FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4377015749> ?p ?o ?g. }

• W4377015749 endingPage "S572" @default.
• W4377015749 startingPage "S571" @default.
• W4377015749 abstract "Arrhythmogenic right ventricular cardiomyopathy (ARVC) is a genetically determined cardiomyopathy with a high risk of malignant ventricular arrhythmias (MVA). Determining an individual patient’s arrhythmic risk remains however challenging. Artificial intelligence (AI) has been opted as a tool to detect slight variances in electrocardiograms (ECGs), which could help predicting VA in this population. To predict ventricular arrhythmias in patients with definite ARVC using an in-house developed explainable AI-tool. All patients with definite ARVC by 2010 Task Force Criteria (TFC), no history of prior sustained ventricular arrhythmia, and with a12-lead ECG available for analysis were included from our center. Demographics, clinical phenotype and outcomes were ascertained. MVA was defined as (aborted) sudden cardiac death, sustained ventricular tachycardia, or appropriate implantable cardioverter-defibrillator therapy. We utilized FactorECG, an in-house developed explainable deep neural network algorithm that was trained on 1.1 million ECGs from 251473 patients, as our AI-tool. In short, FactorECG compresses the median beat ECG, thereby summarizing the ECG into 21 explainable factors. An online version has been made available to provide interactive visualizations (https://decoder.ecgx.ai). Univariable Cox proportional hazard regression was used to evaluate the prognostic value of the 21 ECG factors for predicting MVA. We included 79 ARVC patients (49% male, age at diagnosis 41.6 (interquartile range (IQR) 32.8-51.9) years, 39% proband, 57% plakophilin-2 carriers). During 8.5 (IQR 4.2-11.2) years of follow-up, 21 (27%) developed a MVA. Male patients were more likely to develop MVA (71% vs. 41%; p=0.035). Age, symptoms, and presence of a class 4/5 variant were not associated with development of MVA (p>0.05). Figure 1 shows ECG features associated with 5-year predicted risk of MVA. With univariate Cox proportional hazard regression, the factor corresponding for lateral ST deviation was statistically significant (hazard ratio 0.52 (95% confidence interval: 0.30-0.89); p=0.017). Factors corresponding with inferolateral ST deviation and onset of depolarization had a tendency towards significance (Table 1). Explainable AI has the potential to detect features associated with development of MVA in patients with definite ARVC. Future studies with a larger sample size should focus on the incremental value of FactorECG over currently available risk stratification schemes." @default.
• W4377015749 created "2023-05-19" @default.
• W4377015749 creator A5025295873 @default.
• W4377015749 creator A5034268510 @default.
• W4377015749 creator A5056895279 @default.
• W4377015749 creator A5090978835 @default.
• W4377015749 date "2023-05-01" @default.
• W4377015749 modified "2023-09-26" @default.
• W4377015749 title "PO-04-197 EXPLAINABLE ARTIFICIAL INTELLIGENCE IN ELECTROCARDIOGRAPHY TO PREDICT VENTRICULAR ARRHYTHMIAS IN PATIENTS WITH ARRHYTHMOGENIC RIGHT VENTRICULAR CARDIOMYOPATHY" @default.
• W4377015749 doi "https://doi.org/10.1016/j.hrthm.2023.03.1213" @default.
• W4377015749 hasPublicationYear "2023" @default.
• W4377015749 type Work @default.
• W4377015749 citedByCount "0" @default.
• W4377015749 crossrefType "journal-article" @default.
• W4377015749 hasAuthorship W4377015749A5025295873 @default.
• W4377015749 hasAuthorship W4377015749A5034268510 @default.
• W4377015749 hasAuthorship W4377015749A5056895279 @default.
• W4377015749 hasAuthorship W4377015749A5090978835 @default.
• W4377015749 hasBestOaLocation W43770157491 @default.
• W4377015749 hasConcept C119060515 @default.
• W4377015749 hasConcept C126322002 @default.
• W4377015749 hasConcept C164705383 @default.
• W4377015749 hasConcept C2775935837 @default.
• W4377015749 hasConcept C2776331378 @default.
• W4377015749 hasConcept C2777093960 @default.
• W4377015749 hasConcept C2778198053 @default.
• W4377015749 hasConcept C2778797674 @default.
• W4377015749 hasConcept C2781005686 @default.
• W4377015749 hasConcept C71924100 @default.
• W4377015749 hasConceptScore W4377015749C119060515 @default.
• W4377015749 hasConceptScore W4377015749C126322002 @default.
• W4377015749 hasConceptScore W4377015749C164705383 @default.
• W4377015749 hasConceptScore W4377015749C2775935837 @default.
• W4377015749 hasConceptScore W4377015749C2776331378 @default.
• W4377015749 hasConceptScore W4377015749C2777093960 @default.
• W4377015749 hasConceptScore W4377015749C2778198053 @default.
• W4377015749 hasConceptScore W4377015749C2778797674 @default.
• W4377015749 hasConceptScore W4377015749C2781005686 @default.
• W4377015749 hasConceptScore W4377015749C71924100 @default.
• W4377015749 hasIssue "5" @default.
• W4377015749 hasLocation W43770157491 @default.
• W4377015749 hasOpenAccess W4377015749 @default.
• W4377015749 hasPrimaryLocation W43770157491 @default.
• W4377015749 hasRelatedWork W108924682 @default.
• W4377015749 hasRelatedWork W1534090624 @default.
• W4377015749 hasRelatedWork W2092097770 @default.
• W4377015749 hasRelatedWork W2471786597 @default.
• W4377015749 hasRelatedWork W2581104969 @default.
• W4377015749 hasRelatedWork W2789805016 @default.
• W4377015749 hasRelatedWork W2896445697 @default.
• W4377015749 hasRelatedWork W2945088353 @default.
• W4377015749 hasRelatedWork W4280530690 @default.
• W4377015749 hasRelatedWork W4378233385 @default.
• W4377015749 hasVolume "20" @default.
• W4377015749 isParatext "false" @default.
• W4377015749 isRetracted "false" @default.
• W4377015749 workType "article" @default.