Please use this identifier to cite or link to this item: http://repo.lib.jfn.ac.lk/ujrr/handle/123456789/10042
Full metadata record
DC FieldValueLanguage
dc.contributor.authorGowtham, P.-
dc.contributor.authorNiranjan, M.-
dc.contributor.authorKaneswaran, A.-
dc.date.accessioned2024-01-16T04:42:56Z-
dc.date.available2024-01-16T04:42:56Z-
dc.date.issued2022-
dc.identifier.urihttp://repo.lib.jfn.ac.lk/ujrr/handle/123456789/10042-
dc.description.abstractImpressive high performance reported in the use of machine learning on computer vision problems is often due to the availability of very large datasets with which deep neural networks can be trained. With inference from medical images, however, this is not the case and available data is often only a small fraction in size in comparison to benchmark natural scene recognition problems. To circumvent this problem, transfer learning is often applied, where a model trained on a large natural image corpus is adapted, or pre-trained, to model the medical problem. In this work, we consider transfer learning applied to a specific medical diagnostics problem, that of abnormality detection in the gastrointestinal tract of a human body using images obtained during endoscopy. We carry out a search over several image recognition architectures and adapt pretrained models to the endoscopy problem. Using the benchmark KVASIR dataset, we show that transfer learning is effective in outperforming previously reported results, at an accuracy of 98.5±0.27.en_US
dc.language.isoenen_US
dc.publisherIEEEen_US
dc.subjectEndoscopyen_US
dc.subjectGastrointestinal abnormalitiesen_US
dc.subjectTransfer Learningen_US
dc.titleAutomated gastrointestinal abnormalities detection from endoscopic imagesen_US
dc.typeArticleen_US
Appears in Collections:Computer Engineering

Files in This Item:
File Description SizeFormat 
Automated gastrointestinal abnormalities detection from endoscopic images.pdf5.45 MBAdobe PDFView/Open


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