Computed tomography super-resolution using deep convolutional neural network

Abstract

The objective of this study is to develop a convolutional neural network (CNN) for computed tomography (CT) image super-resolution. The network learns an end-to-end mapping between low (thick-slice thickness) and high (thin-slice thickness) resolution images using the modified U-Net. To verify the proposed method, we train and test the CNN using axially averaged data of existing thin-slice CT images as input and their middle slice as the label. Fifty-two CT studies are used as the CNN training set, and 13 CT studies are used as the test set. We perform five-fold cross-validation to confirm the performance consistency. Because all input and output images are used in two-dimensional slice format, the total number of slices for training the CNN is 7670. We assess the performance of the proposed method with respect to the resolution and contrast, as well as the noise properties. The CNN generates output images that are virtually equivalent to the ground truth. The most remarkable image-recovery improvement by the CNN is deblurring of boundaries of bone structures and air cavities. The CNN output yields an approximately 10% higher peak signal-to-noise ratio and lower normalized root mean square error than the input (thicker slices). The CNN output noise level is lower than the ground truth and equivalent to the iterative image reconstruction result. The proposed deep learning method is useful for both super-resolution and de-noising.

Generation of PET Attenuation Map for Whole-Body Time-of-Flight 18F-FDG PET/MRI Using a Deep Neural Network Trained with Simultaneously Reconstructed Activity and Attenuation Maps

“…The peak signal-to-noise ratio and normalized root-mean-square error values were calculated as additional image quality metrics to quantify the similarity of attenuation and activity maps (30).…”

Current Applications, Opportunities, and Limitations of AI for 3D Imaging in Dental Research and Practice

“…Table 1 exhibits the included studies regarding the use of AI for 3D imaging in DMFR. These studies focused on three main applications, including automated diagnosis of dental and maxillofacial diseases [16][17][18][19][20][28][29][30][31][32], localization of anatomical landmarks for orthodontic and orthognathic treatment planning [21,22,[33][34][35], and improvement of image quality [23,36].…”

“…The balance between the radiation dose and CT image resolution is the biggest concern for radiologists. To address this issue, Park et al proposed a deep learning algorithm to enhance the thick-slice CT image resolution similar to that of a thin slice [36]. It is reported that the noise level of the enhanced CT images is even lower than the original images.…”

Anisotropic Super Resolution In Prostate Mri Using Super Resolution Generative Adversarial Networks

“…[5] and Park et.al. [6] use convolutional neural networks to achieve SR. While the previous works attack the SR problem from the 2D perspective, others such as Chaudhari et.al.…”