JMIR Publications

ABSTRACT

Background:

Although catheter ablation (CA) is currently the most effective clinical treatment for atrial fibrillation (AF), its variable therapeutic effects among individual patients present numerous problems. Machine learning (ML) shows promising potential in optimizing the management and clinical outcomes of AFCA patients.

Objective:

This scoping review aims to evaluate the current scientific evidence on the application of ML for managing AFCA patients, compare the performance of various models across specific clinical tasks within AFCA, and summarize the strengths and limitations of ML in this field.

Methods:

Adhering to the PRISMA extension for Scoping Reviews guidelines, relevant studies were searched from PubMed, Web of Science, Embase, the Cochrane Library, and ScienceDirect. The final included studies were confirmed based on inclusion and exclusion criteria and manual review. Methodological quality assessment tools (QUADAS-2 and PROBAST) were used to review the included studies, and narrative data synthesis was performed on the modeled results provided by these studies.

Results:

The analysis of 23 included studies showcased ML's contributions in (1) identifying potential ablation targets, (2) improving ablation strategies, and (3) patient prognosis. The patient data utilized in these studies comprised demographics, clinical characteristics, various types of imaging (n=9, 39%), and electrophysiological signals (n=7, 30%). In terms of model type, deep learning, represented by CNN, was most frequently applied (n=14, 61%). Compared with traditional clinical scoring models or human clinicians, the model performance reported in the included studies was generally satisfactory, but most models showed a high risk of bias due to lack of external validation (n=14).

Conclusions:

Our evidence-based findings suggest that ML is a promising tool for improving the effectiveness and efficiency of AFCA patient management. While guiding data preparation and model selection for future studies, this review highlights the need to address prevalent limitations, including lack of external validation and further exploration of model generalization and interpretability.