Complementing sparse vascular imaging data by physiological adaptation rules.

Area of Science:

• Biomedical Engineering
• Computational Biology
• Cardiovascular Physiology

Background:

• Personalized pulse wave propagation (PWP) models require complete vascular geometry data, which is often unavailable in clinical practice.
• Existing PWP models struggle with incomplete patient-specific datasets, limiting their clinical utility.
• Vascular adaptation to mechanical load is a physiological process that can potentially fill data gaps.

Purpose of the Study:

• To develop and validate an adaptive PWP model that estimates missing vascular geometries using physiological adaptation rules.
• To assess the accuracy of the adaptive model in predicting arterial dimensions and waveform characteristics compared to a complete model.
• To enable personalized PWP modeling even with incomplete patient data.

Main Methods:

• Implemented homeostatic feedback loops in a validated PWP model to simulate vessel geometry adaptation.
• Used vascular MRI and ultrasound data from 10 healthy subjects.
• Compared adapted models (using limited data) with reference models (using complete data) for geometry and waveform accuracy.

Main Results:

• The adaptive PWP model accurately predicted arterial radius and wall thickness, with limits of agreement of 0.2 ± 2.6 mm and -140 ± 557 µm, respectively.
• Pressure and flow waveforms from the adapted models more closely resembled those from the reference models.
• The adaptation-based model demonstrated improved accuracy over non-adapted models with incomplete data.

Conclusions:

• The developed adaptation-based PWP model successfully personalizes vascular geometries using physiological adaptation rules, even with incomplete datasets.
• This approach offers a valuable tool for facilitating personalized cardiovascular modeling in clinical settings where data is often missing.
• The method enhances the feasibility of using PWP models for clinical decision-making by overcoming data limitations.