Matches in SemOpenAlex for { <https://semopenalex.org/work/W4384162023> ?p ?o ?g. }
Showing items 1 to 77 of
77
with 100 items per page.
- W4384162023 endingPage "1" @default.
- W4384162023 startingPage "1" @default.
- W4384162023 abstract "Medical contrastive vision-language pretraining has shown great promise in many downstream tasks, such as data-efficient/zero-shot recognition. Current studies pretrain the network with contrastive loss by treating the paired image-reports as positive samples and the unpaired ones as negative samples. However, unlike natural datasets, many medical images or reports from different cases could have large similarity especially for the normal cases, and treating all the unpaired ones as negative samples could undermine the learned semantic structure and impose an adverse effect on the representations. Therefore, we design a simple yet effective approach for better contrastive learning in medical vision-language field. Specifically, by simplifying the computation of similarity between medical image-report pairs into the calculation of the inter-report similarity, the image-report tuples are divided into positive, negative, and additional neutral groups. With this better categorization of samples, more suitable contrastive loss is constructed. For evaluation, we perform extensive experiments by applying the proposed model-agnostic strategy to two state-of-the-art pretraining frameworks. The consistent improvements on four common downstream tasks, including cross-modal retrieval, zero-shot/data-efficient image classification, and image segmentation, demonstrate the effectiveness of the proposed strategy in medical field." @default.
- W4384162023 created "2023-07-14" @default.
- W4384162023 creator A5008168090 @default.
- W4384162023 creator A5014222179 @default.
- W4384162023 creator A5021769901 @default.
- W4384162023 creator A5027053844 @default.
- W4384162023 creator A5051649145 @default.
- W4384162023 creator A5065417380 @default.
- W4384162023 creator A5090815103 @default.
- W4384162023 date "2023-01-01" @default.
- W4384162023 modified "2023-09-27" @default.
- W4384162023 title "Improving Medical Vision-Language Contrastive Pretraining with Semantics-aware Triage" @default.
- W4384162023 doi "https://doi.org/10.1109/tmi.2023.3294980" @default.
- W4384162023 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/37440389" @default.
- W4384162023 hasPublicationYear "2023" @default.
- W4384162023 type Work @default.
- W4384162023 citedByCount "0" @default.
- W4384162023 crossrefType "journal-article" @default.
- W4384162023 hasAuthorship W4384162023A5008168090 @default.
- W4384162023 hasAuthorship W4384162023A5014222179 @default.
- W4384162023 hasAuthorship W4384162023A5021769901 @default.
- W4384162023 hasAuthorship W4384162023A5027053844 @default.
- W4384162023 hasAuthorship W4384162023A5051649145 @default.
- W4384162023 hasAuthorship W4384162023A5065417380 @default.
- W4384162023 hasAuthorship W4384162023A5090815103 @default.
- W4384162023 hasConcept C103278499 @default.
- W4384162023 hasConcept C115961682 @default.
- W4384162023 hasConcept C153180895 @default.
- W4384162023 hasConcept C154945302 @default.
- W4384162023 hasConcept C1667742 @default.
- W4384162023 hasConcept C184337299 @default.
- W4384162023 hasConcept C195324797 @default.
- W4384162023 hasConcept C199360897 @default.
- W4384162023 hasConcept C202444582 @default.
- W4384162023 hasConcept C204321447 @default.
- W4384162023 hasConcept C33923547 @default.
- W4384162023 hasConcept C41008148 @default.
- W4384162023 hasConcept C69505689 @default.
- W4384162023 hasConcept C75294576 @default.
- W4384162023 hasConcept C94124525 @default.
- W4384162023 hasConcept C9652623 @default.
- W4384162023 hasConceptScore W4384162023C103278499 @default.
- W4384162023 hasConceptScore W4384162023C115961682 @default.
- W4384162023 hasConceptScore W4384162023C153180895 @default.
- W4384162023 hasConceptScore W4384162023C154945302 @default.
- W4384162023 hasConceptScore W4384162023C1667742 @default.
- W4384162023 hasConceptScore W4384162023C184337299 @default.
- W4384162023 hasConceptScore W4384162023C195324797 @default.
- W4384162023 hasConceptScore W4384162023C199360897 @default.
- W4384162023 hasConceptScore W4384162023C202444582 @default.
- W4384162023 hasConceptScore W4384162023C204321447 @default.
- W4384162023 hasConceptScore W4384162023C33923547 @default.
- W4384162023 hasConceptScore W4384162023C41008148 @default.
- W4384162023 hasConceptScore W4384162023C69505689 @default.
- W4384162023 hasConceptScore W4384162023C75294576 @default.
- W4384162023 hasConceptScore W4384162023C94124525 @default.
- W4384162023 hasConceptScore W4384162023C9652623 @default.
- W4384162023 hasFunder F4320335777 @default.
- W4384162023 hasLocation W43841620231 @default.
- W4384162023 hasLocation W43841620232 @default.
- W4384162023 hasOpenAccess W4384162023 @default.
- W4384162023 hasPrimaryLocation W43841620231 @default.
- W4384162023 hasRelatedWork W1541271503 @default.
- W4384162023 hasRelatedWork W2186284405 @default.
- W4384162023 hasRelatedWork W2189043660 @default.
- W4384162023 hasRelatedWork W2365213443 @default.
- W4384162023 hasRelatedWork W2509146328 @default.
- W4384162023 hasRelatedWork W2732542196 @default.
- W4384162023 hasRelatedWork W2887442125 @default.
- W4384162023 hasRelatedWork W2941732455 @default.
- W4384162023 hasRelatedWork W3190424778 @default.
- W4384162023 hasRelatedWork W1872130062 @default.
- W4384162023 isParatext "false" @default.
- W4384162023 isRetracted "false" @default.
- W4384162023 workType "article" @default.