Matches in SemOpenAlex for { <https://semopenalex.org/work/W4221011406> ?p ?o ?g. }
Showing items 1 to 89 of
89
with 100 items per page.
- W4221011406 abstract "Precise and robust multimodal image registration helps in spatially aligning medical finding from multiple sources for accurate medical diagnosis. We present MedRegNet, a lightweight descriptor module for registration multimodal retinal images. We utilize generative adversary networks (GANs) to learn generator networks which during training can synthesize structurally consistent multimodal image-pairs. From these image-pairs MedRegNet learns to predict stable point descriptors via a relaxed ranking loss. As no dataset-specific incentives are given, MedRegNet learns stable representations by itself. Most importantly, both mono- and multimodal training – including training the GANs – is entirely unsupervised, hence no costly expert annotation is required. Through evaluation on the publicly available Fundus Image Registration Dataset (FIRE) as well as on our own multimodal dataset containing 340 retinal fundus, autofluorescence and fluorescein angiography image-pairs from 24 patients we show MedRegNet to improve robustness and registration capabilities of classical detector/descriptor algorithms like SIFT, ORB, KAZE and AKAZE. Despite using the same interest points, MedRegNet is matching more points successfully than either baseline even in the monomodal use case. In the multimodal case, where the classical baselines fail due to large visual difference between the modalities, MedRegNet’s registration performance stays consistent to the monomodal case. Furthermore, MedRegNet shows to be adaptable to point detectors it was not trained on with no or only little cost in performance. MedRegNet can easily be integrated into any feature-based registration pipeline and due to the lack of dataset-specific incentives has the potential to be applied to fields outside retinal imaging." @default.
- W4221011406 created "2022-04-03" @default.
- W4221011406 creator A5022955518 @default.
- W4221011406 creator A5045272470 @default.
- W4221011406 creator A5053803665 @default.
- W4221011406 creator A5061926759 @default.
- W4221011406 creator A5072469553 @default.
- W4221011406 creator A5074395199 @default.
- W4221011406 creator A5075779604 @default.
- W4221011406 creator A5081604705 @default.
- W4221011406 date "2022-04-04" @default.
- W4221011406 modified "2023-09-25" @default.
- W4221011406 title "MedRegNet: unsupervised multimodal retinal-image registration with GANs and ranking loss" @default.
- W4221011406 cites W2008040967 @default.
- W4221011406 cites W2056425777 @default.
- W4221011406 cites W2058333183 @default.
- W4221011406 cites W2085261163 @default.
- W4221011406 cites W2103585593 @default.
- W4221011406 cites W2107823820 @default.
- W4221011406 cites W2117228865 @default.
- W4221011406 cites W2121068789 @default.
- W4221011406 cites W2125956239 @default.
- W4221011406 cites W2141085506 @default.
- W4221011406 cites W2146063841 @default.
- W4221011406 cites W2148900397 @default.
- W4221011406 cites W2151103935 @default.
- W4221011406 cites W2538579510 @default.
- W4221011406 cites W2544587078 @default.
- W4221011406 cites W2755033654 @default.
- W4221011406 cites W2766309214 @default.
- W4221011406 cites W2963752842 @default.
- W4221011406 cites W2994240761 @default.
- W4221011406 cites W3038202136 @default.
- W4221011406 cites W3194075191 @default.
- W4221011406 cites W4200179413 @default.
- W4221011406 cites W4237101215 @default.
- W4221011406 cites W4244382019 @default.
- W4221011406 cites W4248767311 @default.
- W4221011406 cites W4250685322 @default.
- W4221011406 doi "https://doi.org/10.1117/12.2607653" @default.
- W4221011406 hasPublicationYear "2022" @default.
- W4221011406 type Work @default.
- W4221011406 citedByCount "2" @default.
- W4221011406 countsByYear W42210114062022 @default.
- W4221011406 crossrefType "proceedings-article" @default.
- W4221011406 hasAuthorship W4221011406A5022955518 @default.
- W4221011406 hasAuthorship W4221011406A5045272470 @default.
- W4221011406 hasAuthorship W4221011406A5053803665 @default.
- W4221011406 hasAuthorship W4221011406A5061926759 @default.
- W4221011406 hasAuthorship W4221011406A5072469553 @default.
- W4221011406 hasAuthorship W4221011406A5074395199 @default.
- W4221011406 hasAuthorship W4221011406A5075779604 @default.
- W4221011406 hasAuthorship W4221011406A5081604705 @default.
- W4221011406 hasConcept C104317684 @default.
- W4221011406 hasConcept C115961682 @default.
- W4221011406 hasConcept C153180895 @default.
- W4221011406 hasConcept C154945302 @default.
- W4221011406 hasConcept C166704113 @default.
- W4221011406 hasConcept C185592680 @default.
- W4221011406 hasConcept C31972630 @default.
- W4221011406 hasConcept C41008148 @default.
- W4221011406 hasConcept C55493867 @default.
- W4221011406 hasConcept C63479239 @default.
- W4221011406 hasConceptScore W4221011406C104317684 @default.
- W4221011406 hasConceptScore W4221011406C115961682 @default.
- W4221011406 hasConceptScore W4221011406C153180895 @default.
- W4221011406 hasConceptScore W4221011406C154945302 @default.
- W4221011406 hasConceptScore W4221011406C166704113 @default.
- W4221011406 hasConceptScore W4221011406C185592680 @default.
- W4221011406 hasConceptScore W4221011406C31972630 @default.
- W4221011406 hasConceptScore W4221011406C41008148 @default.
- W4221011406 hasConceptScore W4221011406C55493867 @default.
- W4221011406 hasConceptScore W4221011406C63479239 @default.
- W4221011406 hasLocation W42210114061 @default.
- W4221011406 hasOpenAccess W4221011406 @default.
- W4221011406 hasPrimaryLocation W42210114061 @default.
- W4221011406 hasRelatedWork W1999222583 @default.
- W4221011406 hasRelatedWork W2035976912 @default.
- W4221011406 hasRelatedWork W2036807459 @default.
- W4221011406 hasRelatedWork W2109974539 @default.
- W4221011406 hasRelatedWork W2125927971 @default.
- W4221011406 hasRelatedWork W2541791370 @default.
- W4221011406 hasRelatedWork W2738084969 @default.
- W4221011406 hasRelatedWork W2954664659 @default.
- W4221011406 hasRelatedWork W3163375306 @default.
- W4221011406 hasRelatedWork W4293868162 @default.
- W4221011406 isParatext "false" @default.
- W4221011406 isRetracted "false" @default.
- W4221011406 workType "article" @default.