Identification of the Unloaded Heart Configuration Including External Interactions

Area of Science:

• Computational mechanics
• Biomedical engineering
• Cardiac modeling

Background:

• Accurate patient-specific cardiac models require precise determination of the heart's unloaded reference configuration.
• Current inverse mechanics methods for estimating this configuration rely on accurate boundary conditions, often simplified as spring elements.
• Existing boundary conditions do not fully capture localized external forces from surrounding structures like the ribs and diaphragm.

Purpose of the Study:

• To develop a novel inverse mechanics approach for determining the unloaded cardiac geometry.
• To integrate localized external forces, not directly measurable from medical images, into the inverse mechanics formulation.
• To improve the physiological relevance and accuracy of estimated cardiac strains and stresses.

Main Methods:

• The proposed method modifies the inverse mechanics formulation by penalizing large deformations, implicitly accounting for external forces.
• This approach avoids complex optimization procedures typically required in inverse problems.
• The method was validated using a series of computational test problems.

Main Results:

• The novel approach successfully generated a reference cardiac configuration that closely matched the ground truth in test cases.
• Circumferential strain estimations were improved by an order of magnitude compared to standard inverse mechanics methods.
• The integration of external force effects through deformation penalization enhanced the accuracy of the unloaded geometry.

Conclusions:

• The developed inverse mechanics method provides a more accurate and physiologically meaningful reference configuration for cardiac models.
• Penalizing large deformations is an effective strategy for incorporating unmeasured external forces into cardiac mechanics.
• This advancement holds significant potential for improving the accuracy of patient-specific cardiac modeling and clinical applications.