Magnetic resonance imaging apparatus for generating parametric map from magnetic resonance signal data and method of operating the same
AssigneesSamsung Electronics Co., Ltd., Seoul National University R&Db Foundation
InventorsYoungbeom KIM, Doohee Lee, Jongho Lee, Junki Lee, Sangyoung Zho
Life of this patent
Estimating the term…
Abstract
Provided are a magnetic resonance imaging (MRI) apparatus and method of operating the same, whereby a characteristic parameter value may be acquired from MR signal data via training using an artificial neural network (ANN) and a parametric map may be generated based on the acquired characteristic parameter value.
Claims20
A method of generating a parametric map based on magnetic resonance (MR) signal data, the method comprising:
- acquiring a trained network model by performing training in which MR signal data generated using a magnetic resonance imaging (MRI) method and an MR signal model is applied to an artificial neural network as input and a first characteristic parameter value calculated via the MR signal model is applied to the artificial neural network as groundtruth;
- acquiring MR image data by scanning an object;
- applying the MR image data to the trained network model;
- and
- calculating, using the trained network model and based on the MR image data, a second characteristic parameter value with respect to the MR image data.
A magnetic resonance imaging (MRI) apparatus for generating a parametric map based on magnetic resonance (MR) signal data, the apparatus comprising:
- a radio frequency coil configured to acquire, from at least one radio frequency channel coil, an MR signal emitted from an object;
- a memory storing instructions for processing the MR signal received from the radio frequency coil;
- and
- a processor configured to execute the stored instructions to:
- generate a trained network model by performing training in which MR signal data generated using an MRI method and an MR signal model is applied to an artificial neural network as input and a first characteristic parameter value calculated via the MR signal model is applied to the artificial neural network as groundtruth;
- acquire MR image data by scanning an object;
- apply the MR image data to the trained network model;
- and
- calculate, using the trained network model and based on the MR image data, a second characteristic parameter value with respect to the MR image data.
A computer program product comprising a non-transitory computer-readable storage medium, wherein the non-transitory computer-readable storage medium comprises instructions for performing a method of operating a magnetic resonance imaging (MRI) apparatus, the method comprising:
- acquiring a trained network model by performing training in which magnetic resonance (MR) signal data generated using an MRI method and an MR signal model is applied to an artificial neural network as input and a first characteristic parameter value calculated via the MR signal model is applied to the artificial neural network as groundtruth;
- acquiring MR image data by scanning an object;
- applying the MR image data to the trained network model;
- and
- calculating, using the trained network model and based on the MR image data, a second characteristic parameter value with respect to the MR image data.
Dependencies are inferred from the claim text.