Key result
An explainable machine learning model based on the Random Forest algorithm predicted atrial fibrillation recurrence after catheter ablation with an area under the curve of 66.7% in the testing cohort.
Why the study?
Existing machine learning models predicting AF recurrence after catheter ablation often suffer from a black-box effect, making it difficult to explain how variables affect model output.
Can an explainable machine learning model effectively identify patients with paroxysmal atrial fibrillation at high risk for recurrence after catheter ablation and reveal its decision-making process?
Cohort (n=471)
No
Can an explainable machine learning model effectively identify patients with paroxysmal atrial fibrillation at high risk for recurrence after catheter ablation and reveal its decision-making process?
Effect estimate: AUC 66.7%
An explainable machine learning model using SHAP analysis can effectively predict AF recurrence after catheter ablation and provide transparent, individualized risk thresholds to aid clinical decision-making.
May support individualized post-ablation risk assessment in paroxysmal AF; leaves open prospective validation and outcome impact.
Background A number of models have been reported for predicting atrial fibrillation (AF) recurrence after catheter ablation. Although many machine learning (ML) models were developed among them, black-box effect existed widely. It was always difficult to explain how variables affect model output. We sought to implement an explainable ML model and then reveal its decision-making process in identifying patients with paroxysmal AF at high risk for recurrence after catheter ablation. Methods Between January 2018 and December 2020, 471 consecutive patients with paroxysmal AF who had their first catheter ablation procedure were retrospectively enrolled. Patients were randomly assigned into training cohort (70%) and testing cohort (30%). The explainable ML model based on Random Forest (RF) algorithm was developed and modified on training cohort, and tested on testing cohort. In order to gain insight into the association between observed values and model output, Shapley additive explanations (SHAP) analysis was used to visualize the ML model. Results In this cohort, 135 patients (14.2/100 patient-years) experienced tachycardias recurrence. With hyperparameters adjusted, the ML model predicted AF recurrence with an area under the curve (AUC) of 66.7% in the testing cohort. Based on SHAP analysis, the ML model's decision-making process was revealed: (i) summary plot listed the top 15 features in descending order and preliminary showed the association between features and outcome prediction; (ii) dependence plots combined with force plots showed the impact of single feature on model output, and helped determine high risk cut-off points; (iii) decision plot recognized significant outliers. Conclusion An explainable ML model effectively revealed its decision-making process in identifying patients with paroxysmal atrial fibrillation at high risk for recurrence after catheter ablation. Physicians can combine model output, visualization of model and clinical experience to make better decision.
No takes yet. Share an insight, caveat, or question.
Ma et al. (2022) conducted a cohort in paroxysmal atrial fibrillation (n=471). Explainable Machine Learning model (Random Forest) was evaluated on Prediction of atrial fibrillation recurrence in the testing cohort (AUC 66.7%). An explainable machine learning model based on the Random Forest algorithm predicted atrial fibrillation recurrence after catheter ablation with an area under the curve of 66.7% in the testing cohort.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: