Predicting thrombus formation and growth in patient-specific models of aortic dissection: a multiphase approach based on porous media theory

Aortic dissection represents a life-threatening vascular emergency with significant morbidity and mortality. Traditional models for predicting aortic thrombosis often depend on complex biochemical parameters, lack clearly defined phase interfaces, and require extensive computational time. Existing porous media algorithms are limited in their ability to accurately capture the dynamic processes of thrombus growth and hemodynamic changes, largely due to imprecise physical formulations. This study presents a novel multiphase porous media approach for predicting thrombus formation in various types of aortic dissection, which is innovatively applied to a large number of patient-specific aortic models. By incorporating an extended Darcy–Brinkman–Stokes (DBS) equation to explicitly model the interaction between solid and liquid phases, and introducing a novel porosity equation to simplify platelet transport and deposition, the method achieves substantial improvements in computational efficiency. Applied to computed tomography-based reconstructions, the algorithm demonstrated high predictive accuracy, achieving a correlation coefficient of 0.97 between predicted and actual thrombus volumes in 12 cases of partial false lumen and 9 cases of complete false lumen. The average prediction time per case was reduced to 40 min, representing a 70 % improvement in efficiency. Furthermore, the study investigated mechanical factors underlying enhanced postoperative recovery in patients with complete false lumens and introduced an acceleration factor to align simulation time with actual thrombus progression. By integrating a mechanically grounded thrombus evolution model, this method enables rapid, dynamic predictions, thereby supporting timely clinical decision-making and facilitating the development of personalized treatment strategies for patients with aortic dissection.