A flexible generative algorithm for growing in silico placentas.

Area of Science:

• Biomedical Engineering
• Computational Biology
• Reproductive Medicine

Background:

• The placenta is vital for fetal development, with abnormalities linked to pregnancy complications like pre-eclampsia.
• Understanding placental vascular structure is key to diagnosing and managing fetal growth restriction and other issues.
• Current computational models lack precise control over vascular morphology, limiting their predictive power for placental dysfunction.

Purpose of the Study:

• To introduce a novel generative algorithm for creating customizable in silico placental vascular networks.
• To enable user control over key morphological parameters of the feto-placental vasculature.
• To provide a tool for investigating the relationship between placental structure and function.

Main Methods:

• Developed a generative algorithm based on physiological branching laws (e.g., Murray's Law).
• Algorithm defined by vessel diameter, length, branching angle, and asymmetry for customisation.
• Generated synthetic placental vascular structures with user-controlled parameters and stochastic variations.

Main Results:

• The algorithm successfully generated in silico placentas consistent with in vivo and ex vivo measurements.
• Sensitivity analysis revealed vessel length and branching angles significantly influence vascular network architecture.
• The stochastic nature of the algorithm produces diverse topological metrics for identical input parameters.

Conclusions:

• The novel algorithm offers direct control over key morphological parameters, unlike previous methods.
• This approach generates realistic vascular densities, facilitating studies on placental function.
• The tool enables detailed investigation into how specific vascular parameters impact placental health and pregnancy outcomes.