Deep learning with convolutional neural network in radiology
- 1k Downloads
Deep learning with a convolutional neural network (CNN) is gaining attention recently for its high performance in image recognition. Images themselves can be utilized in a learning process with this technique, and feature extraction in advance of the learning process is not required. Important features can be automatically learned. Thanks to the development of hardware and software in addition to techniques regarding deep learning, application of this technique to radiological images for predicting clinically useful information, such as the detection and the evaluation of lesions, etc., are beginning to be investigated. This article illustrates basic technical knowledge regarding deep learning with CNNs along the actual course (collecting data, implementing CNNs, and training and testing phases). Pitfalls regarding this technique and how to manage them are also illustrated. We also described some advanced topics of deep learning, results of recent clinical studies, and the future directions of clinical application of deep learning techniques.
KeywordsDeep learning Convolutional neural network CT MRI PET
No specific funding was disclosed.
Compliance with ethical standards
Conflict of interest
The authors declare that they have no conflict of interest.
This article does not contain any studies with human participants or animals performed by any of the authors.
- 4.Krizhevsky A, Sutskever I, Hinton G. ImageNet classification with deep convolutional neural networks. Advances in Neural Information Processing System 25 (NIPS 2012). 2012. https://papers.nips.cc/paper/4824-imagenet-classification-with-deep-convolutional-neural-networks:Published. Accessed 14 Dec 2017.
- 18.Le QV, Ranzato M, Monga R, Devin M, Chen K, Corrado GS, et al. Building high-level features using large scale unsupervised learning. International Conference on Machine Learning. 2012. http://icml.cc/2012/papers. Accessed 14 Dec 2017.
- 21.Nair V, Hinton G. Rectified linear units improve restricted Boltzmann machines. International Conference on Machine Learning. 2010. Accessed 14 Dec 2017.Google Scholar
- 22.Srivastava N, Hinton GE, Krizhevsky A, Sutskever I, Salakhutdinov R. Dropout: a simple way to prevent neural networks from overfitting. J Mach Learn Res. 2014;15:1929–58.Google Scholar
- 23.Ioffe S, Szegedy C. Batch normalization: accelerating deep network training by reducing internal covariate shift. Cornell University Library. 2015. http://arxiv.org/abs/1502.03167. Accessed 30 April 2017.
- 25.Duchi J, Hazan E, Singer Y. Adaptive subgradient methods for online learning and stochastic optimization. J Mach Learn Res. 2011;12:2121–59.Google Scholar
- 26.Kingma DP, Ba JL. Adam: a method for stochastic optimization. Cornell University Library. 2014. http://arxiv.org/abs/1412.6980. Accessed 30 April 2017.
- 27.Szegedy C, Liu W, Jia Y, Sermanet P, Reed S, Anguelov D, et al. Going deeper with convolutions. Cornell University Library. 2014. https://arxiv.org/abs/1409.4842. Accessed 14 Dec 2017.
- 28.He K, Zhang X, Ren S, Sun J. Deep residual learning for image recognition. Cornell University Library. 2015. https://arxiv.org/abs/1512.03385. Accessed 14 Dec 2017.
- 30.Andrearczyk V, Whelan PF. Using filter banks in convolutional neural networks for texture classification. Cornell University Library. 2016. https://arxiv.org/abs/1601.02919. Accessed 30 April 2017.
- 31.Long J, Shelhamer E, Darrell T. Fully convolutional networks for semantic segmentation. 2015. http://ieeexplore.ieee.org/document/7298965/?reload=true. Accessed 14 Dec 2017.
- 32.Ronneberger O, Fischer P, Brox T. U-Net: convolutional networks for biomedical image segmentation. Cornell University Library. 2015. https://arxiv.org/abs/1505.04597. Accessed 14 Dec 2017.
- 33.Badrinarayanan V, Kendall A, Cipolla R. SegNet: a deep convolutional encoder-decoder architecture for image segmentation. Cornell University Library. 2015. https://arxiv.org/pdf/1511.00561. Accessed 14 Dec 2017.
- 34.Mao X, Shen C, Yang Y. Image restoration using very deep convolutional encoder-decoder networks with symmetric skip connections. Cornell University Library. 2016. https://arxiv.org/abs/1603.09056. Accessed 14 Dec 2017.
- 37.Bengio Y. Deep learning of representations for unsupervised and transfer learning. In: JMLR: Workshop and Conference Proceedings. 2012;17–37.Google Scholar
- 63.Leynes AP, Yang J, Wiesinger F, Kaushik SS, Shanbhag DD, Seo Y, et al. Direct PseudoCT generation for pelvis PET/MRI attenuation correction using deep convolutional neural networks with multi-parametric MRI: zero echo-time and dixon deep pseudoCT (ZeDD-CT). J Nucl Med. 2017. https://doi.org/10.2967/jnumed.117.198051.PubMedGoogle Scholar
- 66.Pickhardt PJ, Lubner MG, Kim DH, Tang J, Ruma JA, del Rio AM, et al. Abdominal CT with model-based iterative reconstruction (MBIR): initial results of a prospective trial comparing ultralow-dose with standard-dose imaging. AJR Am J Roentgenol. 2012;199:1266–74.CrossRefPubMedPubMedCentralGoogle Scholar
- 69.Yasaka K, Katsura M, Hanaoka S, Sato J, Ohtomo K. High-resolution CT with new model-based iterative reconstruction with resolution preference algorithm in evaluations of lung nodules: comparison with conventional model-based iterative reconstruction and adaptive statistical iterative reconstruction. Eur J Radiol. 2016;85:599–606.CrossRefPubMedGoogle Scholar