Feasibility of image quality improvement for high-speed CBCT imaging using deep convolutional neural network for...

High-speed cone-beam computed tomography (CBCT) scan for image-guided radiotherapy (IGRT) can reduce both the scan time and the exposure dose. However, it causes noise and artifacts in the reconstructed images due to the lower number of acquired projection data. The purpose of this study is to improve the image quality of high-speed CBCT using a deep convolutional neural network (DCNN).

CBCT images of 36 prostate cancer patients were selected. The CBCT images acquired at normal scan speed were defined as CBCT_100%. Simulated high-speed CBCT images acquired at twofold and fourfold scan speed were created, which were defined as CBCT_50% and CBCT_25%, respectively. The image quality of the CBCT_50% was treated as the requirement for IGRT in this study because previous studies reported that its image is sufficient with respect to IGRT. The DCNN model was trained to learn direct mapping from CBCT_25% to the corresponding CBCT_100%. The performance of the DCNN model was evaluated using the sixfold cross-validation method. CBCT images generated by DCNN (CBCT_25%+DCNN) were evaluated for voxel value accuracy and image quality.

The DCNN model can process CBCT_25% of a new patient within 0.06 s/slice. The CBCT_25%+DCNN was comparable to the CBCT_50% in terms of both voxel value accuracy and image quality.

We developed a DCNN model to remove noise and artifacts from high-speed CBCT. We emphasize that it is possible to reduce exposure to one quarter and to increase the CBCT scan speed by a factor of four.