Matches in SemOpenAlex for { <https://semopenalex.org/work/W4387211617> ?p ?o ?g. }
- W4387211617 endingPage "496" @default.
- W4387211617 startingPage "486" @default.
- W4387211617 abstract "Recent advancements in large-scale Vision Transformers have made significant strides in improving pre-trained models for medical image segmentation. However, these methods face a notable challenge in acquiring a substantial amount of pre-training data, particularly within the medical field. To address this limitation, we present Masked Multi-view with Swin Transformers (SwinMM), a novel multi-view pipeline for enabling accurate and data-efficient self-supervised medical image analysis. Our strategy harnesses the potential of multi-view information by incorporating two principal components. In the pre-training phase, we deploy a masked multi-view encoder devised to concurrently train masked multi-view observations through a range of diverse proxy tasks. These tasks span image reconstruction, rotation, contrastive learning, and a novel task that employs a mutual learning paradigm. This new task capitalizes on the consistency between predictions from various perspectives, enabling the extraction of hidden multi-view information from 3D medical data. In the fine-tuning stage, a cross-view decoder is developed to aggregate the multi-view information through a cross-attention block. Compared with the previous state-of-the-art self-supervised learning method Swin UNETR, SwinMM demonstrates a notable advantage on several medical image segmentation tasks. It allows for a smooth integration of multi-view information, significantly boosting both the accuracy and data-efficiency of the model. Code and models are available at https://github.com/UCSC-VLAA/SwinMM/ ." @default.
- W4387211617 created "2023-10-01" @default.
- W4387211617 creator A5009911035 @default.
- W4387211617 creator A5013038112 @default.
- W4387211617 creator A5022344556 @default.
- W4387211617 creator A5025852924 @default.
- W4387211617 creator A5048043129 @default.
- W4387211617 creator A5053401529 @default.
- W4387211617 creator A5067640436 @default.
- W4387211617 creator A5072796155 @default.
- W4387211617 creator A5077206511 @default.
- W4387211617 creator A5086706224 @default.
- W4387211617 date "2023-01-01" @default.
- W4387211617 modified "2023-10-01" @default.
- W4387211617 title "SwinMM: Masked Multi-view with Swin Transformers for 3D Medical Image Segmentation" @default.
- W4387211617 cites W1667869507 @default.
- W4387211617 cites W1901129140 @default.
- W4387211617 cites W1986649315 @default.
- W4387211617 cites W2065320551 @default.
- W4387211617 cites W2620998106 @default.
- W4387211617 cites W2804047627 @default.
- W4387211617 cites W2890236360 @default.
- W4387211617 cites W2921629474 @default.
- W4387211617 cites W2963418739 @default.
- W4387211617 cites W2964309882 @default.
- W4387211617 cites W2971013993 @default.
- W4387211617 cites W2979967919 @default.
- W4387211617 cites W3037260676 @default.
- W4387211617 cites W3049757379 @default.
- W4387211617 cites W3138516171 @default.
- W4387211617 cites W3165635828 @default.
- W4387211617 cites W3172681723 @default.
- W4387211617 cites W3183302989 @default.
- W4387211617 cites W3195361525 @default.
- W4387211617 cites W3204255739 @default.
- W4387211617 cites W4212875960 @default.
- W4387211617 cites W4284711968 @default.
- W4387211617 cites W4295940432 @default.
- W4387211617 cites W4298110661 @default.
- W4387211617 cites W4312428231 @default.
- W4387211617 cites W4313156423 @default.
- W4387211617 cites W4379619220 @default.
- W4387211617 cites W4382998948 @default.
- W4387211617 cites W4386352556 @default.
- W4387211617 doi "https://doi.org/10.1007/978-3-031-43898-1_47" @default.
- W4387211617 hasPublicationYear "2023" @default.
- W4387211617 type Work @default.
- W4387211617 citedByCount "0" @default.
- W4387211617 crossrefType "book-chapter" @default.
- W4387211617 hasAuthorship W4387211617A5009911035 @default.
- W4387211617 hasAuthorship W4387211617A5013038112 @default.
- W4387211617 hasAuthorship W4387211617A5022344556 @default.
- W4387211617 hasAuthorship W4387211617A5025852924 @default.
- W4387211617 hasAuthorship W4387211617A5048043129 @default.
- W4387211617 hasAuthorship W4387211617A5053401529 @default.
- W4387211617 hasAuthorship W4387211617A5067640436 @default.
- W4387211617 hasAuthorship W4387211617A5072796155 @default.
- W4387211617 hasAuthorship W4387211617A5077206511 @default.
- W4387211617 hasAuthorship W4387211617A5086706224 @default.
- W4387211617 hasConcept C111919701 @default.
- W4387211617 hasConcept C118505674 @default.
- W4387211617 hasConcept C119857082 @default.
- W4387211617 hasConcept C121332964 @default.
- W4387211617 hasConcept C153180895 @default.
- W4387211617 hasConcept C154945302 @default.
- W4387211617 hasConcept C165801399 @default.
- W4387211617 hasConcept C31972630 @default.
- W4387211617 hasConcept C41008148 @default.
- W4387211617 hasConcept C46686674 @default.
- W4387211617 hasConcept C62520636 @default.
- W4387211617 hasConcept C66322947 @default.
- W4387211617 hasConcept C89600930 @default.
- W4387211617 hasConceptScore W4387211617C111919701 @default.
- W4387211617 hasConceptScore W4387211617C118505674 @default.
- W4387211617 hasConceptScore W4387211617C119857082 @default.
- W4387211617 hasConceptScore W4387211617C121332964 @default.
- W4387211617 hasConceptScore W4387211617C153180895 @default.
- W4387211617 hasConceptScore W4387211617C154945302 @default.
- W4387211617 hasConceptScore W4387211617C165801399 @default.
- W4387211617 hasConceptScore W4387211617C31972630 @default.
- W4387211617 hasConceptScore W4387211617C41008148 @default.
- W4387211617 hasConceptScore W4387211617C46686674 @default.
- W4387211617 hasConceptScore W4387211617C62520636 @default.
- W4387211617 hasConceptScore W4387211617C66322947 @default.
- W4387211617 hasConceptScore W4387211617C89600930 @default.
- W4387211617 hasLocation W43872116171 @default.
- W4387211617 hasOpenAccess W4387211617 @default.
- W4387211617 hasPrimaryLocation W43872116171 @default.
- W4387211617 hasRelatedWork W1669643531 @default.
- W4387211617 hasRelatedWork W1982826852 @default.
- W4387211617 hasRelatedWork W2005437358 @default.
- W4387211617 hasRelatedWork W2008656436 @default.
- W4387211617 hasRelatedWork W2023558673 @default.
- W4387211617 hasRelatedWork W2110230079 @default.
- W4387211617 hasRelatedWork W2134924024 @default.
- W4387211617 hasRelatedWork W2517104666 @default.
- W4387211617 hasRelatedWork W2613186388 @default.
- W4387211617 hasRelatedWork W1967061043 @default.