Bone Suppression of Chest Radiographs With Cascaded Convolutional Networks in Wavelet Domain

Bone suppression of chest radiographs (CXRs) is potentially useful for diagnosing lung diseases for radiologists and computer–aided diagnosis. This paper presents a cascaded convolutional network model in wavelet domain (Wavelet-CCN) for bone suppression in single conventional CXR. Wavelet coefficients are sparse and suitable as the output of convolutional network. The convolutional networks are trained to predict the wavelet coefficients of bone images from the wavelet coefficients of CXRs, using real two-exposure dual energy subtraction (DES) CXRs as training data. By combining the multi-level wavelet decomposition and a cascaded refinement framework, the Wavelet-CCN model can work automatically with a multi-scale approach and progressively refine the prediction in terms of accuracy and spatial resolution. Compared with previous work of CamsNet model which preforms bone prediction in gradient domain, the Wavelet-CCN model predicts the wavelet coefficients to reconstruct bone images and can avoid the inconsistent background intensity caused by 2D integration of gradients. The predicted bone image is subtracted from the original CXR to produce a soft-tissue image. The Wavelet-CCN model and its variants with different wavelet basis are evaluated on a dataset that consists of 504 cases of real two-exposure DES CXRs (404 cases for training and 100 cases for test). Experimental results show that among all the variants and different wavelet bases, the Wavelet-CCN model with Haar wavelet performs best. The average peak signal-to-noise ratio and structural similarity index of the soft-tissue images produced by the proposed Wavelet-CCN model are both higher than those of the previous CamsNet model in gradient domain, reaching values of 39.4 (±0.94) dB and 0.977 (±0.004), respectively. The results also demonstrate that the Wavelet-CCN model can process the CXRs acquired by four types of X-ray machines.

[1]  Liqing Zhang,et al.  Rectifier Neural Network with a Dual-Pathway Architecture for Image Denoising , 2016, ArXiv.

[2]  Jong Chul Ye,et al.  Wavelet Domain Residual Network (WavResNet) for Low-Dose X-ray CT Reconstruction , 2017, ArXiv.

[3]  Williams Travis,et al.  Advanced Image Classification Using Wavelets and Convolutional Neural Networks , 2016 .

[4]  Tae-Seong Kim,et al.  Rib suppression in frontal chest radiographs: A blind source separation approach , 2007, 2007 9th International Symposium on Signal Processing and Its Applications.

[5]  E. Candès.  Ridgelets: estimating with ridge functions , 2003 .

[6]  Hasan Ogul,et al.  Eliminating rib shadows in chest radiographic images providing diagnostic assistance , 2016, Comput. Methods Programs Biomed..

[7]  Kunio Doi,et al.  Image-processing technique for suppressing ribs in chest radiographs by means of massive training artificial neural network (MTANN) , 2006, IEEE Transactions on Medical Imaging.

[8]  Qianjin Feng,et al.  Cascade of multi-scale convolutional neural networks for bone suppression of chest radiographs in gradient domain , 2017, Medical Image Anal..

[9]  Alessandro Foi,et al.  Image Denoising by Sparse 3-D Transform-Domain Collaborative Filtering , 2007, IEEE Transactions on Image Processing.

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

[11]  Jian Sun,et al.  Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification , 2015, 2015 IEEE International Conference on Computer Vision (ICCV).

[12]  Jan Kautz,et al.  Loss Functions for Neural Networks for Image Processing , 2015, ArXiv.

[13]  Bram van Ginneken,et al.  Segmentation of the posterior ribs in chest radiographs using iterated contextual pixel classification , 2006, IEEE Transactions on Medical Imaging.

[14]  Kunio Doi,et al.  Suppression of the contrast of ribs in chest radiographs by means of massive training artificial neural network , 2004, SPIE Medical Imaging.

[15]  Neeraj Kumar,et al.  Convolutional neural networks for wavelet domain super resolution , 2017, Pattern Recognit. Lett..

[16]  Stéphane Mallat,et al.  Multifrequency channel decompositions of images and wavelet models , 1989, IEEE Trans. Acoust. Speech Signal Process..

[17]  Adin Ramirez Rivera,et al.  Deep learning models for bone suppression in chest radiographs , 2017, 2017 IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology (CIBCB).

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

[19]  Bram van Ginneken,et al.  Suppression of Translucent Elongated Structures: Applications in Chest Radiography , 2013, IEEE Transactions on Medical Imaging.

[20]  Minh N. Do,et al.  Contourlets: a directional multiresolution image representation , 2002, Proceedings. International Conference on Image Processing.

[21]  L. Hogeweg,et al.  Automatic detection of tuberculosis in chest radiographs , 2013 .

[22]  Bram van Ginneken,et al.  Filter learning: Application to suppression of bony structures from chest radiographs , 2006, Medical Image Anal..

[23]  Rakesh Sharma.  Wavelet Thresholding for Image De-noising , 2011 .

[24]  Kenji Suzuki,et al.  Computerized Detection of Lung Nodules by Means of “Virtual Dual-Energy” Radiography , 2013, IEEE Transactions on Biomedical Engineering.

[25]  Peter Vock,et al.  Dual energy subtraction: principles and clinical applications. , 2009, European journal of radiology.

[26]  Jian Sun,et al.  Guided Image Filtering , 2010, ECCV.

[27]  Max A. Viergever,et al.  Segmenting the posterior ribs in chest radiographs by iterated contextual pixel classification , 2003, SPIE Medical Imaging.

[28]  Kenji Suzuki,et al.  Separation of Bones From Chest Radiographs by Means of Anatomically Specific Multiple Massive-Training ANNs Combined With Total Variation Minimization Smoothing , 2014, IEEE Transactions on Medical Imaging.

[29]  Hieu Xuan Nguyen,et al.  Ribs Suppression in Chest X-Ray Images by Using ICA Method , 2015 .

[30]  Jong Chul Ye,et al.  Multi-Scale Wavelet Domain Residual Learning for Limited-Angle CT Reconstruction , 2017, ArXiv.

[31]  Tieniu Tan,et al.  Wavelet-SRNet: A Wavelet-Based CNN for Multi-scale Face Super Resolution , 2017, 2017 IEEE International Conference on Computer Vision (ICCV).

[32]  Peter Maday,et al.  Elimination of clavicle shadows to help automatic lung nodule detection on chest radiographs , 2009 .

[33]  Tong Fang,et al.  Gradient domain layer separation under independent motion , 2009, 2009 IEEE 12th International Conference on Computer Vision.