The Journal of Supercomputing

, Volume 75, Issue 2, pp 704–718 | Cite as

Medical image denoising using convolutional neural network: a residual learning approach

  • Worku Jifara
  • Feng JiangEmail author
  • Seungmin Rho
  • Maowei Cheng
  • Shaohui Liu


In medical imaging, denoising is very important for analysis of images, diagnosis and treatment of diseases. Currently, image denoising methods based on deep learning are effective, where the methods are however limited for the requirement of training sample size (i.e., not successful enough for small data size). Using small sample size, we design deep feed forward denoising convolutional neural networks by studying the model in deep framework, learning approach and regularization approach for medical image denoising. More specifically, we use residual learning as a learning approach and batch normalization as regularization in the deep model. Unlike most of the other image denoising approaches which directly learn the latent clean images, the residual learning approach learns the noise from the noisy images instead of the latent clean images where the denoised images are obtained by subtracting the learned residual from the noisy image. Moreover, batch normalization is integrated with residual learning to improve model learning accuracy and training time. We compute the quality of the reconstructed or denoised image in standard image quality metrics, peak signal to noise ratio and structural similarity and compare our model performance with some medical image denoising techniques. Experimental results reveal that our approach has better performance than some other methods.


Medical image Image denoising Residual learning CNN Batch normalization 



This work is partially funded by the MOE-Microsoft Key Laboratory of Natural Language Processing and Speech, Harbin Institute of Technology, the Major State Basic Research Development Program of China (973 Program 2015CB351804) and the National Natural Science Foundation of China under Grant Nos. 61572155, 61672188 and 61272386. We would also like to acknowledge NVIDIA Corporation who kindly provided two sets of GPU. We would like to acknowledge the editors and the anonymous reviewers whose important comments and suggestions led to greatly improved the manuscript.


  1. 1.
    Gao Q, Olgac N (2016) Determination of the bounds of imaginary spectra of LTI systems with multiple time delays. Automatica 72:235–241CrossRefzbMATHGoogle Scholar
  2. 2.
    Mondal T, Maitra M (2015) Denoising and compression of medical image in wavelet 2D. Int J Recent Innov Trends Comput Commun 6:173–178Google Scholar
  3. 3.
    Mustafa N, Li JP, Khan SA, Giess M (2015) Medical image de-noising schemes using wavelet transform with fixed form thresholding. Int J Adv Comput Sci Appl 6:173–178Google Scholar
  4. 4.
    Bahendwar YS, Sinh GR (2012) Efficient algorithm for denoising of medical images using discrete wavelet transforms. Math Methods Syst Sci Eng 142:158–162Google Scholar
  5. 5.
    Zhang X (2015) Image denoising using shearlet transform and nonlinear diffusion. Proc Sci 20:033016Google Scholar
  6. 6.
    Starck JL, Cands EJ, Donoho DL (2002) The curvelet transform for image denoising. IEEE Tran Image Process 11(6):670–684MathSciNetCrossRefzbMATHGoogle Scholar
  7. 7.
    Hu J, Pu Y, Wu X, Zhang Y, Zhou J (2012) Improved DCT-based nonlocal means filter for MR images denoising. Comput Math Methods Med 2012:85–91Google Scholar
  8. 8.
    Sameh Arif A, Mansor S, Logeswaran R (2011) Combined bilateral and anisotropic-diffusion filters for medical image de-noising. IEEE Student Conf Res Dev SCOReD 2:420–424Google Scholar
  9. 9.
    Bhonsle D, Chandra V, Sinha GR (2012) Medical image denoising using bilateral filter. Int J Image Gr Signal Process 4:36–43CrossRefGoogle Scholar
  10. 10.
    Dabov K, Foi A, Katkovnik V, Katkovnik V, Egiazarian K (2007) Image denoising by sparse 3-D transform-domain collaborative filtering. IEEE Trans Image Process 16:2080–2095MathSciNetCrossRefGoogle Scholar
  11. 11.
    Elad M, Aharon M (2006) Image Denoising via learned dictionaries and sparse representation. Comput Vis Pattern Recogn 1:1063–6919Google Scholar
  12. 12.
    Bengio Y, Lamblin P, Popovici D, Larochelle H (2007) Greedy Layer-Wise training of deep networks. Adv Neural Inf Process Syst 19:153Google Scholar
  13. 13.
    Schmidhuber J (2014) Deep learning in neural networks: an overview. J Neural Netw 61:85–117CrossRefGoogle Scholar
  14. 14.
    Tan Z, Wei H, Chen Y, Du M, Ye S (2016) Design for medical imaging services platform based on cloud computing. Int J Big Data Intell 3(4):270–278CrossRefGoogle Scholar
  15. 15.
    Vincent P, Larochelle H, Bengio Y, Manzagol PA (2011) Extracting and composing robust features with denoising autoencoders. In: IEEE Student Conference on Research and Development, pp 1096–1103Google Scholar
  16. 16.
    Gondara L (2016) Medical image denoising using convolutional denoising autoencoders. In: IEEE Conference on Computer Vision and Pattern RecognitionGoogle Scholar
  17. 17.
    Buades A, Coll B, Morel JM (2005) A review of image denoising algorithms with a new one. SIAM J Multiscale Model Simul A SIAM Interdiscip 4:490–530MathSciNetCrossRefzbMATHGoogle Scholar
  18. 18.
    Agostinelli F, Anderson MR, Lee H (2013) Adaptive multi-column deep neural networks with application to robust image denoising. Adv Neural Inf Process Syst 26:1493–1501Google Scholar
  19. 19.
    Burger HC, Schuler CJ, Harmeling S (2012) Image denoising: can plain neural networks compete with BM3D? In: IEEE Computer Society Conference on Computer Vision and Pattern Recognition, pp 2392–2399Google Scholar
  20. 20.
    Zhang K, Zuo W, Chen Y, Meng D, Zhang L (2016) Beyond a Gaussian denoiser: residual learning of deep CNN for image denoising, Tech report Computer Vision and Pattern Recognition, pp 1–13Google Scholar
  21. 21.
    Girshick NR (2015) Fast R-CNN. Int Conf Comput Vis Pattern Recogn 1440–1448 (arXiv:1504.08083)
  22. 22.
    Girshick R, Donahue J, Darrell T, Malik J (2014) Rich feature hierarchies for accurate object detection and semantic segmentation. In: Proceedings of the IEEE Computer Society Conference on Computer Vision Pattern Recognition, pp 580–587Google Scholar
  23. 23.
    Oliveira TP, Barbar JS, Soares Alexsandro Santos (2016) Computer network traffic prediction: a comparison between traditional and deep learning neural networks. Int J Big Data Intell 3(1):28–37CrossRefGoogle Scholar
  24. 24.
    Krizhevsky A, Sutskever I, Geoffrey EH (2012) Imagenet classification with deep convolutional neural networks. Adv Neural Inf Process Syst 25:1097–1105Google Scholar
  25. 25.
    He K, Zhang X, Ren S, Sun J (2014) Spatial pyramid pooling in deep convolutional networks for visual recognition. In: European Conference on Computer Vision, Springer, BerlinGoogle Scholar
  26. 26.
    Glorot X, Bengio Y (2010) Understanding the difficulty of training deep feedforward neural networks. In: Proceedings of the International Conference on Artificial Intelligence and Statistics vol. 9, pp 249–256Google Scholar
  27. 27.
    Ioffe S, Szegedy C (2015) Batch normalization: accelerating deep network training by reducing internal covariate shift. In: Proceedings of the 32nd International Conference on Machine Learning, pp 448–456Google Scholar
  28. 28.
    He K, Zhang X, Ren Sh, Sun J (2015) Deep residual learning for image recognition. In: Proceedings of the IEEE Conference on Computer Vision Pattern Recognition, pp 770–778 (arXiv preprint arXiv:1512.03385)
  29. 29.
    Xing C, Ma L, Yang X (2016) Stacked denoise autoencoder based feature extraction and classification for hyperspectral images. J Sens 2016:1–10. doi: 10.1155/2016/3632943
  30. 30.
    Masci J, Meier U, Ciresan D, Schmidhuber J (2011) Stacked convolutional auto-encoders for hierarchical feature extraction. Springer, BerlinCrossRefGoogle Scholar
  31. 31.
    Jiang F, Rho S, Chen B, Du X, Zhao D (2015) Face hallucination and recognition in social network services. J Supercomput 71(6):2035–2049CrossRefGoogle Scholar
  32. 32.
    Jiang F, Chen B, Li K, Zhao D (2014) Big data driven decision making and multi-prior models collaboration for media restoration. Multimedia Tools Appl 75:12967–12982Google Scholar
  33. 33.
    Jiang F, Chen BW, Rho S et al (2016) Optimal filter based on scale-invariance generation of natural images. J Supercomput 72(1):5–23CrossRefGoogle Scholar
  34. 34.
    Jiang F, Ji X, Hu C, Liu S, Zhao D (2014) Compressed vision information restoration based on cloud prior and local prior. IEEE 2:1117–1127Google Scholar
  35. 35.
    He K, Zhang X, Ren S, Sun J (2015) Delving deep into rectifiers: surpassing human-level performance on imagenet classification. In: IEEE International Conference on Computer Vision pp 1026–1034Google Scholar
  36. 36.
    Kingma D, Ba J (2015) Adam: a method for stochastic optimization. Paper at the 3rd International Conference for Learning RepresentationsGoogle Scholar
  37. 37.
    Vedaldi A, Lenc K (2015) Matconvnet: convolutional neural networks for matlab. In: Proceedings of the 23rd Annual ACM Conference on Multimedia Conference, pp 689–692Google Scholar
  38. 38.
    Chen BW, Wang JC, Wang JF (2009) A novel video summarization based on mining the story-structure and semantic relations among concept entities. IEEE Trans Multimed 11(2):295–312CrossRefGoogle Scholar
  39. 39.
    Chen BW, Chen CY, Wang JF (2013) Smart homecare surveillance system: behavior identification based on state transition support vector machines and sound directivity pattern analysis. IEEE Trans Syst Man Cybern Syst 43(6):1279–1289CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media New York 2017

Authors and Affiliations

  • Worku Jifara
    • 1
  • Feng Jiang
    • 1
    Email author
  • Seungmin Rho
    • 2
  • Maowei Cheng
    • 3
  • Shaohui Liu
    • 1
  1. 1.School of Computer Science and TechnologyHarbin Institute of TechnologyHarbinChina
  2. 2.Department of Media SoftwareSungkyul UniversityAnyangKorea
  3. 3.College of Command Information SystemPLA University of Science and TechnologyNanjingChina

Personalised recommendations