FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3048137319> ?p ?o ?g. }

• W3048137319 endingPage "103527" @default.
• W3048137319 startingPage "103527" @default.
• W3048137319 abstract "To present a Machine Learning pipeline for automatically relabeling anatomical structure sets in the Digital Imaging and Communications in Medicine (DICOM) format to a standard nomenclature that will enable data abstraction for research and quality improvement. DICOM structure sets from approximately 1200 lung and prostate cancer patients across 40 treatment centers were used to build predictive models to automate the relabeling of clinically specified structure labels to standardized labels as defined by the American Association of Physics in Medicine’s (AAPM) Task Group 263 (TG-263). Volumetric bitmaps were created based on the delineated volumes and were combined with associated bony anatomy data to build feature vectors. Feature reduction was performed with singular value decomposition and the resulting vectors were used for predicting the label of each structure using five different classifier algorithms on the Apache Spark platform with 5-fold cross-validation. Undersampling methods were used to deal with underlying class imbalance that hindered the performance of classifiers. Experiments were performed on both a curated version of the data, which included only annotated structures, and the non-curated data that included all structures from the original treatment plans. Random Forest provided the highest accuracies with F 1 scores of 98.77 for lung and 95.06 for prostate on the curated data sets. Scores were lower with 95.67 for lung and 90.22 for prostate on the non-curated data sets, highlighting some of the challenges of classifying real clinical data. Including bony anatomy data and pooling information from all structures for the same patient both increased accuracies. In some cases, undersampling with k -Means clustering for class balancing improved classifier accuracy but in all experiments it significantly reduced run time compared to random undersampling. This work shows that structure sets can be relabeled using our approach with accuracies over 95% for many structure types when presented with curated data. Although accuracies dropped when using the full non-curated data sets, some structure types were still correctly labeled over 90% of the time. With similar results obtained on an external test data set, we can infer that the proposed models are likely to work on other clinical data sets. • The best F1 scores were 98.77 for lung and 95.06 for prostate. • Bony anatomy and using patient dependence improved classifier accuracy. • k-Means undersampling can improve classifier predictive performance and run time. • A high percentage of structure labels were not common between training and test data. • Classifying real world clinical data sets remains a challenging problem." @default.
• W3048137319 created "2020-08-13" @default.
• W3048137319 creator A5021439006 @default.
• W3048137319 creator A5046617944 @default.
• W3048137319 creator A5047762320 @default.
• W3048137319 creator A5050819040 @default.
• W3048137319 creator A5054879396 @default.
• W3048137319 creator A5064188373 @default.
• W3048137319 creator A5073284026 @default.
• W3048137319 creator A5080886331 @default.
• W3048137319 date "2020-09-01" @default.
• W3048137319 modified "2023-09-29" @default.
• W3048137319 title "A Machine Learning method for relabeling arbitrary DICOM structure sets to TG-263 defined labels" @default.
• W3048137319 cites W1978425155 @default.
• W3048137319 cites W2001881071 @default.
• W3048137319 cites W2012708547 @default.
• W3048137319 cites W2073171674 @default.
• W3048137319 cites W2094409975 @default.
• W3048137319 cites W2117756735 @default.
• W3048137319 cites W2172000360 @default.
• W3048137319 cites W2174357492 @default.
• W3048137319 cites W2176028050 @default.
• W3048137319 cites W2338318698 @default.
• W3048137319 cites W2413354483 @default.
• W3048137319 cites W2551560790 @default.
• W3048137319 cites W2774576524 @default.
• W3048137319 cites W2808391149 @default.
• W3048137319 cites W2896804544 @default.
• W3048137319 cites W2903870823 @default.
• W3048137319 cites W2911964244 @default.
• W3048137319 cites W2948548380 @default.
• W3048137319 cites W3002921634 @default.
• W3048137319 cites W3003839960 @default.
• W3048137319 cites W4239510810 @default.
• W3048137319 doi "https://doi.org/10.1016/j.jbi.2020.103527" @default.
• W3048137319 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/32777484" @default.
• W3048137319 hasPublicationYear "2020" @default.
• W3048137319 type Work @default.
• W3048137319 sameAs 3048137319 @default.
• W3048137319 citedByCount "10" @default.
• W3048137319 countsByYear W30481373192021 @default.
• W3048137319 countsByYear W30481373192022 @default.
• W3048137319 countsByYear W30481373192023 @default.
• W3048137319 crossrefType "journal-article" @default.
• W3048137319 hasAuthorship W3048137319A5021439006 @default.
• W3048137319 hasAuthorship W3048137319A5046617944 @default.
• W3048137319 hasAuthorship W3048137319A5047762320 @default.
• W3048137319 hasAuthorship W3048137319A5050819040 @default.
• W3048137319 hasAuthorship W3048137319A5054879396 @default.
• W3048137319 hasAuthorship W3048137319A5064188373 @default.
• W3048137319 hasAuthorship W3048137319A5073284026 @default.
• W3048137319 hasAuthorship W3048137319A5080886331 @default.
• W3048137319 hasBestOaLocation W30481373191 @default.
• W3048137319 hasConcept C118552586 @default.
• W3048137319 hasConcept C119857082 @default.
• W3048137319 hasConcept C124101348 @default.
• W3048137319 hasConcept C153180895 @default.
• W3048137319 hasConcept C154945302 @default.
• W3048137319 hasConcept C155261790 @default.
• W3048137319 hasConcept C2780035454 @default.
• W3048137319 hasConcept C3115412 @default.
• W3048137319 hasConcept C41008148 @default.
• W3048137319 hasConcept C71924100 @default.
• W3048137319 hasConcept C77331912 @default.
• W3048137319 hasConceptScore W3048137319C118552586 @default.
• W3048137319 hasConceptScore W3048137319C119857082 @default.
• W3048137319 hasConceptScore W3048137319C124101348 @default.
• W3048137319 hasConceptScore W3048137319C153180895 @default.
• W3048137319 hasConceptScore W3048137319C154945302 @default.
• W3048137319 hasConceptScore W3048137319C155261790 @default.
• W3048137319 hasConceptScore W3048137319C2780035454 @default.
• W3048137319 hasConceptScore W3048137319C3115412 @default.
• W3048137319 hasConceptScore W3048137319C41008148 @default.
• W3048137319 hasConceptScore W3048137319C71924100 @default.
• W3048137319 hasConceptScore W3048137319C77331912 @default.
• W3048137319 hasFunder F4320311096 @default.
• W3048137319 hasLocation W30481373191 @default.
• W3048137319 hasOpenAccess W3048137319 @default.
• W3048137319 hasPrimaryLocation W30481373191 @default.
• W3048137319 hasRelatedWork W1911936269 @default.
• W3048137319 hasRelatedWork W2075251596 @default.
• W3048137319 hasRelatedWork W2111883592 @default.
• W3048137319 hasRelatedWork W2891267967 @default.
• W3048137319 hasRelatedWork W2961085424 @default.
• W3048137319 hasRelatedWork W3048137319 @default.
• W3048137319 hasRelatedWork W4286629047 @default.
• W3048137319 hasRelatedWork W4306321456 @default.
• W3048137319 hasRelatedWork W4306674287 @default.
• W3048137319 hasRelatedWork W4224009465 @default.
• W3048137319 hasVolume "109" @default.
• W3048137319 isParatext "false" @default.
• W3048137319 isRetracted "false" @default.
• W3048137319 magId "3048137319" @default.
• W3048137319 workType "article" @default.