computed tomography reconstruction by combining motion estimation, motion-compensated reconstruction, biomechanical modeling and deep learning

Abstract

4-Dimensional cone-beam computed tomography (4D-CBCT) offers several key advantages over conventional 3D-CBCT in moving target localization/delineation, structure de-blurring, target motion tracking, treatment dose accumulation and adaptive radiation therapy. However, the use of the 4D-CBCT in current radiation therapy practices has been limited, mostly due to its sub-optimal image quality from limited angular sampling of conebeam projections. In this study, we summarized the recent developments of 4D-CBCT reconstruction techniques for image quality improvement, and introduced our developments of a new 4D-CBCT reconstruction technique which features simultaneous motion estimation and image reconstruction (SMEIR). Based on the original SMEIR scheme, biomechanical modeling-guided SMEIR (SMEIR-Bio) was introduced to further improve the reconstruction accuracy of fine details in lung 4D-CBCTs. To improve the efficiency of reconstruction, we recently developed a U-net-based deformation-vector-field (DVF) optimization technique to leverage a population-based deep learning scheme to improve the accuracy of intra-lung DVFs (SMEIR-Unet), without explicit biomechanical modeling. Details of each of the SMEIR, SMEIR-Bio and SMEIR-Unet techniques were included in this study, along with the corresponding results comparing the reconstruction accuracy in terms of CBCT images and the DVFs. We also discussed the application prospects of the SMEIR-type techniques in image-guided radiation therapy and adaptive radiation therapy, and presented potential schemes on future developments to achieve faster and more accurate 4D-CBCT imaging.

Real‐time liver motion estimation via deep learning‐based angle‐agnostic X‐ray imaging

“…The details of the DL‐based liver biomechanical model can be found in Shao et al 30 . In addition, to supplement the single X‐ray projection which has a limited field‐of‐view (FOV), optical body surface imaging can be combined with the X‐ray imaging to estimate liver boundary motion 30,46 . We integrated X360 into a previously developed framework of liver tumor motion tracking 30 (Figure 3) that cascades the DL‐based surface imaging, X‐ray imaging, and biomechanical modeling.…”

“…The details of the DL-based liver biomechanical model can be found in Shao et al 30 In addition, to supplement the single X-ray projection which has a limited field-of -view (FOV), optical body surface imaging can be combined with the X-ray imaging to estimate liver boundary motion. 30,46 We integrated X360 into a previously developed framework of liver tumor motion tracking 30 (Figure 3) that cascades the DL-based surface imaging, X-ray imaging, and biomechanical modeling. The resulting Surf-X360-Bio framework first estimated the liver boundary motion by a DL-based surface imaging model (Surf), which uses the external-internal motion correlation to infer liver boundary motion from external body surface motion.…”

Motion compensated cone-beam CT reconstruction using an a priori motion model from CT simulation: a pilot study

“…While this study provided a great improvement to CBCT reconstruction, its limitations included sensitivity to the accuracy of the initial motion model and to breathing irregularities. Several studies were subsequently published using biomechanical modeling or deep learning (Zhang et al 2019) to improve the registration accuracy of SMEIR-based approaches. Additionally, many other groups developed similar approaches and alternatives to SMEIR with various implementations of simultaneous modeling and reconstruction (Liu et al 2015, Riblett et al 2018, Sauppe et al 2018, Jailin et al 2021, Mo et al 2021.…”

Automatic liver tumor localization using deep learning‐based liver boundary motion estimation and biomechanical modeling (DL‐Bio)

“…The motion model, which often takes the form of deformation‐vector‐fields (DVFs), allows automatic tumor propagation and localization even if such tasks may be visually challenging. Out of the many deformable registration methods, 2D‐3D deformable registration has proved effective to estimate on‐board CBCTs from prior CT/CBCT images, especially for scenarios where the number of available cone‐beam projections is insufficient to reconstruct a high‐quality CBCT for direct 3D‐3D registration 28–32 . The 2D‐3D deformable registration method estimates DVFs between a prior reference image and new on‐board target images, by matching the pixel‐wise intensities between 2D digitally reconstructed‐radiographs (DRRs) of the deformed 3D prior image and acquired 2D on‐board projections.…”

“…Out of the many deformable registration methods, 2D-3D deformable registration has proved effective to estimate on-board CBCTs from prior CT/CBCT images, especially for scenarios where the number of available cone-beam projections is insufficient to reconstruct a high-quality CBCT for direct 3D-3D registration. [28][29][30][31][32] The 2D-3D deformable registration method estimates DVFs between a prior reference image and new onboard target images, by matching the pixel-wise intensities between 2D digitally reconstructed-radiographs (DRRs) of the deformed 3D prior image and acquired 2D on-board projections. It can be especially effective toward scenarios of limited-angle or sparse-view F I G U R E 1 A cone-beam computed tomography (CBCT) projection for the liver site.…”