Statistical techniques for predicting rupture risk in abdominal aortic aneurysms: A contribution based on bootstrap.

Area of Science:

• Biomedical Engineering
• Cardiovascular Research
• Medical Statistics

Background:

• Abdominal aortic aneurysm (AAA) morphometry is a key factor in rupture risk.
• Patient-specific quantification of AAA morphology aids in rupture risk prediction.
• Previous studies correlated AAA hemodynamics and morphometry using Pearson coefficient.

Purpose of the Study:

• To investigate the correlation between AAA morphology and hemodynamic stresses.
• To explore the bootstrap technique for predicting patient-specific AAA rupture risk.
• To enhance understanding of AAA rupture and improve clinical management.

Main Methods:

• Simulated 13 patient-specific AAA models.
• Applied the bootstrap technique to analyze morphological correlations with hemodynamic stresses.
• Compared bootstrap results with regression analysis using the Pearson coefficient.

Main Results:

• The bootstrap technique provides more reliable and robust results than Pearson regression, particularly for small, variable datasets.
• Morphological indices like AAA length, deformation rate, saccular index, and asymmetry are significant predictors.
• The developed methodology predicts hemodynamic stresses from size and shape indices.

Conclusions:

• The bootstrap technique is a valuable tool for predicting AAA rupture risk from morphology, especially with limited patient data.
• This approach offers improved reliability over traditional methods.
• Findings can lead to better clinical management of patients with abdominal aortic aneurysms.