Classification of SD-OCT Volumes using Local Binary Patterns: Experimental Validation for DME detection

Lemaitre, Guillaume and Rastgoo, Mojdeh and Massich, Joan and Cheung, Carol Y. and Wong, Tien Y. and Lamoureux, Ecosse and Milea, Dan and Meriaudeau, Fabrice and Sidibe, Desire

@article{lemaitre2016classification,
  title={Classification of SD-OCT Volumes using Local Binary Patterns: Experimental Validation for DME detection},
  author={Lema\^{i}tre, Guillaume and Rastgoo, Mojdeh and Massich, Joan and Cheung, Carol Y. and Wong, Tien Y. and Lamoureux, Ecosse and Milea, Dan and M\'{e}riaudeau, Fabrice and Sidib\'{e}, D\'{e}sir\'{e}},
  journal={Journal of Ophthalmology},
  volume={2016},
  year={2016},
  publisher={Hindawi Publishing Corporation}
}

Graphical abstract

Image comes here.

Abstract

This paper addresses the problem of automatic classification of SD-OCT data for automatic identification of patients with DME versus normal subjects. OCT has been a valuable diagnostic tool for DME, which is among the most common causes of irreversible vision loss in individuals with diabetes. Here, a classification framework with five distinctive steps is proposed and we present an extensive study of each step. Our method considers combination of various pre-processings in conjunction with LBP features and different mapping strategies. Using linear and non-linear classifiers, we tested the developed framework on a balanced cohort of 32 patients.

Experimental results show that the proposed method outperforms the previous studies by achieving a SE and SP of 81.2% and 93.7%, respectively. Our study concludes that the 3D features and high-level representation of 2D features using patches achieve the best results. However, the effects of pre-processing is inconsistent with respect to different classifiers and feature configurations.

Notes

Notes come here.