Matches in SemOpenAlex for { <https://semopenalex.org/work/W4367011678> ?p ?o ?g. }
Showing items 1 to 88 of
88
with 100 items per page.
- W4367011678 endingPage "646" @default.
- W4367011678 startingPage "635" @default.
- W4367011678 abstract "Children and adults with brain tumors are among the top causes of death in the world. Deep learning-based approaches have seen considerable success in brain tumor segmentation in recent years. Medical image tasks like tumor segmentation are severely limited by the time and effort necessary to collect paired medical imaging datasets. When gathering multi-modal image data, the problem becomes more complicated. However, this problem can be overcome by leveraging generative adversarial networks (GAN), which can generate synthesized images. The proposed automated brain tumor segmentation is made up of two modules. The first is a DCGAN network, which produces a binary tumor mask that is to be overlayed on a healthy brain image. Then, the overlayed brain image is used as an input for the pix2pix GAN network that applies for style transfer and generates realistic output that is indistinguishable from a real input. Thus, we are able to generate synthetic paired data to augment our dataset. The next module is the segmentation model that comprises a U-Net architecture that is optimized with residual blocks, the addition of residual blocks improves training time, training accuracy, and validation accuracy. The experimental results show that the synthesized dataset created by our proposed method and optimized U-Net model significantly improve the tumor segmentation performance and achieves the dice and IOU scores of 0.93 and 0.87, respectively." @default.
- W4367011678 created "2023-04-27" @default.
- W4367011678 creator A5010828356 @default.
- W4367011678 creator A5033606133 @default.
- W4367011678 creator A5039050408 @default.
- W4367011678 creator A5042603399 @default.
- W4367011678 creator A5051882518 @default.
- W4367011678 date "2023-01-01" @default.
- W4367011678 modified "2023-10-18" @default.
- W4367011678 title "Automated Brain Tumor Segmentation Using GAN Augmentation and Optimized U-Net" @default.
- W4367011678 cites W1901129140 @default.
- W4367011678 cites W2791575870 @default.
- W4367011678 cites W2793678581 @default.
- W4367011678 cites W2806051477 @default.
- W4367011678 cites W2884065486 @default.
- W4367011678 cites W2911631536 @default.
- W4367011678 cites W2963073614 @default.
- W4367011678 cites W2979754953 @default.
- W4367011678 cites W2984306354 @default.
- W4367011678 cites W2992511782 @default.
- W4367011678 cites W3000658149 @default.
- W4367011678 cites W3045950046 @default.
- W4367011678 cites W3120990782 @default.
- W4367011678 cites W3201722080 @default.
- W4367011678 cites W3209037005 @default.
- W4367011678 cites W4288086194 @default.
- W4367011678 doi "https://doi.org/10.1007/978-981-19-5191-6_51" @default.
- W4367011678 hasPublicationYear "2023" @default.
- W4367011678 type Work @default.
- W4367011678 citedByCount "0" @default.
- W4367011678 crossrefType "book-chapter" @default.
- W4367011678 hasAuthorship W4367011678A5010828356 @default.
- W4367011678 hasAuthorship W4367011678A5033606133 @default.
- W4367011678 hasAuthorship W4367011678A5039050408 @default.
- W4367011678 hasAuthorship W4367011678A5042603399 @default.
- W4367011678 hasAuthorship W4367011678A5051882518 @default.
- W4367011678 hasConcept C108583219 @default.
- W4367011678 hasConcept C11413529 @default.
- W4367011678 hasConcept C115961682 @default.
- W4367011678 hasConcept C124504099 @default.
- W4367011678 hasConcept C14166107 @default.
- W4367011678 hasConcept C142724271 @default.
- W4367011678 hasConcept C153180895 @default.
- W4367011678 hasConcept C154945302 @default.
- W4367011678 hasConcept C155512373 @default.
- W4367011678 hasConcept C22029948 @default.
- W4367011678 hasConcept C2524010 @default.
- W4367011678 hasConcept C2779130545 @default.
- W4367011678 hasConcept C31972630 @default.
- W4367011678 hasConcept C33923547 @default.
- W4367011678 hasConcept C41008148 @default.
- W4367011678 hasConcept C71924100 @default.
- W4367011678 hasConcept C89600930 @default.
- W4367011678 hasConceptScore W4367011678C108583219 @default.
- W4367011678 hasConceptScore W4367011678C11413529 @default.
- W4367011678 hasConceptScore W4367011678C115961682 @default.
- W4367011678 hasConceptScore W4367011678C124504099 @default.
- W4367011678 hasConceptScore W4367011678C14166107 @default.
- W4367011678 hasConceptScore W4367011678C142724271 @default.
- W4367011678 hasConceptScore W4367011678C153180895 @default.
- W4367011678 hasConceptScore W4367011678C154945302 @default.
- W4367011678 hasConceptScore W4367011678C155512373 @default.
- W4367011678 hasConceptScore W4367011678C22029948 @default.
- W4367011678 hasConceptScore W4367011678C2524010 @default.
- W4367011678 hasConceptScore W4367011678C2779130545 @default.
- W4367011678 hasConceptScore W4367011678C31972630 @default.
- W4367011678 hasConceptScore W4367011678C33923547 @default.
- W4367011678 hasConceptScore W4367011678C41008148 @default.
- W4367011678 hasConceptScore W4367011678C71924100 @default.
- W4367011678 hasConceptScore W4367011678C89600930 @default.
- W4367011678 hasLocation W43670116781 @default.
- W4367011678 hasOpenAccess W4367011678 @default.
- W4367011678 hasPrimaryLocation W43670116781 @default.
- W4367011678 hasRelatedWork W1669643531 @default.
- W4367011678 hasRelatedWork W2110230079 @default.
- W4367011678 hasRelatedWork W2117664411 @default.
- W4367011678 hasRelatedWork W2117933325 @default.
- W4367011678 hasRelatedWork W2122581818 @default.
- W4367011678 hasRelatedWork W2159066190 @default.
- W4367011678 hasRelatedWork W2739874619 @default.
- W4367011678 hasRelatedWork W2790662084 @default.
- W4367011678 hasRelatedWork W2948658236 @default.
- W4367011678 hasRelatedWork W4243168368 @default.
- W4367011678 isParatext "false" @default.
- W4367011678 isRetracted "false" @default.
- W4367011678 workType "book-chapter" @default.