A generalized additive model using clinical parameters accurately detected atrial fibrillation with an area under the curve of 0.964, sensitivity of 0.879, and specificity of 0.920.
Cross-Sectional (n=173,328)
Does an explainable machine learning model using a generalized additive model accurately predict atrial fibrillation in the general population using routine health checkup data?
An explainable machine learning model using a generalized additive model accurately predicts atrial fibrillation from routine health checkup data, highlighting non-linear relationships between clinical parameters and AF risk.
Tasa de eventos absoluta: 0.964% vs 0.962%
Background:Atrial fibrillation (AF) is the most common arrhythmia and is associated with increased thromboembolic stroke risk and heart failure. Although various prediction models for AF risk have been developed using machine learning, their output cannot be accurately explained to doctors and patients. Therefore, we developed an explainable model with high interpretability and accuracy accounting for the non-linear effects of clinical characteristics on AF incidence.
Kawakami et al. (Mon,) conducted a cross-sectional in Atrial Fibrillation (n=173,328). Generalized Additive Model (GAM) vs. Generalized Linear Model (GLM) was evaluated on Area under the curve (AUC) for atrial fibrillation detection. A generalized additive model using clinical parameters accurately detected atrial fibrillation with an area under the curve of 0.964, sensitivity of 0.879, and specificity of 0.920.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: