Convolutional Neural Networks With Intermediate Loss for 3D Super-Resolution of CT and MRI Scans

Computed Tomography (CT) scanners that are commonly-used in hospitals and medical centers nowadays produce low-resolution images, e.g. one voxel in the image corresponds to at most one-cubic millimeter of tissue. In order to accurately segment tumors and make treatment plans, radiologists and oncologists need CT scans of higher resolution. The same problem appears in Magnetic Resonance Imaging (MRI). In this paper, we propose an approach for the single-image super-resolution of 3D CT or MRI scans. Our method is based on deep convolutional neural networks (CNNs) composed of 10 convolutional layers and an intermediate upscaling layer that is placed after the first 6 convolutional layers. Our first CNN, which increases the resolution on two axes (width and height), is followed by a second CNN, which increases the resolution on the third axis (depth). Different from other methods, we compute the loss with respect to the ground-truth high-resolution image right after the upscaling layer, in addition to computing the loss after the last convolutional layer. The intermediate loss forces our network to produce a better output, closer to the ground-truth. A widely-used approach to obtain sharp results is to add Gaussian blur using a fixed standard deviation. In order to avoid overfitting to a fixed standard deviation, we apply Gaussian smoothing with various standard deviations, unlike other approaches. We evaluate the proposed method in the context of 2D and 3D super-resolution of CT and MRI scans from two databases, comparing it to related works from the literature and baselines based on various interpolation schemes, using <inline-formula> <tex-math notation="LaTeX">$2\times $ </tex-math></inline-formula> and <inline-formula> <tex-math notation="LaTeX">$4\times $ </tex-math></inline-formula> scaling factors. The empirical study shows that our approach attains superior results to all other methods. Moreover, our subjective image quality assessment by human observers reveals that both doctors and regular annotators chose our method in favor of Lanczos interpolation in 97.55% cases for an upscaling factor of <inline-formula> <tex-math notation="LaTeX">$2\times $ </tex-math></inline-formula> and in 96.69% cases for an upscaling factor of <inline-formula> <tex-math notation="LaTeX">$4\times $ </tex-math></inline-formula>. In order to allow others to reproduce our state-of-the-art results, we provide our code as open source at <uri>https://github.com/lilygeorgescu/3d-super-res-cnn</uri>.

[1]  J. C. Elliott,et al.  X‐ray microtomography , 1982, Journal of microscopy.

[2]  George Cybenko,et al.  Approximation by superpositions of a sigmoidal function , 1989, Math. Control. Signals Syst..

[3]  Kurt Hornik,et al.  Approximation capabilities of multilayer feedforward networks , 1991, Neural Networks.

[4]  Thomas G. Dietterich Approximate Statistical Tests for Comparing Supervised Classification Learning Algorithms , 1998, Neural Computation.

[5]  Hong Chang,et al.  Super-resolution through neighbor embedding , 2004, Proceedings of the 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2004. CVPR 2004..

[6]  Eero P. Simoncelli,et al.  Image quality assessment: from error visibility to structural similarity , 2004, IEEE Transactions on Image Processing.

[7]  Gustavo de Veciana,et al.  An information fidelity criterion for image quality assessment using natural scene statistics , 2005, IEEE Transactions on Image Processing.

[8]  Alan C. Bovik,et al.  Mean squared error: Love it or leave it? A new look at Signal Fidelity Measures , 2009, IEEE Signal Processing Magazine.

[9]  Emilio Quaia,et al.  The SYRMEP Beamline of Elettra: Clinical Mammography and Bio‐medical Applications , 2010 .

[10]  Dariush Sardari,et al.  Calculation of Externally Applied Electric Field Intensity for Disruption of Cancer Cell Proliferation , 2010, Electromagnetic biology and medicine.

[11]  Geoffrey E. Hinton,et al.  Rectified Linear Units Improve Restricted Boltzmann Machines , 2010, ICML.

[12]  Thomas S. Huang,et al.  Image Super-Resolution Via Sparse Representation , 2010, IEEE Transactions on Image Processing.

[13]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[14]  Aline Roumy,et al.  Low-Complexity Single-Image Super-Resolution based on Nonnegative Neighbor Embedding , 2012, BMVC.

[15]  Andrew Zisserman,et al.  Speeding up Convolutional Neural Networks with Low Rank Expansions , 2014, BMVC.

[16]  Yoshua Bengio,et al.  Generative Adversarial Nets , 2014, NIPS.

[17]  Carol Davila,et al.  ADVANTAGES OF LASER PHOTOACOUSTIC SPECTROSCOPY IN RADIOTHERAPY CHARACTERIZATION , 2014 .

[18]  Chih-Yuan Yang,et al.  Single-Image Super-Resolution: A Benchmark , 2014, ECCV.

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

[20]  Xiaogang Wang,et al.  Visual Tracking with Fully Convolutional Networks , 2015, 2015 IEEE International Conference on Computer Vision (ICCV).

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

[22]  Luc Van Gool,et al.  Jointly Optimized Regressors for Image Super‐resolution , 2015, Comput. Graph. Forum.

[23]  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).

[24]  Jian Sun,et al.  Deep Residual Learning for Image Recognition , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[25]  Guigang Zhang,et al.  Deep Learning , 2016, Int. J. Semantic Comput..

[26]  Xiaoou Tang,et al.  Image Super-Resolution Using Deep Convolutional Networks , 2014, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[27]  Konstantinos Kamnitsas,et al.  Multi-input Cardiac Image Super-Resolution Using Convolutional Neural Networks , 2016, MICCAI.

[28]  Franz Pfeiffer,et al.  Advanced Non-Destructive Ocular Visualization Methods by Improved X-Ray Imaging Techniques , 2017, PloS one.

[29]  Trevor Darrell,et al.  Fully Convolutional Networks for Semantic Segmentation , 2017, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[30]  Bo Chen,et al.  MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications , 2017, ArXiv.

[31]  Ling Shao,et al.  Simultaneous Super-Resolution and Cross-Modality Synthesis of 3D Medical Images Using Weakly-Supervised Joint Convolutional Sparse Coding , 2017, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[32]  Kyoung Mu Lee,et al.  Enhanced Deep Residual Networks for Single Image Super-Resolution , 2017, 2017 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW).

[33]  Thomas S. Huang,et al.  Computed tomography super-resolution using convolutional neural networks , 2017, 2017 IEEE International Conference on Image Processing (ICIP).

[34]  David Maintz,et al.  Image quality evaluation of dual-layer spectral detector CT of the chest and comparison with conventional CT imaging. , 2017, European journal of radiology.

[35]  Yun Fu,et al.  Image Super-Resolution Using Very Deep Residual Channel Attention Networks , 2018, ECCV.

[36]  Houshang Darabi,et al.  LSTM Fully Convolutional Networks for Time Series Classification , 2017, IEEE Access.

[37]  Yu Yang,et al.  Simultaneous single- and multi-contrast super-resolution for brain MRI images based on a convolutional neural network , 2018, Comput. Biol. Medicine.

[38]  Lulu Wang,et al.  Accelerated Super-resolution MR Image Reconstruction via a 3D Densely Connected Deep Convolutional Neural Network , 2018, 2018 IEEE International Conference on Bioinformatics and Biomedicine (BIBM).

[39]  Debiao Li,et al.  Efficient and Accurate MRI Super-Resolution using a Generative Adversarial Network and 3D Multi-Level Densely Connected Network , 2018, MICCAI.

[40]  Dwarikanath Mahapatra,et al.  Image super-resolution using progressive generative adversarial networks for medical image analysis , 2019, Comput. Medical Imaging Graph..

[41]  Shihui Ying,et al.  MR Image Super-Resolution via Wide Residual Networks With Fixed Skip Connection , 2019, IEEE Journal of Biomedical and Health Informatics.

[42]  Nicolas Passat,et al.  Multiscale brain MRI super-resolution using deep 3D convolutional networks , 2019, Comput. Medical Imaging Graph..

[43]  Jean-Yves Tourneret,et al.  A Tensor Factorization Method for 3-D Super Resolution With Application to Dental CT , 2018, IEEE Transactions on Medical Imaging.

[44]  Xiaofeng Du,et al.  Gradient-Guided Convolutional Neural Network for MRI Image Super-Resolution , 2019, Applied Sciences.

[45]  Eser Sert,et al.  A new approach for brain tumor diagnosis system: Single image super resolution based maximum fuzzy entropy segmentation and convolutional neural network. , 2019, Medical hypotheses.

[46]  Mohsen Guizani,et al.  Super-Resolution of Brain MRI Images Using Overcomplete Dictionaries and Nonlocal Similarity , 2019, IEEE Access.

[47]  Adrian Basarab,et al.  Deep Learning-Based Super-Resolution Applied to Dental Computed Tomography , 2019, IEEE Transactions on Radiation and Plasma Medical Sciences.

[48]  Ling Shao,et al.  Object-Centric Auto-Encoders and Dummy Anomalies for Abnormal Event Detection in Video , 2018, 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[49]  Tao Zhang,et al.  Channel Splitting Network for Single MR Image Super-Resolution , 2018, IEEE Transactions on Image Processing.

[50]  Guang Li,et al.  CT Super-Resolution GAN Constrained by the Identical, Residual, and Cycle Learning Ensemble (GAN-CIRCLE) , 2018, IEEE Transactions on Medical Imaging.

[51]  Zhongshi He,et al.  Super-resolution reconstruction of single anisotropic 3D MR images using residual convolutional neural network , 2020, Neurocomputing.

[52]  Jakub Jurek,et al.  CNN-based superresolution reconstruction of 3D MR images using thick-slice scans , 2020, Biocybernetics and Biomedical Engineering.

[53]  Corrado Mencar,et al.  Crowd Detection for Drone Safe Landing Through Fully-Convolutional Neural Networks , 2020, SOFSEM.