An algorithm for generating biophysically realistic three-dimensional arteriolar networks applied to rat skeletal muscle.

The microcirculation comprises small vessel networks that regulate blood perfusion within tissues. The relationship between tissue shape or size and its microvascular properties is not yet clear. This study develops an algorithm for computationally simulating branching arteriolar networks within ellipsoidal tissue volumes, including user-adjustable parameters (e.g., tissue width-length-height dimensions and microvessel density) for application within different rodent skeletal muscles. The algorithm is developed using principles from constrained constructive optimization, an iterative network generation framework based on proposed mechanisms of vascular growth. Networks generated within muscles of varying shapes and sizes were analyzed over a range of geometric (e.g., mean diameter, length, and number of bifurcations per Strahler's and centrifugal order, fractal dimension) and hemodynamic (e.g., Murray's law exponent, hematocrit) properties. Statistical similarity was observed across different skeletal muscle tissues, with differences due to tissue shape being observed only above a vessel diameter threshold of ~25 μm (varying at large or small tissue volumes at the scale m3 or mm3). The algorithm was comprehensively validated against in vivo data using different modeling approaches (whole tissue vs. subsection simulations). The algorithm's accuracy and adaptability support a wide range of research objectives and contributes to advancing current understanding of perfusion distribution in healthy tissue.