21 November 2017 Classification of images based on small local features: a case applied to microaneurysms in fundus retina images
Author Affiliations +
Abstract
Convolutional neural networks (CNNs), the state of the art in image classification, have proven to be as effective as an ophthalmologist, when detecting referable diabetic retinopathy. Having a size of < 1 % of the total image, microaneurysms are early lesions in diabetic retinopathy that are difficult to classify. A model that includes two CNNs with different input image sizes, 60 × 60 and 420 × 420    pixels , was developed. These models were trained using the Kaggle and Messidor datasets and tested independently against the Kaggle dataset, showing a sensitivity > 91 % , a specificity > 93 % , and an area under the receiver operating characteristics curve > 93 % . Furthermore, by combining these trained models, there was a reduction of false positives for complete images by about 50% and a sensitivity of 96% when tested against the DiaRetDB1 dataset. In addition, a powerful image preprocessing procedure was implemented, improving not only images for annotations, but also decreasing the number of epochs during training. Finally, a feedback method was developed increasing the accuracy of the CNN 420 × 420    pixel input model.
© 2017 Society of Photo-Optical Instrumentation Engineers (SPIE)
Pablo F. Ordóñez, Carlos M. Cepeda, Jose Garrido, Sumit Chakravarty, "Classification of images based on small local features: a case applied to microaneurysms in fundus retina images," Journal of Medical Imaging 4(4), 041309 (21 November 2017). https://doi.org/10.1117/1.JMI.4.4.041309 . Submission: Received: 9 May 2017; Accepted: 25 October 2017
Received: 9 May 2017; Accepted: 25 October 2017; Published: 21 November 2017
JOURNAL ARTICLE
17 PAGES


SHARE
Back to Top