Multi-scale deep feature fusion for automated classification of macular pathologies from OCT images. (September 2019)

• Record Type: Journal Article
• Authors: Das, Vineeta; Dandapat, Samarendra; Bora, Prabin Kumar
• Is Part Of: Biomedical signal processing and control. Volume 54(2019)
• Publication Date: 2019-09
• Subjects: Optical coherence tomography (OCT) -- Multi-scale -- Convolutional neural network (CNN) -- Feature fusion -- Classification -- Macular pathology; Signal processing -- Periodicals; Biomedical engineering -- Periodicals; Signal Processing, Computer-Assisted -- Periodicals; Image Processing, Computer-Assisted -- Periodicals; Biomedical Engineering -- Periodicals; 610.28
• DOI: 10.1016/j.bspc.2019.101605
• Languages: English
• ISSNs: 1746-8094

Abstract

Highlights: We present a new multi-scale deep feature fusion (MDFF) based approach for classification of macular pathologies. Optical coherence tomography images are used for pathology classification. The proposed method fuses information from multiple image scales to capture the inter-scale variations among the pathological manifestations. The method does not require any speckle removal step. The proposed method provides promising classification accuracy which makes it highly suitable for use in eye clinics. Abstract: Identification of the macular pathologies at an early stage can prevent vision loss. Similarity in the pathological manifestations of common macular disorders like age related macular degeneration (AMD) and diabetic macular edema (DME) can make manual screening fallible. There is a growing interest among researchers for reliable automated detection of macular pathologies using computer methods. Therefore, in this paper we present a novel method for classification of DME and two stages of AMD namely the drusens (early stage) and the choroidal neo vascularization (CNV) (late stage) from healthy optical coherence tomography (OCT) images. The proposed method introduces a multi-scale deep feature fusion (MDFF) based classification approach using convolutional neural network (CNN) for reliable diagnosis. The MDFF captures the inter-scale variations in images to introduce discriminative and complementary information to the classifier. The proposed method is evaluated on an OCT dataset containing 84, 484 images with different class distributions. The imbalance in the dataset is handled by introducing the cost sensitive loss function during the learning of the classifier. The proposed method achieves an average sensitivity, specificity and accuracy of 99.6%, 99.87% and 99.6% on the test set. The promising classification results make the proposed method highly suitable for preliminary automated diagnosis of macular pathologies in health care centres and eye clinics.