FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3035993984> ?p ?o ?g. }

• W3035993984 endingPage "e206" @default.
• W3035993984 startingPage "e205" @default.
• W3035993984 abstract "Prior large-scale studies have shown that approximately 20% of patients develop progression of carotid stenosis over 28 months. We recently reported that under 4% of patients experience rapid progression of carotid stenosis within 18 months. We sought to determine whether a novel machine learning algorithm could identify clinical features that predict rapid progression of carotid stenosis. Patients with two or more carotid duplex ultrasounds from August 2010 to August 2018 were identified. Our algorithm included patients by analyzing the text of carotid duplex ultrasound reports. We stratified patients into the following levels of carotid stenosis: (1) 0% to 39%; (2) 40% to 59%; (3) 60% to 79%; (4) 80% to 99%; and (5) occlusion. Carotid stenosis rapid progression was defined as an increase in two or more levels in less than 18 months. Value of health covariates for prediction of carotid stenosis rapid progression were evaluated using least absolute shrinkage and selection operator, a machine learning regression method that analyzes all possible combinations of terms, creating a set of terms which best predict rapid progression. The Results were cross-validated by testing predictive terms against randomly chosen mixtures of rapid progressors and nonprogressing controls. Predictions based on the fitted model were evaluated against true (validated) rapid progression status to calculate the C-Statistic. Area under the receiver operating characteristic curve was plotted to visualize model fit. From an initial cohort of over 4 million unique patients, 4452 patients with two or more duplex ultrasound examinations and complete health covariates were identified. A group of 180 patients experienced rapid progression of carotid stenosis and were compared to the remainder of our cohort. The machine learning algorithm predicted rapid progression of carotid stenosis with a C statistic of 0.71 (Figure). The algorithm found a history of peripheral arterial disease (coefficient = 1.01), transient ischemic attack (0.55), and use of antihyperlipidemics (0.52) predicted rapid progression of carotid stenosis. Former tobacco use (–0.98) predicted against rapid progression of carotid stenosis (Table). Machine learning identifies a mathematically useful and unbiased set of factors for prediction of carotid stenosis progression, independent of clinical significance. Our method analyzes all possible combinations of clinical features and generates a set of features which predict rapid progression of carotid stenosis; the process is then cross-validated by testing the predictive features against a random mixture of rapid progressors and nonprogressing controls. This methodology could be applied to other diseases to understand factors that may predict progression or development of disease.TableClinical features associated with rapid progression of carotid stenosisPredictive of development Native American race (1.34) Hispanic race (0.31) History of peripheral arterial disease (1.01) History of transient ischemic attack (0.55) Use of antihyperlipidemics (0.52) Use of antiplatelet drugs (0.30) Use of anticoagulant drugs (0.39) White race (0.22) Use of antihyperglycemics (0.03)Considered by algorithm, but not predictive Male sex Asian race East Indian race Filipino race Korean Middle Eastern Hawaiian South American Vietnamese Unknown race Diabetes Hyperlipidemia Hypertension Male sexPredictive against development Former tobacco use (–0.98) Increasing age at time of examination (–0.002) Increasing body mass index (–0.02) Ischemic heart disease (–0.18) Open table in a new tab" @default.
• W3035993984 created "2020-06-25" @default.
• W3035993984 creator A5017735925 @default.
• W3035993984 creator A5024980646 @default.
• W3035993984 creator A5032232094 @default.
• W3035993984 creator A5034337969 @default.
• W3035993984 creator A5056025545 @default.
• W3035993984 creator A5063933737 @default.
• W3035993984 creator A5064252055 @default.
• W3035993984 creator A5082341569 @default.
• W3035993984 date "2020-07-01" @default.
• W3035993984 modified "2023-09-26" @default.
• W3035993984 title "Machine Learning Predicts Carotid Stenosis Rapid Progression in an 8-Year Dataset With 4452 Patients" @default.
• W3035993984 doi "https://doi.org/10.1016/j.jvs.2020.04.350" @default.
• W3035993984 hasPublicationYear "2020" @default.
• W3035993984 type Work @default.
• W3035993984 sameAs 3035993984 @default.
• W3035993984 citedByCount "0" @default.
• W3035993984 crossrefType "journal-article" @default.
• W3035993984 hasAuthorship W3035993984A5017735925 @default.
• W3035993984 hasAuthorship W3035993984A5024980646 @default.
• W3035993984 hasAuthorship W3035993984A5032232094 @default.
• W3035993984 hasAuthorship W3035993984A5034337969 @default.
• W3035993984 hasAuthorship W3035993984A5056025545 @default.
• W3035993984 hasAuthorship W3035993984A5063933737 @default.
• W3035993984 hasAuthorship W3035993984A5064252055 @default.
• W3035993984 hasAuthorship W3035993984A5082341569 @default.
• W3035993984 hasBestOaLocation W30359939841 @default.
• W3035993984 hasConcept C119043178 @default.
• W3035993984 hasConcept C119857082 @default.
• W3035993984 hasConcept C126322002 @default.
• W3035993984 hasConcept C126838900 @default.
• W3035993984 hasConcept C143753070 @default.
• W3035993984 hasConcept C151956035 @default.
• W3035993984 hasConcept C2780007028 @default.
• W3035993984 hasConcept C41008148 @default.
• W3035993984 hasConcept C54355233 @default.
• W3035993984 hasConcept C552990157 @default.
• W3035993984 hasConcept C58471807 @default.
• W3035993984 hasConcept C71924100 @default.
• W3035993984 hasConcept C76318530 @default.
• W3035993984 hasConcept C86803240 @default.
• W3035993984 hasConcept C99611785 @default.
• W3035993984 hasConceptScore W3035993984C119043178 @default.
• W3035993984 hasConceptScore W3035993984C119857082 @default.
• W3035993984 hasConceptScore W3035993984C126322002 @default.
• W3035993984 hasConceptScore W3035993984C126838900 @default.
• W3035993984 hasConceptScore W3035993984C143753070 @default.
• W3035993984 hasConceptScore W3035993984C151956035 @default.
• W3035993984 hasConceptScore W3035993984C2780007028 @default.
• W3035993984 hasConceptScore W3035993984C41008148 @default.
• W3035993984 hasConceptScore W3035993984C54355233 @default.
• W3035993984 hasConceptScore W3035993984C552990157 @default.
• W3035993984 hasConceptScore W3035993984C58471807 @default.
• W3035993984 hasConceptScore W3035993984C71924100 @default.
• W3035993984 hasConceptScore W3035993984C76318530 @default.
• W3035993984 hasConceptScore W3035993984C86803240 @default.
• W3035993984 hasConceptScore W3035993984C99611785 @default.
• W3035993984 hasIssue "1" @default.
• W3035993984 hasLocation W30359939841 @default.
• W3035993984 hasOpenAccess W3035993984 @default.
• W3035993984 hasPrimaryLocation W30359939841 @default.
• W3035993984 hasRelatedWork W1975167538 @default.
• W3035993984 hasRelatedWork W1992584557 @default.
• W3035993984 hasRelatedWork W2016949197 @default.
• W3035993984 hasRelatedWork W2090977008 @default.
• W3035993984 hasRelatedWork W2317154179 @default.
• W3035993984 hasRelatedWork W2364708368 @default.
• W3035993984 hasRelatedWork W2972896016 @default.
• W3035993984 hasRelatedWork W3201665455 @default.
• W3035993984 hasRelatedWork W4287500247 @default.
• W3035993984 hasRelatedWork W4206894848 @default.
• W3035993984 hasVolume "72" @default.
• W3035993984 isParatext "false" @default.
• W3035993984 isRetracted "false" @default.
• W3035993984 magId "3035993984" @default.
• W3035993984 workType "article" @default.