What does this research mean for the field?

Surface ECG measurements can be used to optimally group atrial fibrillation into five distinct sub-classes using clustering algorithms, with classification separability improving when spectral concentration is discarded as a feature. Novelty: ClaimNovelty.METHODOLOGICAL. Consensus alignment: ConsensusAlignment.NEUTRAL.

July 1, 2013

Clustering of atrial fibrillation based on surface ECG measurements

View Full Paper

Ask AI

Bookmark

Structured PICO

Population

218 real 12-lead ECG recordings, from which a subset of 136 atrial activity signals displaying known properties of atrial fibrillation were extracted

Intervention

k-means and hierarchical clustering algorithms using peak frequency mean value (PFM), peak frequency standard deviation (PFSD) and spectral concentration (SC) as features

Outcome

Optimal number of clusters and clustering quality (separability)

Machine learning clustering of surface ECG features can identify 5 distinct sub-classes of atrial fibrillation, providing a basis for future automated classification.

Discussion

No takes yet. Share an insight, caveat, or question.

Implication

Hypothesis-generating for ECG-based AF subtyping; prospective validation required before clinical consideration.

Cite This Study

Donoso et al. (2013) studied this question.

synapsesocial.com/papers/6a1bd37cea84844e355f0966 https://doi.org/10.1109/embc.2013.6610472

Topics

Atrial fibrillation Persistent AF management AF ablation outcomes

View Full Paper

Ask AI

Bookmark

Clustering of atrial fibrillation based on surface ECG measurements

Structured PICO

Discussion

Implication

Cite This Study

Topics

Clustering of atrial fibrillation based on surface ECG measurements

Structured PICO

Discussion

Implication

Cite This Study

Also Consider

Topics

Also Consider