Matches in SemOpenAlex for { <https://semopenalex.org/work/W2912358227> ?p ?o ?g. }
Showing items 1 to 75 of
75
with 100 items per page.
- W2912358227 endingPage "230a" @default.
- W2912358227 startingPage "230a" @default.
- W2912358227 abstract "When assessing whether an otherwise healthy individual is at risk of developing an arrhythmia, the most common assessment is based on the electrocardiographic QT interval. Although this trait predicts susceptibility, there remains a percentage of patients who still succumb to an arrhythmia with a normal QT, or at the cellular level a normal action potential duration (APD). Machine learning (ML) can unmask hidden features and provide greater insight needed to distinguish between healthy and arrhythmia-prone patients. We sought to use this technique along with systems pharmacology models that describe the electrophysiology of ventricular myocytes to determine key ionic channel abnormalities and experimentally quantifiable metrics of the AP shape that are unique to highly susceptible cells. We generated heterogeneous populations of cells using three human ventricular myocyte models, and examined the effect in each population of applying various arrhythmic triggers (increased L-type calcium current, hypokalemia, block of hERG channel, increased inward current). We then trained ML algorithms (SVM, Random Forest, logistic regression, artificial neural network) on features pertaining to the pre-perturbed AP and the outcome of the applied trigger. Our main goal was to find a set of features that could predict risk no matter which perturbation was applied. We found that: (1) ∼11% of each population was susceptible to all four perturbations, (2) APD was a strong predictor of susceptibility, confirming the utility of testing QT interval (3) incorporating additional features (besides APD) describing the AP and the calcium transient improves prediction accuracy, and (4) the accuracy of the ML algorithms changes based on the perturbation applied. Overall, this computational pipeline can reveal patients who appear to be healthy but are highly susceptible to an arrhythmia, thus providing an automated approach to screen for patient risk." @default.
- W2912358227 created "2019-02-21" @default.
- W2912358227 creator A5000485841 @default.
- W2912358227 creator A5052680047 @default.
- W2912358227 creator A5079389528 @default.
- W2912358227 date "2019-02-01" @default.
- W2912358227 modified "2023-09-30" @default.
- W2912358227 title "Combining Systems Pharmacology Modeling with Machine Learning to Identify Sub-Populations at Risk of Arrhythmia" @default.
- W2912358227 doi "https://doi.org/10.1016/j.bpj.2018.11.1266" @default.
- W2912358227 hasPublicationYear "2019" @default.
- W2912358227 type Work @default.
- W2912358227 sameAs 2912358227 @default.
- W2912358227 citedByCount "0" @default.
- W2912358227 crossrefType "journal-article" @default.
- W2912358227 hasAuthorship W2912358227A5000485841 @default.
- W2912358227 hasAuthorship W2912358227A5052680047 @default.
- W2912358227 hasAuthorship W2912358227A5079389528 @default.
- W2912358227 hasBestOaLocation W29123582271 @default.
- W2912358227 hasConcept C118441451 @default.
- W2912358227 hasConcept C119857082 @default.
- W2912358227 hasConcept C126322002 @default.
- W2912358227 hasConcept C146403970 @default.
- W2912358227 hasConcept C151956035 @default.
- W2912358227 hasConcept C154945302 @default.
- W2912358227 hasConcept C164705383 @default.
- W2912358227 hasConcept C169258074 @default.
- W2912358227 hasConcept C185263204 @default.
- W2912358227 hasConcept C2777796741 @default.
- W2912358227 hasConcept C2778242168 @default.
- W2912358227 hasConcept C2779362680 @default.
- W2912358227 hasConcept C2908647359 @default.
- W2912358227 hasConcept C41008148 @default.
- W2912358227 hasConcept C47423411 @default.
- W2912358227 hasConcept C71924100 @default.
- W2912358227 hasConcept C83743174 @default.
- W2912358227 hasConcept C99454951 @default.
- W2912358227 hasConceptScore W2912358227C118441451 @default.
- W2912358227 hasConceptScore W2912358227C119857082 @default.
- W2912358227 hasConceptScore W2912358227C126322002 @default.
- W2912358227 hasConceptScore W2912358227C146403970 @default.
- W2912358227 hasConceptScore W2912358227C151956035 @default.
- W2912358227 hasConceptScore W2912358227C154945302 @default.
- W2912358227 hasConceptScore W2912358227C164705383 @default.
- W2912358227 hasConceptScore W2912358227C169258074 @default.
- W2912358227 hasConceptScore W2912358227C185263204 @default.
- W2912358227 hasConceptScore W2912358227C2777796741 @default.
- W2912358227 hasConceptScore W2912358227C2778242168 @default.
- W2912358227 hasConceptScore W2912358227C2779362680 @default.
- W2912358227 hasConceptScore W2912358227C2908647359 @default.
- W2912358227 hasConceptScore W2912358227C41008148 @default.
- W2912358227 hasConceptScore W2912358227C47423411 @default.
- W2912358227 hasConceptScore W2912358227C71924100 @default.
- W2912358227 hasConceptScore W2912358227C83743174 @default.
- W2912358227 hasConceptScore W2912358227C99454951 @default.
- W2912358227 hasIssue "3" @default.
- W2912358227 hasLocation W29123582271 @default.
- W2912358227 hasOpenAccess W2912358227 @default.
- W2912358227 hasPrimaryLocation W29123582271 @default.
- W2912358227 hasRelatedWork W1508014482 @default.
- W2912358227 hasRelatedWork W2034172298 @default.
- W2912358227 hasRelatedWork W2040918531 @default.
- W2912358227 hasRelatedWork W2118871889 @default.
- W2912358227 hasRelatedWork W2412761636 @default.
- W2912358227 hasRelatedWork W2749104498 @default.
- W2912358227 hasRelatedWork W2910323577 @default.
- W2912358227 hasRelatedWork W2982201518 @default.
- W2912358227 hasRelatedWork W3033579662 @default.
- W2912358227 hasRelatedWork W3141561812 @default.
- W2912358227 hasVolume "116" @default.
- W2912358227 isParatext "false" @default.
- W2912358227 isRetracted "false" @default.
- W2912358227 magId "2912358227" @default.
- W2912358227 workType "article" @default.