Automatic Laplacian-based shape optimization for patient-specific vascular grafts.

Area of Science:

• Biomedical Engineering
• Computational Fluid Dynamics
• Medical Device Design

Background:

• Congenital heart disease is a major cause of newborn mortality.
• Current tissue-engineered vascular grafts often lack personalization and require extensive manual design.
• Patient-specific grafts are crucial for effective treatment of aortic arch defects.

Purpose of the Study:

• To develop a computational framework for automatic shape optimization of patient-specific tissue-engineered vascular grafts.
• To reduce manual intervention in the design process for congenital heart disease treatment.
• To improve hemodynamic performance of vascular grafts for aortic arch repair.

Main Methods:

• Utilized a computational framework combining Bayesian optimization with OpenFOAM and a novel graft deformation algorithm.
• Employed Laplacian mode computation and Gaussian process surrogate modeling for efficient optimization.
• Evaluated the framework using imaging and flow data from six patients with congenital heart disease.

Main Results:

• The automated framework successfully optimized graft shapes, reducing inlet-outlet pressure drop (PD) and maximum wall shear stress (WSS).
• Automated designs achieved at least a 16% reduction in blood flow pressure drop compared to human-optimized geometries.
• Demonstrated the potential of online training and hemodynamic surrogate model optimization for personalized graft design.

Conclusions:

• The developed computational framework offers an automated solution for designing patient-specific vascular grafts.
• This approach significantly improves hemodynamic efficiency compared to traditional methods.
• The findings support the use of automated, personalized designs for treating congenital heart disease, reducing mortality and improving patient outcomes.