Structural and practical identifiability of contrast transport models for DCE-MRI

Area of Science:

• Medical Imaging
• Biophysics
• Pharmacokinetics

Background:

• Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) relies on contrast transport models to analyze blood flow and agent concentration over time.
• Accurate estimation of model parameters is crucial for the diagnostic and prognostic value of DCE-MRI in various biological systems.

Purpose of the Study:

• To investigate the structural and practical identifiability of nested compartment models used in DCE-MRI data analysis.
• To assess the impact of data noise on the reliability of parameter estimation in these models.

Main Methods:

• Analysis of structural and practical identifiability for a class of nested compartment models.
• Utilization of both artificial and real DCE-MRI data to evaluate parameter estimation under varying noise conditions.

Main Results:

• All analyzed models were found to be structurally identifiable.
• Practical identifiability was significantly influenced by data characteristics, particularly noise levels.
• Increased data quality was demonstrated to recover parameter identifiability, even with higher noise.

Conclusions:

• The identifiability of DCE-MRI contrast transport models is robustly dependent on data quality rather than specific tissue types or contrast enhancement patterns.
• Ensuring high-quality data is paramount for reliable parameter estimation and the clinical utility of DCE-MRI.
• These findings are critical for standardizing DCE-MRI analysis and improving diagnostic accuracy.