FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4315630643> ?p ?o ?g. }

• W4315630643 endingPage "4268" @default.
• W4315630643 startingPage "4255" @default.
• W4315630643 abstract "Machine learning algorithms are best trained with large quantities of accurately annotated samples. While natural scene images can often be labeled relatively cheaply and at large scale, obtaining accurate annotations for medical images is both time consuming and expensive. In this study, we propose a cooperative labeling method that allows us to make use of weakly annotated medical imaging data for the training of a machine learning algorithm. As most clinically produced data are weakly-annotated - produced for use by humans rather than machines and lacking information machine learning depends upon - this approach allows us to incorporate a wider range of clinical data and thereby increase the training set size.Our pseudo-labeling method consists of multiple stages. In the first stage, a previously established network is trained using a limited number of samples with high-quality expert-produced annotations. This network is used to generate annotations for a separate larger dataset that contains only weakly annotated scans. In the second stage, by cross-checking the two types of annotations against each other, we obtain higher-fidelity annotations. In the third stage, we extract training data from the weakly annotated scans, and combine it with the fully annotated data, producing a larger training dataset. We use this larger dataset to develop a computer-aided detection (CADe) system for nodule detection in chest CT.We evaluated the proposed approach by presenting the network with different numbers of expert-annotated scans in training and then testing the CADe using an independent expert-annotated dataset. We demonstrate that when availability of expert annotations is severely limited, the inclusion of weakly-labeled data leads to a 5% improvement in the competitive performance metric (CPM), defined as the average of sensitivities at different false-positive rates.Our proposed approach can effectively merge a weakly-annotated dataset with a small, well-annotated dataset for algorithm training. This approach can help enlarge limited training data by leveraging the large amount of weakly labeled data typically generated in clinical image interpretation." @default.
• W4315630643 created "2023-01-12" @default.
• W4315630643 creator A5006593300 @default.
• W4315630643 creator A5036912867 @default.
• W4315630643 creator A5050183993 @default.
• W4315630643 creator A5072858561 @default.
• W4315630643 creator A5073468417 @default.
• W4315630643 creator A5074995495 @default.
• W4315630643 creator A5083553427 @default.
• W4315630643 creator A5089767317 @default.
• W4315630643 date "2023-01-27" @default.
• W4315630643 modified "2023-10-16" @default.
• W4315630643 title "Semi‐supervised training using cooperative labeling of weakly annotated data for nodule detection in chest CT" @default.
• W4315630643 cites W130099911 @default.
• W4315630643 cites W1901129140 @default.
• W4315630643 cites W1972514162 @default.
• W4315630643 cites W1986649315 @default.
• W4315630643 cites W2071717081 @default.
• W4315630643 cites W2079480962 @default.
• W4315630643 cites W2322371438 @default.
• W4315630643 cites W2524399695 @default.
• W4315630643 cites W2584017349 @default.
• W4315630643 cites W2751665805 @default.
• W4315630643 cites W2791142503 @default.
• W4315630643 cites W2793409683 @default.
• W4315630643 cites W2889646458 @default.
• W4315630643 cites W2899912692 @default.
• W4315630643 cites W2946185430 @default.
• W4315630643 cites W2963351448 @default.
• W4315630643 cites W2964317695 @default.
• W4315630643 cites W2979433110 @default.
• W4315630643 cites W3035406940 @default.
• W4315630643 cites W3054666633 @default.
• W4315630643 cites W3102190365 @default.
• W4315630643 cites W3121913585 @default.
• W4315630643 cites W3159288224 @default.
• W4315630643 doi "https://doi.org/10.1002/mp.16219" @default.
• W4315630643 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/36630691" @default.
• W4315630643 hasPublicationYear "2023" @default.
• W4315630643 type Work @default.
• W4315630643 citedByCount "1" @default.
• W4315630643 countsByYear W43156306432023 @default.
• W4315630643 crossrefType "journal-article" @default.
• W4315630643 hasAuthorship W4315630643A5006593300 @default.
• W4315630643 hasAuthorship W4315630643A5036912867 @default.
• W4315630643 hasAuthorship W4315630643A5050183993 @default.
• W4315630643 hasAuthorship W4315630643A5072858561 @default.
• W4315630643 hasAuthorship W4315630643A5073468417 @default.
• W4315630643 hasAuthorship W4315630643A5074995495 @default.
• W4315630643 hasAuthorship W4315630643A5083553427 @default.
• W4315630643 hasAuthorship W4315630643A5089767317 @default.
• W4315630643 hasConcept C108583219 @default.
• W4315630643 hasConcept C119857082 @default.
• W4315630643 hasConcept C153180895 @default.
• W4315630643 hasConcept C154945302 @default.
• W4315630643 hasConcept C2776459999 @default.
• W4315630643 hasConcept C41008148 @default.
• W4315630643 hasConcept C51632099 @default.
• W4315630643 hasConcept C58489278 @default.
• W4315630643 hasConcept C76155785 @default.
• W4315630643 hasConceptScore W4315630643C108583219 @default.
• W4315630643 hasConceptScore W4315630643C119857082 @default.
• W4315630643 hasConceptScore W4315630643C153180895 @default.
• W4315630643 hasConceptScore W4315630643C154945302 @default.
• W4315630643 hasConceptScore W4315630643C2776459999 @default.
• W4315630643 hasConceptScore W4315630643C41008148 @default.
• W4315630643 hasConceptScore W4315630643C51632099 @default.
• W4315630643 hasConceptScore W4315630643C58489278 @default.
• W4315630643 hasConceptScore W4315630643C76155785 @default.
• W4315630643 hasIssue "7" @default.
• W4315630643 hasLocation W43156306431 @default.
• W4315630643 hasLocation W43156306432 @default.
• W4315630643 hasOpenAccess W4315630643 @default.
• W4315630643 hasPrimaryLocation W43156306431 @default.
• W4315630643 hasRelatedWork W3014300295 @default.
• W4315630643 hasRelatedWork W3164822677 @default.
• W4315630643 hasRelatedWork W4223943233 @default.
• W4315630643 hasRelatedWork W4225161397 @default.
• W4315630643 hasRelatedWork W4250304930 @default.
• W4315630643 hasRelatedWork W4312200629 @default.
• W4315630643 hasRelatedWork W4360585206 @default.
• W4315630643 hasRelatedWork W4364306694 @default.
• W4315630643 hasRelatedWork W4380075502 @default.
• W4315630643 hasRelatedWork W4380086463 @default.
• W4315630643 hasVolume "50" @default.
• W4315630643 isParatext "false" @default.
• W4315630643 isRetracted "false" @default.
• W4315630643 workType "article" @default.