Matches in SemOpenAlex for { <https://semopenalex.org/work/W4310584574> ?p ?o ?g. }
Showing items 1 to 82 of
82
with 100 items per page.
- W4310584574 abstract "Deep learning models can enable more accurate and efficient segmentation of cardiac structures in echocardiography (echo). However, their success depends on the availability of large-scale annotated training data, whose achievement is highly challenging in medical imaging. In this paper, we propose a novel data augmentation approach to tackle the scarcity of annotated data in automatic echocardiography segmentation. The proposed approach termed DF-Aug (Disentanglement and Fusion Augmentation), consists of two main steps: image disentanglement and image fusion. First, given that domain shift in medical image data is usually reflected on the image texture component, cartoon + texture decomposition is applied to split an image into its cartoon component and texture component. Second, every image of the available annotated dataset is embedded with the texture component of a randomly selected unlabeled image, resulting in a new annotated image with different features. The feasibility of the proposed approach in overcoming the shortage of labeled medical image dataset is demonstrated on the task of left ventricle (LV) segmentation in echocardiography. Extensive experiments conducted on two different datasets demonstrate that using the generated samples results in remarkable performance enhancement of the deep learning framework." @default.
- W4310584574 created "2022-12-12" @default.
- W4310584574 creator A5011509629 @default.
- W4310584574 creator A5016504488 @default.
- W4310584574 creator A5020542342 @default.
- W4310584574 creator A5042212002 @default.
- W4310584574 creator A5055727134 @default.
- W4310584574 date "2022-10-10" @default.
- W4310584574 modified "2023-10-01" @default.
- W4310584574 title "A Disentanglement and Fusion Data Augmentation Approach for Echocardiography Segmentation" @default.
- W4310584574 cites W2092663520 @default.
- W4310584574 cites W2166493908 @default.
- W4310584574 cites W2884065486 @default.
- W4310584574 cites W2998635511 @default.
- W4310584574 cites W3026847561 @default.
- W4310584574 cites W3036107862 @default.
- W4310584574 cites W3103215654 @default.
- W4310584574 cites W3213656113 @default.
- W4310584574 cites W4200191315 @default.
- W4310584574 doi "https://doi.org/10.1109/ius54386.2022.9958234" @default.
- W4310584574 hasPublicationYear "2022" @default.
- W4310584574 type Work @default.
- W4310584574 citedByCount "0" @default.
- W4310584574 crossrefType "proceedings-article" @default.
- W4310584574 hasAuthorship W4310584574A5011509629 @default.
- W4310584574 hasAuthorship W4310584574A5016504488 @default.
- W4310584574 hasAuthorship W4310584574A5020542342 @default.
- W4310584574 hasAuthorship W4310584574A5042212002 @default.
- W4310584574 hasAuthorship W4310584574A5055727134 @default.
- W4310584574 hasConcept C108583219 @default.
- W4310584574 hasConcept C115961682 @default.
- W4310584574 hasConcept C121332964 @default.
- W4310584574 hasConcept C124504099 @default.
- W4310584574 hasConcept C138885662 @default.
- W4310584574 hasConcept C153180895 @default.
- W4310584574 hasConcept C154945302 @default.
- W4310584574 hasConcept C168167062 @default.
- W4310584574 hasConcept C194051981 @default.
- W4310584574 hasConcept C2778137410 @default.
- W4310584574 hasConcept C2781195486 @default.
- W4310584574 hasConcept C31972630 @default.
- W4310584574 hasConcept C41008148 @default.
- W4310584574 hasConcept C41895202 @default.
- W4310584574 hasConcept C63099799 @default.
- W4310584574 hasConcept C69744172 @default.
- W4310584574 hasConcept C89600930 @default.
- W4310584574 hasConcept C97355855 @default.
- W4310584574 hasConceptScore W4310584574C108583219 @default.
- W4310584574 hasConceptScore W4310584574C115961682 @default.
- W4310584574 hasConceptScore W4310584574C121332964 @default.
- W4310584574 hasConceptScore W4310584574C124504099 @default.
- W4310584574 hasConceptScore W4310584574C138885662 @default.
- W4310584574 hasConceptScore W4310584574C153180895 @default.
- W4310584574 hasConceptScore W4310584574C154945302 @default.
- W4310584574 hasConceptScore W4310584574C168167062 @default.
- W4310584574 hasConceptScore W4310584574C194051981 @default.
- W4310584574 hasConceptScore W4310584574C2778137410 @default.
- W4310584574 hasConceptScore W4310584574C2781195486 @default.
- W4310584574 hasConceptScore W4310584574C31972630 @default.
- W4310584574 hasConceptScore W4310584574C41008148 @default.
- W4310584574 hasConceptScore W4310584574C41895202 @default.
- W4310584574 hasConceptScore W4310584574C63099799 @default.
- W4310584574 hasConceptScore W4310584574C69744172 @default.
- W4310584574 hasConceptScore W4310584574C89600930 @default.
- W4310584574 hasConceptScore W4310584574C97355855 @default.
- W4310584574 hasFunder F4320322919 @default.
- W4310584574 hasLocation W43105845741 @default.
- W4310584574 hasOpenAccess W4310584574 @default.
- W4310584574 hasPrimaryLocation W43105845741 @default.
- W4310584574 hasRelatedWork W1669643531 @default.
- W4310584574 hasRelatedWork W2117933325 @default.
- W4310584574 hasRelatedWork W2122581818 @default.
- W4310584574 hasRelatedWork W2159066190 @default.
- W4310584574 hasRelatedWork W2739874619 @default.
- W4310584574 hasRelatedWork W2790662084 @default.
- W4310584574 hasRelatedWork W2897195263 @default.
- W4310584574 hasRelatedWork W2948658236 @default.
- W4310584574 hasRelatedWork W4247340124 @default.
- W4310584574 hasRelatedWork W2476566122 @default.
- W4310584574 isParatext "false" @default.
- W4310584574 isRetracted "false" @default.
- W4310584574 workType "article" @default.