Stereo-Correlation and Noise-Distribution Aware ResVoxGAN for Dense Slices Reconstruction and Noise Reduction in Thick Low-Dose CT

The low-dose computed tomography (CT) scan with thick slice thickness (3 mm) dramatically improves the imaging efficiency and reduces the radiation risk in clinical. However, the low-dose CT acquisition inherently compromises the signal-to-noise ratio, and the sparse sampled thick slices poorly reproduce the coronal/sagittal anatomy. We propose a Residual Voxel Generative Adversarial Nets (ResVoxGAN), the first powerful work to densely reconstruct slices into the thin thickness (1 mm), and simultaneously denoise the CT image into the more readable pattern, directly from the widely accessible thick low-dose CT. The framework is achieved in a voxel-wise conditional GAN constituted by the following: (1) a generator is composed of consecutive 3D multi-scale residual blocks that richly extracts multi-scale stereo feature for fine-granted and latent spatial structure mining from the noisy volume, and a followed Subpixel Convnet further interpretively reconstructs dense slices from the features for high-resolution and denoising volume; (2) a stereo-correlation constraint elegantly penalizes gradient deviation in voxel adjacent region (i.e., 3D 26-neighborhoods) to guide structural detail, together with a image-expression constraint on perceptual feature representations transformed from a pretrained deep convolution autoencoder to keep scene content; and (3) a pair of coupled discriminators advantageously fuse the prior-knowledge from the thick low-dose CT with the generated image and residual noise via self-learning to drive the generation towards into both realistic anatomic structure distribution and valid noise-reduction distribution. The experiment validated on Mayo dataset shows that the ResVoxGAN successfully reconstruct the low-dose CT of 3 mm thickness into 1 mm, and meanwhile keeps the with peak signal to noise ratio of 40.80 for noise reduction, and structural similarity index of 0.918 for dense slices reconstruction. These advantages reveal our method a great potential in clinical CT imaging.

[1]  Aaron C. Courville,et al.  Improved Training of Wasserstein GANs , 2017, NIPS.

[2]  Feng Lin,et al.  Low-Dose CT With a Residual Encoder-Decoder Convolutional Neural Network , 2017, IEEE Transactions on Medical Imaging.

[3]  Geoffrey E. Hinton,et al.  Reducing the Dimensionality of Data with Neural Networks , 2006, Science.

[4]  Summer J. Decker,et al.  Computed tomography slice thickness and its effects on three-dimensional reconstruction of anatomical structures , 2016 .

[5]  Jin Liu,et al.  Discriminative Feature Representation to Improve Projection Data Inconsistency for Low Dose CT Imaging , 2017, IEEE Transactions on Medical Imaging.

[6]  B. Movsas,et al.  Dose delivered from Varian's CBCT to patients receiving IMRT for prostate cancer , 2007, Physics in medicine and biology.

[7]  L. Goldman,et al.  Principles of CT: Multislice CT* , 2008, Journal of Nuclear Medicine Technology.

[8]  Marcelo Cypel,et al.  CT-guided Microcoil Pulmonary Nodule Localization prior to Video-assisted Thoracoscopic Surgery: Diagnostic Utility and Recurrence-Free Survival. , 2019, Radiology.

[9]  Daniel Rueckert,et al.  Real-Time Single Image and Video Super-Resolution Using an Efficient Sub-Pixel Convolutional Neural Network , 2016, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[10]  J. W. Chadwick,et al.  The effects of slice thickness and interslice interval on reconstructed cone beam computed tomographic images. , 2010, Oral surgery, oral medicine, oral pathology, oral radiology, and endodontics.

[11]  Orpheus Kolokythas,et al.  Impact of operator-selected image noise index and reconstruction slice thickness on patient radiation dose in 64-MDCT. , 2007, AJR. American journal of roentgenology.

[12]  Alexei A. Efros,et al.  Image-to-Image Translation with Conditional Adversarial Networks , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[13]  C. Gatsonis,et al.  Reduced Lung-Cancer Mortality with Low-Dose Computed Tomographic Screening , 2012 .

[14]  Hyunho Park,et al.  Residual CNN-based Image Super-Resolution for CT Slice Thickness Reduction using Paired CT Scans : Preliminary Validation Study , 2018 .

[15]  Xuanqin Mou,et al.  Low-Dose CT Image Denoising Using a Generative Adversarial Network With Wasserstein Distance and Perceptual Loss , 2017, IEEE Transactions on Medical Imaging.

[16]  Li Fei-Fei,et al.  Perceptual Losses for Real-Time Style Transfer and Super-Resolution , 2016, ECCV.

[17]  P. Paré,et al.  A quantification of the lung surface area in emphysema using computed tomography. , 1999, American journal of respiratory and critical care medicine.

[18]  Raymond H Mak,et al.  CT-based radiomic analysis of stereotactic body radiation therapy patients with lung cancer. , 2016, Radiotherapy and oncology : journal of the European Society for Therapeutic Radiology and Oncology.