FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3127203132> ?p ?o ?g. }

• W3127203132 abstract "Introduction: Accurate risk stratification for patients with coronaryartery disease(CAD) is essential for accurate treatment. The current diagnosticpathway comprises a number of medical examinations , including acomputed tomography scan and positron emission tomography myocardialperfusion, which yield prognostic data that may be utilized for risk stratificationpurposes. The aim of this thesis was to develop a risk model forobstructive CAD with machine learning(ML) algorithms. This model mayprovide an individualized risk score based on a combination of clinical featuresand quantitative parameters derived from imaging.Methods: We retrospectively included 1007 patients with no prior cardiovascularhistory, who were referred for rest and regadenoson-inducedstress Rubidium-82 positron emission tomograpy (PET)/computated tomography(CT). Presence of obstructive CAD was defined as a composite of asignificant fractional flow rate measurement during invasive coronary angiography,percutaneous coronary intervention or a coronary artery bypassgraft procedure, and was acquired via follow-up. Furthermore, each patientwas characterized by a broad array of features, including cardiovascularrisk factors (cigarette smoking, hypertension, hypercholesterolemia, diabetes,positive family history of CAD), prior medical history; current medicationusage age; gender; body mass index (BMI); creatinine serum values;coronary artery calcification (CAC) score and PET/CT derived myocardialblood flows. Additionally, the visual interpretation by a team of two cliniciansof the PET/CT scan was obtained. Two sets of input parameter wereused to train the models. First, the entire set of features except the visualinterpretation. Secondly, the entire set of features, including the visual interpretation.Four different ML algorithms were used, so in total, 8 differentmodels were optimized. These models were developed using a subset of 805cases of the dataset to identify obstructive CAD by using 5-fold cross validationin combination with a grid search, whilst their performance was measuredusing the F1-score. The optimized algorithms were validated on 202cases of the dataset, never previously seen by the models. The performanceon these unseen examples was compared with the current diagnostic performanceby clinicians, as measured by the visual interpretation of the scan.Results: The best performing algorithm to predict obstructive CAD wasXGBoost, an ensemble of gradient boosted decision trees. On the unseendataset this algorithm reached an area under the curve of 0.93 while obtaininga sensitivity of 64% (95% CI: 41-83) and a specificity of 96% (95% CI:91-98). The sensitivity by the clinicians on this same dataset was 77% (95%CI 55-93) and the specificity was 92% (95% CI (87-96). The low prevalenceof obstructive events in evaluation dataset (11%) resulted in wide confidenceintervals, making it so that no significant differences were found. Furthermore,we were able to make a ranking via the XGBoost model of importantpredictors for obstructive CAD. Summarized, CAC-scores and quantitative PET derived features were the most important predictors. Classical risk factorsand medication however, could not be used in the current setup to distinguishobstructive CAD from non-obstructive CAD. We also conclude thatthe visual interpretation by the clinician added incremental prognostic informationto the model. Conclusion: We used a set of clinical and quantitative features to developa ML model. This model is able to provide individualized risk stratificationby predicting the possibility of an obstructive cardiovascular event.Although validation with a larger dataset could result in a more well definedperformance range, this model still shows potential to be implemented in thediagnostic workflow by providing a computer aided second opinion to theclinicians." @default.
• W3127203132 created "2021-02-15" @default.
• W3127203132 creator A5063601073 @default.
• W3127203132 date "2020-01-01" @default.
• W3127203132 modified "2023-09-27" @default.
• W3127203132 title "Prediction of obstructive coronary artery disease using machine learning algorithms" @default.
• W3127203132 hasPublicationYear "2020" @default.
• W3127203132 type Work @default.
• W3127203132 sameAs 3127203132 @default.
• W3127203132 citedByCount "0" @default.
• W3127203132 crossrefType "dissertation" @default.
• W3127203132 hasAuthorship W3127203132A5063601073 @default.
• W3127203132 hasConcept C11413529 @default.
• W3127203132 hasConcept C119857082 @default.
• W3127203132 hasConcept C126322002 @default.
• W3127203132 hasConcept C126838900 @default.
• W3127203132 hasConcept C164705383 @default.
• W3127203132 hasConcept C2775842073 @default.
• W3127203132 hasConcept C2778213512 @default.
• W3127203132 hasConcept C2778405248 @default.
• W3127203132 hasConcept C2780221984 @default.
• W3127203132 hasConcept C41008148 @default.
• W3127203132 hasConcept C71924100 @default.
• W3127203132 hasConceptScore W3127203132C11413529 @default.
• W3127203132 hasConceptScore W3127203132C119857082 @default.
• W3127203132 hasConceptScore W3127203132C126322002 @default.
• W3127203132 hasConceptScore W3127203132C126838900 @default.
• W3127203132 hasConceptScore W3127203132C164705383 @default.
• W3127203132 hasConceptScore W3127203132C2775842073 @default.
• W3127203132 hasConceptScore W3127203132C2778213512 @default.
• W3127203132 hasConceptScore W3127203132C2778405248 @default.
• W3127203132 hasConceptScore W3127203132C2780221984 @default.
• W3127203132 hasConceptScore W3127203132C41008148 @default.
• W3127203132 hasConceptScore W3127203132C71924100 @default.
• W3127203132 hasLocation W31272031321 @default.
• W3127203132 hasOpenAccess W3127203132 @default.
• W3127203132 hasPrimaryLocation W31272031321 @default.
• W3127203132 hasRelatedWork W1491114543 @default.
• W3127203132 hasRelatedWork W1592450369 @default.
• W3127203132 hasRelatedWork W1992158879 @default.
• W3127203132 hasRelatedWork W2350531025 @default.
• W3127203132 hasRelatedWork W2405790345 @default.
• W3127203132 hasRelatedWork W2596707319 @default.
• W3127203132 hasRelatedWork W2767698820 @default.
• W3127203132 hasRelatedWork W2793392961 @default.
• W3127203132 hasRelatedWork W2803377716 @default.
• W3127203132 hasRelatedWork W2885387053 @default.
• W3127203132 hasRelatedWork W2887934894 @default.
• W3127203132 hasRelatedWork W2979924912 @default.
• W3127203132 hasRelatedWork W2996623875 @default.
• W3127203132 hasRelatedWork W2998798271 @default.
• W3127203132 hasRelatedWork W3002916497 @default.
• W3127203132 hasRelatedWork W3011311220 @default.
• W3127203132 hasRelatedWork W3126610102 @default.
• W3127203132 hasRelatedWork W3175478006 @default.
• W3127203132 hasRelatedWork W2512076099 @default.
• W3127203132 hasRelatedWork W3107365476 @default.
• W3127203132 isParatext "false" @default.
• W3127203132 isRetracted "false" @default.
• W3127203132 magId "3127203132" @default.
• W3127203132 workType "dissertation" @default.