Generation of Patient-Specific Cardiac Vascular Networks: A Hybrid Image-Based and Synthetic Geometric Model

Area of Science:

• Biomedical Engineering
• Computational Biology
• Cardiovascular Research

Background:

• Current CT angiography limitations restrict vessel segmentation to larger epicardial arteries.
• Patient-specific models are crucial for simulating coronary artery blood flow for diagnosis and treatment planning.
• Existing methods lack the ability to generate detailed vascular networks down to the arteriolar level.

Purpose of the Study:

• To propose a novel algorithm for generating patient-specific cardiac vascular networks.
• To extend existing tree generation methods to simulate angiogenesis and incorporate functional principles.
• To create comprehensive vascular models from epicardial vessels down to arterioles.

Main Methods:

• Utilized a constrained constructive optimization method extended for multiple, competing vascular trees.
• Simulated angiogenesis incorporating vascular volume minimization, flow-related, and geometrical constraints.
• Adapted simultaneous tree growths to patient-specific data and priors.

Main Results:

• Generated patient-specific vascular networks with 6000 terminal segments for six patients.
• Created networks comprising 33 to 62 synthetic trees, filling the left ventricle myocardium.
• Validated that the generated vascular models accurately match previously described morphometry properties.

Conclusions:

• The developed algorithm successfully generates detailed, patient-specific cardiac vascular networks.
• These models can simulate blood flow from the aorta into the myocardium, overcoming CT image resolution limitations.
• This advancement is significant for improving the diagnosis and treatment planning of coronary artery disease.