Description:
- Magnetic resonance imaging
- Tracer kinetic map generation
- Disease diagnosis and monitoring
Abstract
USC researchers have developed a method of DCE-MRI reconstruction that allows different TK models and solvers to be integrated, as well as joint estimation of patient-specific AIF and TK maps. This is achieved by setting the TK model consistency as a statistical constraint with a penalty, decoupling it from the data consistency. This enables the data fit to deviate from the TK model and allows for situations where the model may be inconsistent. The individualized Pat-AIF can thus be iteratively estimated from the undersampled data, rather than a less accurate population-averaged AIF. This new approach enables much more flexibility for direct model-based reconstruction of accelerated DCE-MRI.
Benefit
- Creation of accurate tracer-kinetic maps and individualized patient arterial input functions with undersampled data
- Multiple tracer-kinetic models allowed
Market Application
Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) is a powerful tool that accurately probes the vascular properties of human tissue and involves capturing images before, during, and after the administration of a contrast agent. From these images, patient-specific arterial input functions (Pat-AIF) can be identified to create accurate tracer-kinetic (TK) maps, which are used for diagnosis and monitoring a patient’s response to treatment. Traditional Nyquist sampling is inadequate for high spatial resolution and whole-brain coverage in DCE-MRI; therefore, current methods involve undersampling and constrained reconstruction. However, this leads to difficult estimation of Pat-AIF from the undersampled data, and the resulting TK maps are based on population-averaged AFI and a single TK model, which may not be accurate.
Stage of Development
- Proof-of-concept demonstrated