FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4377010966> ?p ?o ?g. }

• W4377010966 abstract "Abstract Background Deep learning models applied to healthcare applications including digital pathology have been increasing their scope and importance in recent years. Many of these models have been trained on The Cancer Genome Atlas (TCGA) atlas of digital images, or use it as a validation source. One crucial factor that seems to have been widely ignored is the internal bias that originates from the institutions that contributed WSIs to the TCGA dataset, and its effects on models trained on this dataset. Methods 8,579 paraffin-embedded, hematoxylin and eosin stained, digital slides were selected from the TCGA dataset. More than 140 medical institutions (acquisition sites) contributed to this dataset. Two deep neural networks (DenseNet121 and KimiaNet were used to extract deep features at 20× magnification. DenseNet was pre-trained on non-medical objects. KimiaNet has the same structure but trained for cancer type classification on TCGA images. The extracted deep features were later used to detect each slide’s acquisition site, and also for slide representation in image search. Results DenseNet’s deep features could distinguish acquisition sites with 70% accuracy whereas KimiaNet’s deep features could reveal acquisition sites with more than 86% accuracy. These findings suggest that there are acquisition site specific patterns that could be picked up by deep neural networks. It has also been shown that these medically irrelevant patterns can interfere with other applications of deep learning in digital pathology, namely image search. Summary This study shows that there are acquisition site specific patterns that can be used to identify tissue acquisition sites without any explicit training. Furthermore, it was observed that a model trained for cancer subtype classification has exploited such medically irrelevant patterns to classify cancer types. Digital scanner configuration and noise, tissue stain variation and artifacts, and source site patient demographics are among factors that likely account for the observed bias. Therefore, researchers should be cautious of such bias when using histopathology datasets for developing and training deep networks." @default.
• W4377010966 created "2023-05-19" @default.
• W4377010966 creator A5000774079 @default.
• W4377010966 creator A5017611735 @default.
• W4377010966 creator A5028342220 @default.
• W4377010966 creator A5038566332 @default.
• W4377010966 creator A5041398877 @default.
• W4377010966 creator A5044531867 @default.
• W4377010966 creator A5070926160 @default.
• W4377010966 creator A5077727968 @default.
• W4377010966 date "2023-05-17" @default.
• W4377010966 modified "2023-09-29" @default.
• W4377010966 title "Biased data, biased AI: deep networks predict the acquisition site of TCGA images" @default.
• W4377010966 cites W2015159529 @default.
• W4377010966 cites W2079770794 @default.
• W4377010966 cites W2108598243 @default.
• W4377010966 cites W2123229215 @default.
• W4377010966 cites W2154813452 @default.
• W4377010966 cites W2282821441 @default.
• W4377010966 cites W2329659234 @default.
• W4377010966 cites W2760946358 @default.
• W4377010966 cites W2789277655 @default.
• W4377010966 cites W2809112427 @default.
• W4377010966 cites W2810135932 @default.
• W4377010966 cites W2883147591 @default.
• W4377010966 cites W2955987583 @default.
• W4377010966 cites W2963446712 @default.
• W4377010966 cites W2964324957 @default.
• W4377010966 cites W2990072080 @default.
• W4377010966 cites W3004016611 @default.
• W4377010966 cites W3029272675 @default.
• W4377010966 cites W3033721673 @default.
• W4377010966 cites W3035647551 @default.
• W4377010966 cites W3036901136 @default.
• W4377010966 cites W3037731665 @default.
• W4377010966 cites W3095520330 @default.
• W4377010966 cites W3109918169 @default.
• W4377010966 cites W3133698071 @default.
• W4377010966 cites W3138973186 @default.
• W4377010966 cites W3164525109 @default.
• W4377010966 cites W3171849353 @default.
• W4377010966 cites W3186679119 @default.
• W4377010966 cites W3197765800 @default.
• W4377010966 cites W4295833879 @default.
• W4377010966 cites W4365447539 @default.
• W4377010966 doi "https://doi.org/10.1186/s13000-023-01355-3" @default.
• W4377010966 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/37198691" @default.
• W4377010966 hasPublicationYear "2023" @default.
• W4377010966 type Work @default.
• W4377010966 citedByCount "5" @default.
• W4377010966 countsByYear W43770109662023 @default.
• W4377010966 crossrefType "journal-article" @default.
• W4377010966 hasAuthorship W4377010966A5000774079 @default.
• W4377010966 hasAuthorship W4377010966A5017611735 @default.
• W4377010966 hasAuthorship W4377010966A5028342220 @default.
• W4377010966 hasAuthorship W4377010966A5038566332 @default.
• W4377010966 hasAuthorship W4377010966A5041398877 @default.
• W4377010966 hasAuthorship W4377010966A5044531867 @default.
• W4377010966 hasAuthorship W4377010966A5070926160 @default.
• W4377010966 hasAuthorship W4377010966A5077727968 @default.
• W4377010966 hasBestOaLocation W43770109661 @default.
• W4377010966 hasConcept C108583219 @default.
• W4377010966 hasConcept C153180895 @default.
• W4377010966 hasConcept C154945302 @default.
• W4377010966 hasConcept C2777522853 @default.
• W4377010966 hasConcept C2984842247 @default.
• W4377010966 hasConcept C31972630 @default.
• W4377010966 hasConcept C41008148 @default.
• W4377010966 hasConcept C4144372 @default.
• W4377010966 hasConcept C50644808 @default.
• W4377010966 hasConceptScore W4377010966C108583219 @default.
• W4377010966 hasConceptScore W4377010966C153180895 @default.
• W4377010966 hasConceptScore W4377010966C154945302 @default.
• W4377010966 hasConceptScore W4377010966C2777522853 @default.
• W4377010966 hasConceptScore W4377010966C2984842247 @default.
• W4377010966 hasConceptScore W4377010966C31972630 @default.
• W4377010966 hasConceptScore W4377010966C41008148 @default.
• W4377010966 hasConceptScore W4377010966C4144372 @default.
• W4377010966 hasConceptScore W4377010966C50644808 @default.
• W4377010966 hasFunder F4320313340 @default.
• W4377010966 hasIssue "1" @default.
• W4377010966 hasLocation W43770109661 @default.
• W4377010966 hasLocation W43770109662 @default.
• W4377010966 hasLocation W43770109663 @default.
• W4377010966 hasLocation W43770109664 @default.
• W4377010966 hasOpenAccess W4377010966 @default.
• W4377010966 hasPrimaryLocation W43770109661 @default.
• W4377010966 hasRelatedWork W2139751930 @default.
• W4377010966 hasRelatedWork W2279398222 @default.
• W4377010966 hasRelatedWork W2738221750 @default.
• W4377010966 hasRelatedWork W2791691546 @default.
• W4377010966 hasRelatedWork W2915754718 @default.
• W4377010966 hasRelatedWork W2941846814 @default.
• W4377010966 hasRelatedWork W3082895349 @default.
• W4377010966 hasRelatedWork W3176779650 @default.
• W4377010966 hasRelatedWork W3193983877 @default.
• W4377010966 hasRelatedWork W4299822940 @default.
• W4377010966 hasVolume "18" @default.
• W4377010966 isParatext "false" @default.
• W4377010966 isRetracted "false" @default.

• next