×

Image decomposition algorithm for dual-energy computed tomography via fully convolutional network. (English) Zbl 1431.92107

Summary: Background. Dual-energy computed tomography (DECT) has been widely used due to improved substances identification from additional spectral information. The quality of material-specific image produced by DECT attaches great importance to the elaborated design of the basis material decomposition method. Objective. The aim of this work is to develop and validate a data-driven algorithm for the image-based decomposition problem. Methods. A deep neural net, consisting of a fully convolutional net (FCN) and a fully connected net, is proposed to solve the material decomposition problem. The former net extracts the feature representation of input reconstructed images, and the latter net calculates the decomposed basic material coefficients from the joint feature vector. The whole model was trained and tested using a modified clinical dataset. Results. The proposed FCN delivers image with about 60% smaller bias and 70% lower standard deviation than the competing algorithms, suggesting its better material separation capability. Moreover, FCN still yields excellent performance in case of photon noise. Conclusions. Our deep cascaded network features high decomposition accuracies and noise robust property. The experimental results have shown the strong function fitting ability of the deep neural network. Deep learning paradigm could be a promising way to solve the nonlinear problem in DECT.

MSC:

92C55 Biomedical imaging and signal processing
PDFBibTeX XMLCite
Full Text: DOI

References:

[1] Ying, Z.; Naidu, R.; Crawford, C. R., Dual energy computed tomography for explosive detection, Journal of X-Ray Science and Technology, 14, 4, 235-256, (2006)
[2] Goo, H. W.; Goo, J. M., Dual-energy CT: new horizon in medical imaging, Korean Journal of Radiology, 18, 4, 555-569, (2017) · doi:10.3348/kjr.2017.18.4.555
[3] Hong, C.; Chin, T. Y.; Peh, W. C. G., Dual-energy CT in gout-a review of current concepts and applications, Journal of Medical Radiation Sciences, 64, 1, 41-51, (2017) · doi:10.1002/jmrs.223
[4] Postma, A. A.; Das, M.; Stadler, A. A. R.; Wildberger, J. E., Dual-energy CT: what the neuroradiologist should know, Current Radiology Reports, 3, 5, (2015) · doi:10.1007/s40134-015-0097-9
[5] Li, B.; Zhang, Y., Projection decomposition algorithm of x-ray dual-energy computed tomography based on projection matching, Acta Optica Sinica, 31, 3, (2011) · doi:10.3788/aos20113103.311002
[6] Brendel, B.; Schlomka, J. P., Empirical projection-based basis-component decomposition method, Proceedings of Medical Imaging 2009: Physics of Medical Imaging, International Society for Optics and Photonics · doi:10.1117/12.811006
[7] Schmidt, T. G.; Zimmerman, K. C., Material decomposition of multi-spectral x-ray projections using neural networks, (2015)
[8] Lambert, J. W.; Sun, Y.; Gould, R. G.; Ohliger, M. A.; Li, Z.; Yeh, B. M., An image-domain contrast material extraction method for dual-energy computed tomography, Investigative Radiology, 52, 4, 245-254, (2017) · doi:10.1097/rli.0000000000000335
[9] Niu, T.; Dong, X.; Petrongolo, M.; Zhu, L., Iterative image-domain decomposition for dual-energy CT, Medical Physics, 41, 4, (2014) · doi:10.1118/1.4866386
[10] Dong, X.; Niu, T.; Zhu, L., Combined iterative reconstruction and image-domain decomposition for dual energy CT using total-variation regularization, Medical Physics, 41, 5, (2014) · doi:10.1118/1.4870375
[11] Szczykutowicz, T. P.; Chen, G. H., Dual energy CT using slow kVp switching acquisition and prior image constrained compressed sensing, Physics in Medicine and Biology, 55, 21, 6411-6429, (2010) · doi:10.1088/0031-9155/55/21/005
[12] Granton, P. V.; Pollmann, S. I.; Ford, N. L.; Drangova, M.; Holdsworth, D. W., Implementation of dual- and triple-energy cone-beam micro-CT for postreconstruction material decomposition, Medical Physics, 35, 11, 5030-5042, (2008) · doi:10.1118/1.2987668
[13] Baer, M.; Maaß, C.; Kalender, W. A.; Kachelrieß, M., Image-based dual energy CT using optimized precorrection functions: a practical new approach to material decomposition in the image domain, World Congress on Medical Physics and Biomedical Engineering, Springer Berlin Heidelberg
[14] Brooks, R. A., A quantitative theory of the Hounsfield unit and its application to dual energy scanning, Journal of Computer Assisted Tomography, 1, 4, 487-493, (1977) · doi:10.1097/00004728-197710000-00016
[15] Kalender, W. A.; Klotz, E.; Kostaridou, L., An algorithm for noise suppression in dual energy CT material density images, IEEE Transactions on Medical Imaging, 7, 3, 218-224, (1988) · doi:10.1109/42.7785
[16] Ergun, D. L.; Mistretta, C. A.; Brown, D. . E., Single-exposure dual-energy computed radiography: improved detection and processing, Radiology, 174, 1, 243-249, (1990) · doi:10.1148/radiology.174.1.2294555
[17] Dobbins, J. T., Recent progress in noise reduction and scatter correction in dual-energy imaging, Proceedings of SPIE-the International Society for Optical Engineering
[18] Peppler, W. W.; Dobbins, J. T.; Bellers, E. B., Dual-energy computed radiography: improvements in processing, Proceedings of SPIE-the International Society for Optical Engineering
[19] Kaiser, L.; Gomez, A. N.; Shazeer, N., One model to learn them all
[20] He, K.; Zhang, X.; Ren, S.; Sun, J., Deep residual learning for image recognition
[21] Kong, T.; Yao, A.; Chen, Y.; Sun, F., HyperNet: towards accurate region proposal generation and joint object detection, Proceedings of Computer Vision and Patter Recognition IEEE
[22] Wu, D.; Kim, K.; Fakhri, G. E., A cascaded convolutional neural network for x-ray low-dose CT image denoising
[23] Zhang, H.; Li, L.; Qiao, K., Image prediction for limited-angle tomography via deep learning with convolutional neural network
[24] Wang, G., A perspective on deep imaging, IEEE Access, 4, 99, 8914-8924, (2016) · doi:10.1109/access.2016.2624938
[25] Zimmerman, K. C.; Schmidt, T. G., Experimental comparison of empirical material decomposition methods for spectral CT, Physics in Medicine & Biology, 60, 8, 3175-3191, (2015) · doi:10.1088/0031-9155/60/8/3175
[26] Lee, W. J.; Kim, D. S.; Kang, S. W.; Yi, W. J., Material depth reconstruction method of multi-energy X-ray images using neural network, Proceedings of International Conference of the IEEE Engineering in Medicine and Biology Society
[27] Lu, Y.; Berger, M.; Manhart, M., Bridge to real data: empirical multiple material calibration for learning-based material decomposition, Proceedings of IEEE International Symposium on Biomedical Imaging
[28] Lu, Y.; Geret, J.; Unberath, M., Projection-based material decomposition by machine learning using image-based features for computed tomography, Proceedings of the 13th International Meeting on Fully Three-Dimensional Image Reconstruction in Radiology and Nuclear Medicine
[29] Shelhamer, E.; Long, J.; Darrell, T., Fully convolutional networks for semantic segmentation, Proceedings of IEEE Transactions on Patter Analysis & Machine Intelligence
[30] Abadi, M.; Agarwal, A.; Barham, P., TensorFlow: large-scale machine learning on heterogeneous distributed systems
[31] Kingma, D. P.; Ba, J., Adam: a method for stochastic optimization
[32] Dabov, K.; Foi, A.; Katkovnik, V., Image denoising by sparse 3-D transform-domain collaborative filtering, IEEE Transactions on Image Processing, 16, 8, 2080-2095, (2007) · doi:10.1109/tip.2007.901238
[33] Poludniowski, G.; Landry, G.; Deblois, F.; Evans, P. M.; Verhaegen, F., SpekCalc: a program to calculate photon spectra from tungsten anode x-ray tubes, Physics in Medicine & Biology, 54, 19, N433-N438, (2009) · doi:10.1088/0031-9155/54/19/n01
[34] Zhu, L.; Wang, J.; Xing, L., Noise suppression in scatter correction for cone-beam CT, Medical Physics, 36, 3, 741-752, (2009) · doi:10.1118/1.3063001
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.