Document Type : Original Research Paper


1 Digital Communications Signal Processing (DCSP) Research Lab., Faculty of Electrical and Computer Engineering, Shahid Rajaee Teacher Training University (SRTTU), Tehran, Iran.

2 Department of Biomedical Engineering, College of Engineering and School of Medicine, Wayne State University, Detroit, MI 48201, USA

3 Brain& Intelligent Systems Research Lab., Department of Electrical and Computer Engineering, Shahid Rajaee Teacher Training University, Tehran, Iran


Optical Coherence Tomography (OCT) uses the spatial and temporal coherence properties of optical waves backscattered from a tissue sample to form an image. An inherent characteristic of coherent imaging is the presence of speckle noise. In this study we use a new ensemble framework which is a combination of several Multi-Layer Perceptron (MLP) neural networks to denoise OCT images. The noise is modeled using Rayleigh distribution with the noise parameter, sigma, estimated by the ensemble framework. The input to the framework is a set of intensity and wavelet statistical features computed from the input image, and the output is the estimated sigma value for the noise model. In this article the methodology of this technique is explained.


[1] M.  Nasiriavanaki,  Adrian  GhPodoleanu,  John  B.  Schofield,  Carole  Jones, Manu Sira, Yan Liu, and Ali Hojjat. "Quantitative evaluation of scattering in optical coherence tomography skin  images  using  the  extended  Huygens–Fresnel  theorem."  Applied Optics Journal, Vol. 52, No. 8, pp. 1574‐1580, 2013.  [2] Mohammad  R.  N.  Avanaki,  Adrian  Gh.  Podoleanu,  Mark  C.  Price,  Serena  A.    Corr,  and  S.  A.  Hojjatoleslami,  “Two  applications  of  solid phantoms in  performance assessment  of  optical  coherence  tomography  systems,”  Applied  Optics  Journal, Vol. 52, No. 29, pp. 7054‐7061, 2013. 
[3] Mohammad R. N. Avanaki, Ali Hojjatoleslami, Mano Sira,  John  B.  Schofield,  Carole  Jones,  and  Adrian  GhPodoleanu.  "Investigation  of  basal  cell  carcinoma  using  dynamic  focus  optical  coherence  tomography."  Applied  Optics,  Vol.    52,  No.  10, pp. 2116‐2124, 2013. 
[4] J.  W.  Goodman,  Speckle  Phenomena  in  Optics:  Theory  and  Applications. Roberts & Co, 2006. 
  [5] M.  R.  N.  Avanaki,  P.  Philippe  Laissue,  and  Ali  Hojjatoleslami.  "De‐noising  speckled  optical  coherence  tomography  images  using an algorithm based on artificial neural network." Journal of Neuroscience and Neuroengineering, Vol. 2, No. 4, pp. 347‐ 352, 2013. 
[6] P. A. Magnin, O. T. von Ramm, and F. L. Thurstone, "Frequency  compounding for speckle contrast reduction in   phased array  images," Ultrason. Imaging, Vol. 4, pp. 267‐281, 1982. 
[7] Mohammad R. N. Avanaki, P. Philippe Laissue, Tae  JoongEom,  Adrian  Gh.  Podoleanu,  and  Ali  Hojjatoleslami,  “Speckle  reduction  using  an  artificial  neural  network  algorithm,”  Applied Optics Journal, Vol.52, No. 21, pp. 5050‐5057, 2013. 
[8] T. M. Jorgensen, L. Thrane, M. Mogensen, F. Pedersen, and P. E. Andersen,  “Speckle  reduction  in  optical  coherence  tomography  images  of  human  skin  by  a  spatial  diversity  method,” Proc. SPIE. 6627, 66270P, 2007.
  [9] N. Iftimia, B. E. Bouma, and G. J. Tearney, “Speckle reduction in  optical coherence tomography by path length encoded angular  compounding,” J. Bio. Opt. Vol. 8, pp. 260–263, 2003.  [10] R. K. Wang, "Reduction of speckle noise  for optical coherence  tomography  by  the  use  of  nonlinear  anisotropic  diffusion,"  Proc. SPIE. 5690, pp. 380‐385, 2005. 
[11] Mohammad  R.  N.  Avanaki,  R.  Cernat,  Paul  J.  Tadrous,  TaranTatla,  Adrian  Gh.  Podoleanu,  and  S.  Ali  Hojjatoleslami,  “Spatial  compounding  algorithm  for  speckle  reduction  of  dynamic  focus  OCT  images,”  IEEE  Photonics  Technology  Letters, Vol. 25, No. 15, pp.1439, 2013.  
[12] A.  Ozcan,  A.  Bilenca,  A.  E.  Desjardins,  B.  E.  Bouma,  and  G.  J.  Tearney, "Speckle reduction in optical coherence  tomography  images  using  digital  filtering,"  Scanning,  Vol.  20,  pp.  27‐30,  2007. 
[13] P.  Puvanathasan,  K.  Bizheva,  “Speckle  noise  reduction  algorithm for optical coherence tomography based on interval  type  II  fuzzy  set,” Optics  Express,  Vol.  15, No.  24,  pp.  15747‐ 15758, 2007. 
[14] Mohammad R.N. Avanaki, P.P. Laissue, A. Gh. Podoleanu, and A.  Hojjatoleslami,  "Evaluation  of  wavelet  mother  functions  for  speckle  noise  suppression  in  OCT  images,"  International  Journal  on Graphics, Bioinformatics and Medical Engineering,  Vol. 11, Issue 1, pp. 1‐5, 2011. 
[15] D. A. Nix, A. S. Weigend, “Estimating the Mean and Variance of  the Target Probability Distribution," Proc. of the IEEE Int. Conf.  on Neural Networks (IEEE‐ICNN'94), pp. 55‐60, 1994.  [16] H.  Schioler,  P.  Kulczycki,  “Neural  Network  for  Estimating  Conditional  Distributions,"  IEEE  Trans.  Neural  Networks,  Vol.8, No. 5, pp. 1015‐1025,1997. 
[17] M. R. Nasiriavanaki, Jun Xia, Hanlin Wan, Adam Quentin Bauer,  Joseph  P.  Culver,  and  Lihong  V.  Wang.  "High‐resolution  photoacoustic  tomography  of  resting‐state  functional  connectivity in  the mouse brain." Proceedings of  the National  Academy of Sciences, Vol. 111, No. 1 pp. 21‐26, 2014.
  [18] M.  R.  Nasiri‐Avanaki,  Reza  Ebrahimpour.  "In‐service  video  quality  measurements  in  optical  fiber  links  based  on  neural  network." Neural Network World 17.5 (2007): 457. 


Journal of Electrical and Computer Engineering Innovations (JECEI) welcomes letters to the editor for the post-publication discussions and corrections which allows debate post publication on its site, through the Letters to Editor. Letters pertaining to manuscript published in JECEI should be sent to the editorial office of JECEI within three months of either online publication or before printed publication, except for critiques of original research. Following points are to be considering before sending the letters (comments) to the editor.

[1] Letters that include statements of statistics, facts, research, or theories should include appropriate references, although more than three are discouraged.

[2] Letters that are personal attacks on an author rather than thoughtful criticism of the author’s ideas will not be considered for publication.

[3] Letters can be no more than 300 words in length.

[4] Letter writers should include a statement at the beginning of the letter stating that it is being submitted either for publication or not.

[5] Anonymous letters will not be considered.

[6] Letter writers must include their city and state of residence or work.

[7] Letters will be edited for clarity and length.