A computer‐aided diagnostic system for detecting diabetic retinopathy in optical coherence tomography images

• Published in: Medical physics (Lancaster) Vol. 44; no. 3; pp. 914 - 923
• Main Authors: ElTanboly, Ahmed, Ismail, Marwa, Shalaby, Ahmed, Switala, Andy, El‐Baz, Ayman, Schaal, Shlomit, Gimel’farb, Georgy, El‐Azab, Magdi
• Format: Journal Article
• Language: English
• Published: United States 01.03.2017
• Subjects: Adult; Aged; diabetic retinopathy (DR); Diabetic Retinopathy - diagnostic imaging; Female; Humans; Image Interpretation, Computer-Assisted - methods; joint image‐region‐map model; Machine Learning; Male; Markov–Gibbs random field (MGRF); Middle Aged; non‐negativity‐constrained autoencoder (NCAE); optical coherence tomography (OCT); Pattern Recognition, Automated; Retina - diagnostic imaging; Sensitivity and Specificity; Tomography, Optical Coherence - methods; joint image-region-map model; Markov-Gibbs random field (MGRF); non-negativity-constrained autoencoder (NCAE); optical coherence tomography (OCT); diabetic retinopathy (DR)
• ISSN: 0094-2405; 2473-4209; 2473-4209
• DOI: 10.1002/mp.12071

Purpose Detection (diagnosis) of diabetic retinopathy (DR) in optical coherence tomography (OCT) images for patients with type 2 diabetes, but almost clinically normal retina appearances. Methods The proposed computer‐aided diagnostic (CAD) system detects the DR in three steps: (a) localizing and segmenting 12 distinct retinal layers on the OCT image; (b) deriving features of the segmented layers, and (c) learning most discriminative features and classifying each subject as normal or diabetic. To localise and segment the retinal layers, signals (intensities) of the OCT image are described with a joint Markov‐Gibbs random field (MGRF) model of intensities and shape descriptors. Each segmented layer is characterized with cumulative probability distribution functions (CDF) of its locally extracted features, such as reflectivity, curvature, and thickness. A multistage deep fusion classification network (DFCN) with a stack of non‐negativity‐constrained autoencoders (NCAE) is trained to select the most discriminative retinal layers’ features and use their CDFs for detecting the DR. A training atlas was built using the OCT scans for 12 normal subjects and their maps of layers hand‐drawn by retina experts. Results Preliminary experiments on 52 clinical OCT scans (26 normal and 26 with early‐stage DR, balanced between 40–79 yr old males and females; 40 training and 12 test subjects) gave the DR detection accuracy, sensitivity, and specificity of 92%; 83%, and 100%, respectively. The 100% accuracy, sensitivity, and specificity have been obtained in the leave‐one‐out cross‐validation test for all the 52 subjects. Conclusion Both the quantitative and visual assessments confirmed the high accuracy of the proposed computer‐assisted diagnostic system for early DR detection using the OCT retinal images.