%0 Journal Article
%@ 01692607
%A Sidibé, D.
%A Sankar, S.
%A Lemaître, G.
%A Rastgoo, M.
%A Massich, J.
%A Cheung, C.Y.
%A Tan, G.S.W.
%A Milea, D.
%A Lamoureux, E.
%A Wong, T.Y.
%A Mériaudeau, F.
%D 2017
%F scholars:8893
%I Elsevier Ireland Ltd
%J Computer Methods and Programs in Biomedicine
%K Classification (of information); Coherent light; Eye protection; Gaussian distribution; Signal detection; Statistics; Strain measurement, Anomaly detection; Automatic classification; Classification performance; Diabetic retinopathy; Gaussian Mixture Model; Macular edema; SD-OCT; Spectral domain optical coherence tomographies, Optical tomography, Article; B scan; classification algorithm; clinical article; controlled study; diabetic macular edema; diagnostic test accuracy study; gaussian mixture model; human; image analysis; image processing; optical coherence tomography device; principal component analysis; sensitivity and specificity; spectral domain optical coherence tomography; statistical model; volume; complication; diabetes mellitus; diagnostic imaging; female; macular edema; male; optical coherence tomography, Diabetes Complications; Female; Humans; Macular Edema; Male; Tomography, Optical Coherence
%P 109-117
%R 10.1016/j.cmpb.2016.11.001
%T An anomaly detection approach for the identification of DME patients using spectral domain optical coherence tomography images
%U https://khub.utp.edu.my/scholars/8893/
%V 139
%X This paper proposes a method for automatic classification of spectral domain OCT data for the identification of patients with retinal diseases such as Diabetic Macular Edema (DME). We address this issue as an anomaly detection problem and propose a method that not only allows the classification of the OCT volume, but also allows the identification of the individual diseased B-scans inside the volume. Our approach is based on modeling the appearance of normal OCT images with a Gaussian Mixture Model (GMM) and detecting abnormal OCT images as outliers. The classification of an OCT volume is based on the number of detected outliers. Experimental results with two different datasets show that the proposed method achieves a sensitivity and a specificity of 80 and 93 on the first dataset, and 100 and 80 on the second one. Moreover, the experiments show that the proposed method achieves better classification performance than other recently published works. © 2016 Elsevier Ireland Ltd
%Z cited By 48