CFD-derived radiomics from hemodynamic maps for quantitative assessment of left atrial flow in atrial fibrillation: a proof-of-concept study - Report - MDSpire
Advertisement
CFD-derived radiomics from hemodynamic maps for quantitative assessment of left atrial flow in atrial fibrillation: a proof-of-concept study
Clinical Report: Quantitative Evaluation of Left Atrial Flow in Atrial Fibrillation
Overview
This study demonstrates that radiomics-based quantification of left atrial flow can yield reproducible biomarkers for atrial fibrillation (AF) classification. The classifier achieved high accuracy, sensitivity, and specificity, indicating the potential for improved risk stratification in AF patients.
Background
Atrial fibrillation (AF) is the most common sustained cardiac arrhythmia, significantly increasing the risk of ischemic stroke. Understanding left atrial (LA) hemodynamics is crucial, as abnormal flow patterns can lead to thromboembolic events. Current imaging techniques have limitations in capturing complex flow dynamics, highlighting the need for advanced methodologies like radiomics and computational fluid dynamics (CFD).
Data Highlights
Parameter
Value
Classifier Accuracy
93%
Sensitivity
95%
Specificity
90%
AUC
0.92 (95% CI: 0.78–1.00)
Effect Size (ε2)
0.46–0.60
Key Findings
Five robust radiomics features were identified for LA flow biomarkers in AF.
Four features were Gray-Level Non-Uniformity features from throughflow regions, achieving 100% selection frequency.
The fifth feature was a Dependence Entropy feature from LAA helicity maps, with 80% selection frequency.
The classifier demonstrated 93% accuracy, 95% sensitivity, and 90% specificity.
Significant correlations were found between identified features and LA enlargement (ρ = 0.45–0.64) and mitral regurgitation grade (ρ = 0.49–0.69).
Clinical Implications
The findings suggest that radiomics analysis of CFD-derived hemodynamic maps can enhance the assessment of left atrial flow in AF patients. This approach may provide objective biomarkers for improved risk stratification and management of thromboembolic risk in clinical practice.
Conclusion
Radiomics-based quantification of LA flow offers a promising avenue for identifying biomarkers in AF. Further validation in larger cohorts is essential to establish its clinical utility.
