Fully automated macular pathology detection in retina optical coherence tomography images using sparse coding and dictionary learning

• Published in: Journal of biomedical optics Vol. 22; no. 1; p. 016012
• Main Authors: Sun, Yankui, Li, Shan, Sun, Zhongyang
• Format: Journal Article
• Language: English
• Published: United States Society of Photo-Optical Instrumentation Engineers 01.01.2017
• Subjects: Algorithms; Diabetic Retinopathy - complications; Diabetic Retinopathy - diagnostic imaging; Humans; Macular Degeneration - diagnostic imaging; Macular Degeneration - etiology; Macular Edema - diagnostic imaging; Macular Edema - etiology; Retina - diagnostic imaging; Tomography, Optical Coherence - methods; spatial pyramid matching; max pooling; optical coherence tomography; age-related macular degeneration; sparse coding; diabetic macular edema
• ISSN: 1083-3668; 1560-2281; 1560-2281
• DOI: 10.1117/1.JBO.22.1.016012

We propose a framework for automated detection of dry age-related macular degeneration (AMD) and diabetic macular edema (DME) from retina optical coherence tomography (OCT) images, based on sparse coding and dictionary learning. The study aims to improve the classification performance of state-of-the-art methods. First, our method presents a general approach to automatically align and crop retina regions; then it obtains global representations of images by using sparse coding and a spatial pyramid; finally, a multiclass linear support vector machine classifier is employed for classification. We apply two datasets for validating our algorithm: Duke spectral domain OCT (SD-OCT) dataset, consisting of volumetric scans acquired from 45 subjects-15 normal subjects, 15 AMD patients, and 15 DME patients; and clinical SD-OCT dataset, consisting of 678 OCT retina scans acquired from clinics in Beijing-168, 297, and 213 OCT images for AMD, DME, and normal retinas, respectively. For the former dataset, our classifier correctly identifies 100%, 100%, and 93.33% of the volumes with DME, AMD, and normal subjects, respectively, and thus performs much better than the conventional method; for the latter dataset, our classifier leads to a correct classification rate of 99.67%, 99.67%, and 100.00% for DME, AMD, and normal images, respectively.