Hoyeon Lee

IMAGING DOSE OF HUMAN ORGANS FROM kV-CBCT IN IMAGE-GUIDED RADIATION THERAPY

This study investigates dose distribution due to kV cone-beam computed tomography (CBCT) for the patients undergoing CBCT-based image-guided radiation therapy (IGRT). The kV-CBCT provides an efficient image-guidance tool for acquiring the latest volumetric image of a patients anatomy, and has been being routinely used in clinics for an accurate treatment setup. Imaging radiation doses resulting from six different acquisition protocols of the on-board imager (OBI) were calculated using a Geant4 Application for Tomographic Emission (GATE) Monte Carlo simulation toolkit, and the absorbed doses by various organs were analyzed for the adult and pediatric numerical XCAT phantoms in this study. The calculated organ doses range from 0.1 to 24.1 mGy in the adult phantom, and from 0.1 to 36.8 mGy in the pediatric one. The imaging organ doses to the pediatric phantom turn out to be consistently higher than those to the adult phantom. It is believed that our results would provide reliable data to the clinicians for their making better decisions on CBCT scanning options and would also provide a platform for developing a new kV-CBCT scanning protocol in conjunction with a low-dose capability.

Optimal dose reduction algorithm using an attenuation-based tube current modulation method for cone-beam CT imaging

To reduce the radiation dose given to patients, a tube current modulation (TCM) method has been widely used in diagnostic CT systems. However, the TCM method has not yet been applied to a kV-CBCT system on a LINAC machine. The purpose of this study is to investigate if a TCM method would be desirable in a kV-CBCT system for image-guided radiation therapy (IGRT) or not. We have developed an attenuation–based TCM method using prior knowledge from planning CT images of patients. The TCM method can provide optimized dose reductions without degrading image quality for kV-CBCT imaging. Here, we investigate whether or not our suggested TCM method is desirable to use in kV-CBCT systems to confirm and revise the exact position of a patient for IGRT. Patients go through diagnostic CT scans for RT planning; therefore, using information from prior CT images can enable estimations of the total X-ray attenuation through a patient’s body in a CBCT setting for radiation treatment. Having this planning CT image allows to use the proposed TCM method in RT. The proposed TCM method provides a minimal amount of current for each projection, as well as total current, required to reconstruct the current modulated CBCT image with an image quality similar to that of CBCT. After applying a calculated TCM current for each projection, projection images were acquired and the current modulated CBCT image was reconstructed using a FDK algorithm. To validate the proposed approach, we used a numerical XCAT phantom and a real ATOM phantom and evaluated the performance of the proposed method via visual and quantitative image quality metrics. The organ dose due to imaging radiation was calculated in both cases and compared using the GATE simulation toolkit. As shown in the quantitative evaluation, normalized noise and SSIM values of the TCM were similar to those of conventional CBCT images. In addition, the proposed TCM method yielded comparable image quality to that of conventional CBCT images for both simulations and experimental studies as organ doses were decreased. We have successfully demonstrated the feasibility and dosimetric merit of a prototypical TCM method for kV-CBCT via simulations and experimental study. The results indicate that the proposed TCM method and overall framework can be a viable option for CBCT imaging that utilizes an optimal dose reduction without degrading image quality. Thus, this method reduces the probability for side effects due to radiation exposure.

Sinogram synthesis using convolutional-neural-network for sparsely view-sampled CT.

Reducing the number of projections in computed tomography (CT) has been exploited as a low-dose option in conjunction with advanced iterative image reconstruction algorithms. While such iterative image reconstruction methods do provide useful images and valuable insights of the inverse imaging problems, it is an intriguing issue whether missing view projection data in the sinogram can be successfully recovered. There have been reported several approaches to interpolating the missing sinogram data. Deep-learning based super-resolution techniques in the field of natural image enhancement have been recently introduced and showed promising results. Inspired by the super-resolution techniques, we have earlier proposed a sinogram inpainting method that uses a convolutional-neural-network for sparsely viewsampled CT. Despite of the encouraging initial results, our previously proposed method had two drawbacks. The measured sinogram was contaminated in the process of filling the missing sinogram by the deep learning network. Additionally, the sum of the interpolated sinogram in the direction of detector row at each view angle was not preserved. In this study, we improve our previously developed deep-learning based sinogram synthesis method by adding new layers and modifying the size of receptive field in the deep learning network to overcome the above limitations. From the quantitative evaluations on the image accuracy and quality using real patients’ CT images, we show that the new approach synthesizes more accurate sinogram and thus leads to higher quality of CT image than the previous one.

Investigation on Beam-Blocker-Based Scatter Correction Method for Improving CT Number Accuracy

Cone-beam computed tomography (CBCT) is gaining widespread use in various medical and industrial applications but suffers from substantially larger amount of scatter than that in the conventional diagnostic CT resulting in relatively poor image quality. Various methods that can reduce and/or correct for the scatter in the CBCT have therefore been developed. Scatter correction method that uses a beam-blocker has been considered a direct measurement-based approach providing accurate scatter estimation from the data in the shadows of the beam-blocker. To the best of our knowledge, there has been no record reporting the significance of the scatter from the beam-blocker itself in such correction methods. In this paper, we identified the scatter from the beam-blocker that is detected in the object-free projection data investigated its influence on the image accuracy of CBCT reconstructed images, and developed a scatter correction scheme that takes care of this scatter as well as the scatter from the scanned object.

Scatter correction in CBCT with an offset detector through a deconvolution method using data consistency

Our earlier work has demonstrated that the data consistency condition can be used as a criterion for scatter kernel optimization in deconvolution methods in a full-fan mode cone-beam CT [1]. However, this scheme cannot be directly applied to CBCT system with an offset detector (half-fan mode) because of transverse data truncation in projections. In this study, we proposed a modified scheme of the scatter kernel optimization method that can be used in a half-fan mode cone-beam CT, and have successfully shown its feasibility. Using the first-reconstructed volume image from half-fan projection data, we acquired full-fan projection data by forward projection synthesis. The synthesized full-fan projections were partly used to fill the truncated regions in the half-fan data. By doing so, we were able to utilize the existing data consistency-driven scatter kernel optimization method. The proposed method was validated by a simulation study using the XCAT numerical phantom and also by an experimental study using the ACS head phantom.