A stable and quantitative method for dimensionality reduction of aortic centerline

Area of Science:

• Medical Imaging
• Cardiovascular Research
• Geometric Morphometrics

Background:

• Aortic dissection (AD) is a life-threatening condition requiring accurate aortic morphology assessment for diagnosis and treatment.
• Current aortic centerline projection methods (PPP) show significant discrepancies with individual patient anatomy, limiting statistical analysis.
• Reducing 3D aortic data to 2D is crucial for effective clinical decision-making.

Purpose of the Study:

• To develop an objective method for selecting the optimal projection plane of the aorta.
• To evaluate the efficacy of novel parameters in characterizing aortic morphology for AD analysis.
• To improve the discrimination of aortic tortuosity in AD patients.

Main Methods:

• Introduction of evaluation parameters: contour length (CL), enclosure area, and sum of absolute residuals (SAR).
• Objective evaluation of the optimal projection plane using these parameters, replacing subjective judgments.
• Comparison of the proposed method with traditional primary plane projection (PPP) for AD discrimination.

Main Results:

• The optimal projection plane can be objectively identified using CL, enclosure area, and SAR.
• SAR was found to be the optimal morphological criterion among the evaluated parameters.
• The proposed method demonstrated superior AD discrimination in aortic tortuosity analysis compared to traditional PPP.

Conclusions:

• Objective characterization of the optimal aortic projection plane is feasible using specific evaluation parameters.
• The novel method enhances the analysis of aortic tortuosity in AD, offering clinical advantages.
• This technique shows promise as a preprocessing step for geometric morphology analysis in AD.