CoroEval: a multi-platform, multi-modality tool for the evaluation of 3D coronary vessel reconstructions

Area of Science:

• Medical imaging analysis
• Computational cardiology
• Software development for medical research

Background:

• Accurate evaluation of 3D coronary vessel reconstructions is crucial for advancing cardiovascular imaging.
• Existing methods for assessing reconstruction quality can be subjective and labor-intensive.
• Need for modality-independent software to compare different reconstruction algorithms and protocols.

Purpose of the Study:

• To introduce CoroEval, a novel software tool for evaluating 3D coronary vessel reconstructions.
• To provide objective and automated metrics for assessing vessel sharpness and diameter.
• To facilitate the comparison of various reconstruction algorithms and acquisition protocols.

Main Methods:

• CoroEval utilizes raw intensity data for measurements, ensuring independence from display windowing.
• Supports both imported and manually performed vessel centerline segmentations.
• Incorporates an automated algorithm for correcting imperfect centrelines.
• Performs measurements at 1 mm intervals along the vessel centerline from 10 different angles for noise robustness.

Main Results:

• Demonstrated accurate diameter measurements with an average deviation of 0.03 ± 0.03 mm using phantom datasets.
• Graphical measurement results are exportable for scientific publications.
• Segmentation data can be exported in various formats, including point clouds and mesh formats.

Conclusions:

• CoroEval offers a robust, objective, and user-friendly solution for evaluating 3D coronary vessel reconstructions.
• The software enhances the reproducibility and reliability of quantitative assessments in cardiovascular imaging research.
• CoroEval is freely available for binary and source code download, promoting wider adoption in the research community.