Inverse regression in MR Fingerprinting: reducing dictionary size while increasing parameters accuracy
Résumé
Purpose:
To reduce dictionary size and increase parameter estimate accuracy in MR Fingerprinting (MRF).
Methods:
A dictionary-based learning (DBL) method is investigated to bypass inherent MRF limitations in high dimension: reconstruction time and memory requirement. The DBL method is a 3-step procedure: (1) a quasi-random sampling strategy to produce the dictionary, (2) a statistical inverse regression model to learn from the dictionary a probabilistic mapping between MR fingerprints and parameters, and (3) this mapping to provide both parameter estimates and their confidence levels.
Results:
On synthetic data, experiments show that the quasi-random sampling outperforms the grid when designing the dictionary for inverse regression.
Dictionaries up to 100 times smaller than usually employed in MRF yield more accurate parameter estimates with a 500 time gain.
Estimates are supplied with a confidence index, well correlated with the estimation bias (r~$\ge$~0.89). On microvascular MRI data, results show that dictionary-based methods (MRF and DBL) yield more accurate estimates than the conventional, closed-form equation, method.On MRI signals from tumor bearing rats, the DBL method shows very little sensitivity to the dictionary size in contrast to the MRF method.
Conclusion:
The proposed method efficiently reduces the number of required simulations to produce the dictionary, speeds up parameter estimation, and improve estimates accuracy. The DBL method also introduces a confidence index for each parameter estimate.
Domaines
Statistiques [math.ST]Origine | Fichiers produits par l'(les) auteur(s) |
---|
Loading...