Glaucoma detection using entropy sampling and ensemble learning for automatic optic cup and disc segmentation

• Published in: Computerized medical imaging and graphics Vol. 55; pp. 28 - 41
• Main Authors: Zilly, Julian, Buhmann, Joachim M., Mahapatra, Dwarikanath
• Format: Journal Article
• Language: English
• Published: United States Elsevier Ltd 01.01.2017; Elsevier Science Ltd
• Subjects: Artificial neural networks; Boosting; Classifiers; CNN; Computer applications; Ensemble learning; Entropy; Filters; Glaucoma; Glaucoma - diagnostic imaging; Humans; Image processing; Image segmentation; Internal Medicine; Learning; Logistic Models; Neural networks; Neural Networks (Computer); Optic cup; Optic disc; Optic Disk - diagnostic imaging; Other; Pattern Recognition, Automated - methods; Retina; Retinal images; Sampling; Segmentation; Unsupervised Machine Learning; Glaucoma; CNN; Segmentation; Boosting; Optic disc; Ensemble learning; Optic cup
• ISSN: 0895-6111; 1879-0771; 1879-0771
• DOI: 10.1016/j.compmedimag.2016.07.012

•An ensemble learning based architecture to learn convolutional filters.•Use of boosting as a computationally efficient learning framework.•Accurate networks learned from few sample data. We present a novel method to segment retinal images using ensemble learning based convolutional neural network (CNN) architectures. An entropy sampling technique is used to select informative points thus reducing computational complexity while performing superior to uniform sampling. The sampled points are used to design a novel learning framework for convolutional filters based on boosting. Filters are learned in several layers with the output of previous layers serving as the input to the next layer. A softmax logistic classifier is subsequently trained on the output of all learned filters and applied on test images. The output of the classifier is subject to an unsupervised graph cut algorithm followed by a convex hull transformation to obtain the final segmentation. Our proposed algorithm for optic cup and disc segmentation outperforms existing methods on the public DRISHTI-GS data set on several metrics.