Matches in SemOpenAlex for { <https://semopenalex.org/work/W3182587099> ?p ?o ?g. }
- W3182587099 abstract "Transfer learning is a commonly used strategy for medical image classification, especially via pretraining on source data and fine-tuning on target data. There is currently no consensus on how to choose appropriate source data, and in the literature we can find both evidence of favoring large natural image datasets such as ImageNet, and evidence of favoring more specialized medical datasets. In this paper we perform a systematic study with nine source datasets with natural or medical images, and three target medical datasets, all with 2D images. We find that ImageNet is the source leading to the highest performances, but also that larger datasets are not necessarily better. We also study different definitions of data similarity. We show that common intuitions about similarity may be inaccurate, and therefore not sufficient to predict an appropriate source a priori. Finally, we discuss several steps needed for further research in this field, especially with regard to other types (for example 3D) medical images. Our experiments and pretrained models are available via url{this https URL}" @default.
- W3182587099 created "2021-07-19" @default.
- W3182587099 creator A5010208747 @default.
- W3182587099 creator A5012353780 @default.
- W3182587099 creator A5016794035 @default.
- W3182587099 creator A5031371103 @default.
- W3182587099 creator A5068838617 @default.
- W3182587099 date "2021-07-13" @default.
- W3182587099 modified "2023-09-27" @default.
- W3182587099 title "Cats, not CAT scans: a study of dataset similarity in transfer learning for 2D medical image classification." @default.
- W3182587099 cites W16705017 @default.
- W3182587099 cites W1799366690 @default.
- W3182587099 cites W1920702274 @default.
- W3182587099 cites W2047643928 @default.
- W3182587099 cites W2101234009 @default.
- W3182587099 cites W2108598243 @default.
- W3182587099 cites W2118858186 @default.
- W3182587099 cites W2149933564 @default.
- W3182587099 cites W2165698076 @default.
- W3182587099 cites W2253429366 @default.
- W3182587099 cites W2346062110 @default.
- W3182587099 cites W2510153535 @default.
- W3182587099 cites W2560014990 @default.
- W3182587099 cites W2560476520 @default.
- W3182587099 cites W2577849240 @default.
- W3182587099 cites W2592929672 @default.
- W3182587099 cites W2594206839 @default.
- W3182587099 cites W2626059710 @default.
- W3182587099 cites W2630687088 @default.
- W3182587099 cites W2736422900 @default.
- W3182587099 cites W2788633781 @default.
- W3182587099 cites W2789571202 @default.
- W3182587099 cites W2794221466 @default.
- W3182587099 cites W2794335896 @default.
- W3182587099 cites W2804905867 @default.
- W3182587099 cites W2804935296 @default.
- W3182587099 cites W2806857275 @default.
- W3182587099 cites W2887680499 @default.
- W3182587099 cites W2895531857 @default.
- W3182587099 cites W2896966666 @default.
- W3182587099 cites W2916881227 @default.
- W3182587099 cites W2932083555 @default.
- W3182587099 cites W2946948417 @default.
- W3182587099 cites W2947706148 @default.
- W3182587099 cites W2949650786 @default.
- W3182587099 cites W2961533511 @default.
- W3182587099 cites W2962965870 @default.
- W3182587099 cites W2963234501 @default.
- W3182587099 cites W2963863924 @default.
- W3182587099 cites W2990761674 @default.
- W3182587099 cites W3005343217 @default.
- W3182587099 cites W3020996329 @default.
- W3182587099 cites W3025653827 @default.
- W3182587099 cites W3026809105 @default.
- W3182587099 cites W3035314311 @default.
- W3182587099 cites W3038443040 @default.
- W3182587099 cites W3095809712 @default.
- W3182587099 cites W3102785203 @default.
- W3182587099 cites W3134475970 @default.
- W3182587099 hasPublicationYear "2021" @default.
- W3182587099 type Work @default.
- W3182587099 sameAs 3182587099 @default.
- W3182587099 citedByCount "0" @default.
- W3182587099 crossrefType "posted-content" @default.
- W3182587099 hasAuthorship W3182587099A5010208747 @default.
- W3182587099 hasAuthorship W3182587099A5012353780 @default.
- W3182587099 hasAuthorship W3182587099A5016794035 @default.
- W3182587099 hasAuthorship W3182587099A5031371103 @default.
- W3182587099 hasAuthorship W3182587099A5068838617 @default.
- W3182587099 hasConcept C103278499 @default.
- W3182587099 hasConcept C111472728 @default.
- W3182587099 hasConcept C115961682 @default.
- W3182587099 hasConcept C119857082 @default.
- W3182587099 hasConcept C124101348 @default.
- W3182587099 hasConcept C138885662 @default.
- W3182587099 hasConcept C150899416 @default.
- W3182587099 hasConcept C153180895 @default.
- W3182587099 hasConcept C154945302 @default.
- W3182587099 hasConcept C202444582 @default.
- W3182587099 hasConcept C2983685735 @default.
- W3182587099 hasConcept C33923547 @default.
- W3182587099 hasConcept C41008148 @default.
- W3182587099 hasConcept C75553542 @default.
- W3182587099 hasConcept C9652623 @default.
- W3182587099 hasConceptScore W3182587099C103278499 @default.
- W3182587099 hasConceptScore W3182587099C111472728 @default.
- W3182587099 hasConceptScore W3182587099C115961682 @default.
- W3182587099 hasConceptScore W3182587099C119857082 @default.
- W3182587099 hasConceptScore W3182587099C124101348 @default.
- W3182587099 hasConceptScore W3182587099C138885662 @default.
- W3182587099 hasConceptScore W3182587099C150899416 @default.
- W3182587099 hasConceptScore W3182587099C153180895 @default.
- W3182587099 hasConceptScore W3182587099C154945302 @default.
- W3182587099 hasConceptScore W3182587099C202444582 @default.
- W3182587099 hasConceptScore W3182587099C2983685735 @default.
- W3182587099 hasConceptScore W3182587099C33923547 @default.
- W3182587099 hasConceptScore W3182587099C41008148 @default.
- W3182587099 hasConceptScore W3182587099C75553542 @default.
- W3182587099 hasConceptScore W3182587099C9652623 @default.
- W3182587099 hasLocation W31825870991 @default.