Three-dimensional micro-structurally informed in silico myocardium-Towards virtual imaging trials in cardiac...

Area of Science:

• Biomedical Imaging
• Computational Biology
• Medical Physics

Background:

• In silico tissue models are crucial for validating quantitative magnetic resonance imaging (MRI) models.
• Current models often lack detailed myocardial microstructure representation.
• Accurate phantoms are needed for sensitivity analysis of imaging biomarkers and tissue parameters.

Purpose of the Study:

• To develop a novel method for generating realistic numerical phantoms of myocardial microstructure.
• To enhance existing in silico models by incorporating cardiomyocyte shape variability, water exchange, and sheetlet orientations.
• To provide a more robust tool for MRI research and validation.

Main Methods:

• Generated numerical phantoms by simulating cardiomyocyte shape variability and intercalated disc connections.
• Aggregated and oriented myocardial sheetlets in specified directions.
• Employed morphometric analysis and structural correlation to compare simulated data with real tissue properties.

Main Results:

• No significant difference in cardiomyocyte morphometrics (volume, length, axes) between numerical and real data.
• Validated that the in silico tissue exhibits a similar disorder class to real myocardial tissue.
• Demonstrated good agreement between simulated and experimentally measured helical angles (HA) from cardiac diffusion tensor imaging (cDTI).

Conclusions:

• The proposed method generates richer and more realistic numerical phantoms of myocardial microstructure compared to previous approaches.
• The advanced phantoms facilitate improved validation and sensitivity analysis of quantitative MRI models.
• This work advances the development of computational tools for cardiovascular research.