FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2899974865> ?p ?o ?g. }

• W2899974865 endingPage "209" @default.
• W2899974865 startingPage "201" @default.
• W2899974865 abstract "Purpose: Radiomic studies, where correlations are drawn between patients' medical image features and patient outcomes, often deal with small datasets. Consequently, results can suffer from lack of replicability and stability. This paper establishes a methodology to assess and reduce the impact of statistical fluctuations that may occur in small datasets. Such fluctuations can lead to false discoveries, particularly when applying feature selection or machine learning (ML) methods commonly used in the radiomics literature. Methods: Two feature selection methods were created, one for choosing single predictive features, and another for obtaining features sets that could be combined in a predictive model. The features were combined using ML tools less affected by overfitting (Naïve Bayes, logistic regression, and linear support vector machines). Only three features were allowed to be combined at a time, further limiting overfitting. This methodology was applied to MR images from small datasets in metastatic liver disease (69 samples) and primary uterine adenocarcinoma (93 samples), and the outcomes studied were: desmoplasia (for liver metastases), lymphovascular space invasion (LVSI), cancer staging (FIGO), and tumor grade (for uterine tumors). For outcomes in uterine cancer, the predictive models were tested on independent subsets. Results: With respect to the combined predictive feature approach: for LVSI, a prognostic factor that a human reader cannot detect, the predictive model yielded AUC = 0.87 ± 0.07 and accuracy = 0.84 ± 0.09 in the testing set. For FIGO staging, AUC = 0.81 ± 0.03 and accuracy = 0.79 ± 0.08. For tumor grade, AUC = 0.76 ± 0.05 and accuracy = 0.70 ± 0.08. Conclusion: Despite considering a large set (~104) of texture features, the false discovery avoidance methodology allowed only robust predictive models to be retained. Thus, the stringent false discovery avoidance methods introduced here do not preclude the discovery of promising correlations." @default.
• W2899974865 created "2018-11-16" @default.
• W2899974865 creator A5009033925 @default.
• W2899974865 creator A5013292433 @default.
• W2899974865 creator A5033126601 @default.
• W2899974865 creator A5045019334 @default.
• W2899974865 creator A5050179360 @default.
• W2899974865 creator A5051344867 @default.
• W2899974865 creator A5058751861 @default.
• W2899974865 creator A5071840946 @default.
• W2899974865 creator A5090517775 @default.
• W2899974865 date "2019-03-01" @default.
• W2899974865 modified "2023-10-11" @default.
• W2899974865 title "An Empirical Approach for Avoiding False Discoveries When Applying High-Dimensional Radiomics to Small Datasets" @default.
• W2899974865 cites W1408981388 @default.
• W2899974865 cites W1981494355 @default.
• W2899974865 cites W2003304826 @default.
• W2899974865 cites W2044465660 @default.
• W2899974865 cites W2055575746 @default.
• W2899974865 cites W2097475056 @default.
• W2899974865 cites W2099698084 @default.
• W2899974865 cites W2110065044 @default.
• W2899974865 cites W2117297165 @default.
• W2899974865 cites W2122763860 @default.
• W2899974865 cites W2128739912 @default.
• W2899974865 cites W2135046866 @default.
• W2899974865 cites W2162511520 @default.
• W2899974865 cites W2165927297 @default.
• W2899974865 cites W2174661749 @default.
• W2899974865 cites W2213612645 @default.
• W2899974865 cites W2327203407 @default.
• W2899974865 cites W2333277922 @default.
• W2899974865 cites W2346343836 @default.
• W2899974865 cites W2461805626 @default.
• W2899974865 cites W2470491115 @default.
• W2899974865 cites W2517065464 @default.
• W2899974865 cites W2566749675 @default.
• W2899974865 cites W2589577297 @default.
• W2899974865 doi "https://doi.org/10.1109/trpms.2018.2880617" @default.
• W2899974865 hasPublicationYear "2019" @default.
• W2899974865 type Work @default.
• W2899974865 sameAs 2899974865 @default.
• W2899974865 citedByCount "15" @default.
• W2899974865 countsByYear W28999748652019 @default.
• W2899974865 countsByYear W28999748652020 @default.
• W2899974865 countsByYear W28999748652021 @default.
• W2899974865 countsByYear W28999748652022 @default.
• W2899974865 countsByYear W28999748652023 @default.
• W2899974865 crossrefType "journal-article" @default.
• W2899974865 hasAuthorship W2899974865A5009033925 @default.
• W2899974865 hasAuthorship W2899974865A5013292433 @default.
• W2899974865 hasAuthorship W2899974865A5033126601 @default.
• W2899974865 hasAuthorship W2899974865A5045019334 @default.
• W2899974865 hasAuthorship W2899974865A5050179360 @default.
• W2899974865 hasAuthorship W2899974865A5051344867 @default.
• W2899974865 hasAuthorship W2899974865A5058751861 @default.
• W2899974865 hasAuthorship W2899974865A5071840946 @default.
• W2899974865 hasAuthorship W2899974865A5090517775 @default.
• W2899974865 hasConcept C119857082 @default.
• W2899974865 hasConcept C12267149 @default.
• W2899974865 hasConcept C138885662 @default.
• W2899974865 hasConcept C148483581 @default.
• W2899974865 hasConcept C151956035 @default.
• W2899974865 hasConcept C153180895 @default.
• W2899974865 hasConcept C154945302 @default.
• W2899974865 hasConcept C22019652 @default.
• W2899974865 hasConcept C2776401178 @default.
• W2899974865 hasConcept C2778559731 @default.
• W2899974865 hasConcept C41008148 @default.
• W2899974865 hasConcept C41895202 @default.
• W2899974865 hasConcept C50644808 @default.
• W2899974865 hasConcept C58471807 @default.
• W2899974865 hasConceptScore W2899974865C119857082 @default.
• W2899974865 hasConceptScore W2899974865C12267149 @default.
• W2899974865 hasConceptScore W2899974865C138885662 @default.
• W2899974865 hasConceptScore W2899974865C148483581 @default.
• W2899974865 hasConceptScore W2899974865C151956035 @default.
• W2899974865 hasConceptScore W2899974865C153180895 @default.
• W2899974865 hasConceptScore W2899974865C154945302 @default.
• W2899974865 hasConceptScore W2899974865C22019652 @default.
• W2899974865 hasConceptScore W2899974865C2776401178 @default.
• W2899974865 hasConceptScore W2899974865C2778559731 @default.
• W2899974865 hasConceptScore W2899974865C41008148 @default.
• W2899974865 hasConceptScore W2899974865C41895202 @default.
• W2899974865 hasConceptScore W2899974865C50644808 @default.
• W2899974865 hasConceptScore W2899974865C58471807 @default.
• W2899974865 hasFunder F4320322597 @default.
• W2899974865 hasFunder F4320334506 @default.
• W2899974865 hasFunder F4320334593 @default.
• W2899974865 hasIssue "2" @default.
• W2899974865 hasLocation W28999748651 @default.
• W2899974865 hasOpenAccess W2899974865 @default.
• W2899974865 hasPrimaryLocation W28999748651 @default.
• W2899974865 hasRelatedWork W1996541855 @default.
• W2899974865 hasRelatedWork W2384093694 @default.
• W2899974865 hasRelatedWork W2779605423 @default.
• W2899974865 hasRelatedWork W2941912306 @default.
• W2899974865 hasRelatedWork W3105251098 @default.

• next