1Research Scholar, School of Electronics and Communication Engineering, Sathyabama Institute of Science and Technology, Chennai, India
2Professor, School of Electronics and Communication Engineering, Sathyabama Institute of Science and Technology, Chennai, India
*Corresponding Author: Venkatraman. K., Research Scholar, Sathyabama Institute of Science and Technology, Chennai-119, India, Email: venkat.biomed@gmail.com
Online published on 19 August, 2019.
Optical Coherence Tomography (OCT) being one of the vital diagnostic tool in early detection of Blindness, is most commonly used in analysis of fluid based abnormalities that are caused in the retinal layers. As the need for automations in OCT Image analysis has elevated presently, the proposed system focuses on automated classification of the input retinal image based on the fluid pattern as Normal, Cystoid Macular Edema (CME), Choroidal Neo Vascular Membrane (CNVM), and Macular Hole (MH). The system analyses the images from TOPCONN and ZEISS Equipment, totally 114 in numbers. Median Filters have been implemented for preprocessing/noise removal, followed by active contour segmentation for retinal boundaries. Various Marphological (HOG) Features have been extracted and used for classification with k-NN Classifier. The developed system shows an accuracy of 89.29% with significant Diagnostic Odd Ratio of 2.0.
Retinal Disorder, Optical Coherence Tomography, Fluid pattern
