Real-Time Peripheral Revascularization Planning in Chronic Limb Threatening Ischemia Using HarVI: A Digital Twin...

Area of Science:

• Vascular Surgery
• Medical Imaging
• Computational Fluid Dynamics

Background:

• Peripheral artery disease (PAD) and chronic limb-threatening ischemia (CLTI) pose significant global health challenges, often leading to limb loss.
• Current image-guided endovascular interventions lack real-time, patient-specific hemodynamic outcome prediction tools for intraoperative planning.
• Existing computational fluid dynamics (CFD) methods are too slow for interactive surgical guidance.

Purpose of the Study:

• To extend the HarVI (HARVEY Virtual Intervention) digital twin framework to peripheral circulation.
• To evaluate HarVI's capability for real-time prediction of postoperative blood flow in superficial femoral artery (SFA) lesions.
• To assess HarVI's potential for guiding revascularization decisions in PAD patients.

Main Methods:

• Integration of 1D CFD with machine learning within the HarVI framework for rapid patient-specific analysis.
• Automated boundary condition tuning and simulation of various endovascular interventions using a machine-learned surrogate model.
• Validation of predicted hemodynamics against clinical duplex ultrasound measurements in a cohort of seven SFA disease patients.

Main Results:

• HarVI accurately predicted postoperative peak systolic velocities and replicated 1D CFD results.
• The machine-learned surrogate model provided rapid assessment of intervention scenarios, closely matching high-fidelity simulations.
• Near-real-time evaluation of treatment options was enabled, supporting intraoperative planning.

Conclusions:

• HarVI demonstrates significant promise as a digital twin platform for real-time, patient-specific intervention planning in PAD.
• The framework facilitates rapid, data-driven prediction of postoperative hemodynamics.
• HarVI has the potential to enhance intraoperative decision support and improve revascularization outcomes for CLTI patients.