Estimating dual-energy CT imaging from single-energy CT data with material decomposition convolutional neural network

Dual-energy computed tomography (DECT) is of great significance for clinical practice due to its huge potential to provide material-specific information. However, DECT scanners are usually more expensive than standard single-energy CT (SECT) scanners and thus are less accessible to undeveloped regions. In this paper, we show that the energy-domain correlation and anatomical consistency between standard DECT images can be harnessed by a deep learning model to provide high-performance DECT imaging from fully-sampled low-energy data together with single-view high-energy data. We demonstrate the feasibility of the approach with two independent cohorts (the first cohort including contrast-enhanced DECT scans of 5753 image slices from 22 patients and the second cohort including spectral CT scans without contrast injection of 2463 image slices from other 22 patients) and show its superior performance on DECT applications. The deep-learning-based approach could be useful to further significantly reduce the radiation dose of current premium DECT scanners and has the potential to simplify the hardware of DECT imaging systems and to enable DECT imaging using standard SECT scanners.

[1]  A. Macovski,et al.  Energy-selective reconstructions in X-ray computerised tomography , 1976, Physics in medicine and biology.

[2]  Ernst Klotz,et al.  The assessment of intracranial bleeding with virtual unenhanced imaging by means of dual-energy CT angiography , 2009, European Radiology.

[3]  E. Finkelstein,et al.  Development and Validation of a Deep Learning System for Diabetic Retinopathy and Related Eye Diseases Using Retinal Images From Multiethnic Populations With Diabetes , 2017, JAMA.

[4]  T. Vogl,et al.  Automatic bone removal technique in whole-body dual-energy CT angiography: performance and image quality. , 2012, AJR. American journal of roentgenology.

[5]  A. McMillan,et al.  Deep learning Mr imaging–based attenuation correction for PeT/Mr imaging 1 , 2017 .

[6]  Huiman X Barnhart,et al.  Characterization of adrenal nodules with dual-energy CT: can virtual unenhanced attenuation values replace true unenhanced attenuation values? , 2012, AJR. American journal of roentgenology.

[7]  Daniel T Boll,et al.  Renal stone assessment with dual-energy multidetector CT and advanced postprocessing techniques: improved characterization of renal stone composition--pilot study. , 2009, Radiology.

[8]  Christianne Leidecker,et al.  Dual-energy CT in patients suspected of having renal masses: can virtual nonenhanced images replace true nonenhanced images? , 2009, Radiology.

[9]  D. Sahani,et al.  Virtual monochromatic reconstruction of dual-energy unenhanced head CT at 65-75 keV maximizes image quality compared with conventional polychromatic CT. , 2013, Radiology.

[10]  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.

[11]  Jaejun Yoo,et al.  Deep Convolutional Framelet Denosing for Low-Dose CT via Wavelet Residual Network , 2017, IEEE Transactions on Medical Imaging.

[12]  Changhwan Kim,et al.  A Feasibility Study of Low-Dose Single-Scan Dual-Energy Cone-Beam CT in Many-View Under-Sampling Framework , 2017, IEEE Transactions on Medical Imaging.

[13]  James C. Williams,et al.  Noninvasive differentiation of uric acid versus non-uric acid kidney stones using dual-energy CT. , 2007, Academic radiology.

[14]  Silvio Mazziotti,et al.  Stone-targeted dual-energy CT: a new diagnostic approach to urinary calculosis. , 2010, AJR. American journal of roentgenology.

[15]  C. McCollough,et al.  Dual- and Multi-Energy CT: Principles, Technical Approaches, and Clinical Applications. , 2015, Radiology.

[16]  Maximilian F Reiser,et al.  Cervical and Cranial Computed Tomographic Angiography With Automated Bone Removal: Dual Energy Computed Tomography Versus Standard Computed Tomography , 2009, Investigative radiology.

[17]  K. Stierstorfer,et al.  First performance evaluation of a dual-source CT (DSCT) system , 2006, European Radiology.

[18]  Donghoon Lee,et al.  Development of a deep neural network for generating synthetic dual-energy chest x-ray images with single x-ray exposure , 2019, Physics in medicine and biology.

[19]  Walter Heindel,et al.  Dual-energy CTA with bone removal for transcranial arteries: intraindividual comparison with standard CTA without bone removal and TOF-MRA. , 2009, Academic radiology.

[20]  Yuxiang Xing,et al.  Fully connected neural network for virtual monochromatic imaging in spectral computed tomography , 2018, Journal of medical imaging.

[21]  Martín Abadi,et al.  TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems , 2016, ArXiv.

[22]  Ke Sheng,et al.  Accurate Multi-Material Decomposition in Dual-Energy CT: A Phantom Study , 2019, IEEE Transactions on Computational Imaging.

[23]  Lei Zhu,et al.  Single-Scan Dual-Energy CT Using Primary Modulation , 2018, IEEE Transactions on Medical Imaging.

[24]  M. Reiser,et al.  The Value of Dual-Energy Bone Removal in Maximum Intensity Projections of Lower Extremity Computed Tomography Angiography , 2009, Investigative radiology.

[25]  Bin Yan,et al.  Image domain dual material decomposition for dual-energy CT using butterfly network. , 2018, Medical physics.

[26]  Shuai Leng,et al.  Maximizing Iodine Contrast-to-Noise Ratios in Abdominal CT Imaging through Use of Energy Domain Noise Reduction and Virtual Monoenergetic Dual-Energy CT. , 2015, Radiology.

[27]  Huazhong Shu,et al.  Domain Progressive 3D Residual Convolution Network to Improve Low-Dose CT Imaging , 2019, IEEE Transactions on Medical Imaging.

[28]  Lei Xing,et al.  Markerless pancreatic tumor target localization enabled by deep learning. , 2019, International journal of radiation oncology, biology, physics.

[29]  Can Ozan Tan,et al.  Physics-informed Deep Learning for Dual-Energy Computed Tomography Image Processing , 2019, Scientific Reports.

[30]  Andrea Vedaldi,et al.  MatConvNet: Convolutional Neural Networks for MATLAB , 2014, ACM Multimedia.

[31]  Shuai Leng,et al.  Feasibility of discriminating uric acid from non-uric acid renal stones using consecutive spatially registered low- and high-energy scans obtained on a conventional CT scanner. , 2015, AJR. American journal of roentgenology.

[32]  W. Kalender,et al.  Evaluation of a prototype dual-energy computed tomographic apparatus. I. Phantom studies. , 1986, Medical physics.

[33]  Andreas K. Maier,et al.  Learning with Known Operators reduces Maximum Training Error Bounds , 2019, Nature Machine Intelligence.

[34]  Jing Huang,et al.  Iterative Reconstruction for X-Ray Computed Tomography Using Prior-Image Induced Nonlocal Regularization , 2014, IEEE Transactions on Biomedical Engineering.

[35]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[36]  Elizabeth A. Krupinski,et al.  Artificial intelligence will soon change the landscape of medical physics research and practice. , 2018, Medical physics.

[37]  Subhashini Venugopalan,et al.  Development and Validation of a Deep Learning Algorithm for Detection of Diabetic Retinopathy in Retinal Fundus Photographs. , 2016, JAMA.

[38]  Sebastian Thrun,et al.  Dermatologist-level classification of skin cancer with deep neural networks , 2017, Nature.

[39]  Shuai Leng,et al.  Dual-energy CT-based monochromatic imaging. , 2012, AJR. American journal of roentgenology.

[40]  Zhaoying Bian,et al.  Pseudo dual energy CT imaging using deep learning-based framework: basic material estimation , 2018, Medical Imaging.

[41]  Max A. Viergever,et al.  Generative Adversarial Networks for Noise Reduction in Low-Dose CT , 2017, IEEE Transactions on Medical Imaging.

[42]  Thomas Brox,et al.  U-Net: Convolutional Networks for Biomedical Image Segmentation , 2015, MICCAI.

[43]  M. Reiser,et al.  Material differentiation by dual energy CT: initial experience , 2007, European Radiology.

[44]  Florian Wolf,et al.  Virtual non-contrast in second-generation, dual-energy computed tomography: reliability of attenuation values. , 2012, European journal of radiology.

[45]  Qianjin Feng,et al.  Low-dose computed tomography image restoration using previous normal-dose scan. , 2011, Medical physics.

[46]  Christoph Thomas,et al.  Virtual nonenhanced dual-energy CT urography with tin-filter technology: determinants of detection of urinary calculi in the renal collecting system. , 2012, Radiology.

[47]  E. Merkle,et al.  Calcified vascular plaque specimens: assessment with cardiac dual-energy multidetector CT in anthropomorphically moving heart phantom. , 2008, Radiology.

[48]  Lei Xing,et al.  Incorporating imaging information from deep neural network layers into image guided radiation therapy (IGRT). , 2019, Radiotherapy and oncology : journal of the European Society for Therapeutic Radiology and Oncology.

[49]  C. McCollough,et al.  Dual-energy CT iodine-subtraction virtual unenhanced technique to detect urinary stones in an iodine-filled collecting system: a phantom study. , 2008, AJR. American journal of roentgenology.

[50]  Hengyong Yu,et al.  A spectral CT denoising algorithm based on weighted block matching 3D filtering , 2017, Optical Engineering + Applications.

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

[52]  Wei Zhao,et al.  Patient-specific reconstruction of volumetric computed tomography images from a single projection view via deep learning , 2019, Nature Biomedical Engineering.

[53]  W. Kalender,et al.  Evaluation of a prototype dual-energy computed tomographic apparatus. II. Determination of vertebral bone mineral content. , 1986, Medical physics.

[54]  R. Gupta,et al.  Differentiation of Hemorrhage from Iodinated Contrast in Different Intracranial Compartments Using Dual-Energy Head CT , 2012, American Journal of Neuroradiology.

[55]  L. Xing,et al.  Using edge-preserving algorithm with non-local mean for significantly improved image-domain material decomposition in dual-energy CT , 2016, Physics in medicine and biology.

[56]  Lei Xing,et al.  Obtaining dual-energy computed tomography (CT) information from a single-energy CT image for quantitative imaging analysis of living subjects by using deep learning , 2019, PSB.