Please use this identifier to cite or link to this item: http://repo.lib.jfn.ac.lk/ujrr/handle/123456789/2542
Full metadata record
DC FieldValueLanguage
dc.contributor.authorSiyamalan, M.
dc.date.accessioned2021-04-20T03:02:33Z
dc.date.accessioned2022-06-28T04:51:46Z-
dc.date.available2021-04-20T03:02:33Z
dc.date.available2022-06-28T04:51:46Z-
dc.date.issued2017
dc.identifier.issn0278-0062
dc.identifier.urihttp://repo.lib.jfn.ac.lk/ujrr/handle/123456789/2542-
dc.description.abstractWe propose a novel multiple-instance learning (MIL) method to assess the visibility (visible/not visible) of the retinal nerve fiber layer (RNFL) in fundus camera images. Using only image-level labels, our approach learns to classify the images as well as to localize the RNFL visible regions. We transform the original feature space into a discriminative subspace, and learn a region-level classifier in that subspace.We propose a margin-based loss function to jointly learn this subspace and the region-level classifier. Experiments with an RNFL data set containing 884 images annotatedby two ophthalmologistsgive a system-annotator agreement (kappa values) of 0.73 and 0.72, respectively, with an interannotator agreement of 0.73. Our system agrees better with the more experienced annotator. Comparative tests with three public data sets (MESSIDOR and DR for diabetic retinopathy, and UCSB for breast cancer) show that our novel MIL approach improves performance over the state of the art. Our MATLAB code is publicly available at https://github.com/ManiShiyam/Sub-categoryclassifiers- for-Multiple-Instance-Learning/wiki.
dc.language.isoenen_US
dc.subjectImage classificationen_US
dc.subjectMultiple-instance learning(MIL)en_US
dc.subjectRetinalbiomarkers fordementiaen_US
dc.subjectRetinal image processingen_US
dc.subjectRetinal nerve fiber layer (RNFL)en_US
dc.titleSubcategory classifiers for multiple-instance learning and its application to retinal nerve fiber layer visibility classificationen_US
dc.typeArticleen_US
Appears in Collections:Computer Science

Files in This Item:
File Description SizeFormat 
Subcategory Classifiers for Multiple-Instance.pdf331.12 kBAdobe PDFThumbnail
View/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.