Transfer learning based classification of optical coherence tomography images with diabetic macular edema and dry age-related macular degeneration

• Published in: Biomedical optics express Vol. 8; no. 2; pp. 579 - 592
• Main Authors: Karri, S. P. K., Chakraborty, Debjani, Chatterjee, Jyotirmoy
• Format: Journal Article
• Language: English
• Published: United States Optical Society of America 01.02.2017
• Subjects: (070.5010) Pattern recognition; (170.1610) Clinical applications; (100.2960) Image analysis; (110.4500) Optical coherence tomography; (170.4470) Ophthalmology
• ISSN: 2156-7085; 2156-7085
• DOI: 10.1364/BOE.8.000579

We present an algorithm for identifying retinal pathologies given retinal optical coherence tomography (OCT) images. Our approach fine-tunes a pre-trained convolutional neural network (CNN), GoogLeNet, to improve its prediction capability (compared to random initialization training) and identifies salient responses during prediction to understand learned filter characteristics. We considered a data set containing subjects with diabetic macular edema, or dry age-related macular degeneration, or no pathology. The fine-tuned CNN could effectively identify pathologies in comparison to classical learning. Our algorithm aims to demonstrate that models trained on non-medical images can be fine-tuned for classifying OCT images with limited training data.