Plane selection using localizer images
The present disclosure relates to use of a workflow for automatic prescription of different radiological imaging scan planes across different anatomies and modalities. The automated prescription of such imaging scan planes helps ensure contiguous visualization of the different landmark structures. Unlike prior approaches, the disclosed technique determines the necessary planes using the localizer images itself and does not explicitly segment or delineate the landmark structures to perform plane prescription.
1. A method for imaging an anatomic region of a patient, comprising:
acquiring a plurality of two-dimensional (2D) three-plane localizer images of the patient using a magnetic resonance imaging (MRI) system;
providing the plurality of 2D three-plane localizer images to a localizer network trained to select a subset of the 2D three-plane localizer images for detection of an anatomy-of-interest based on the image contents of the plurality of 2D three-plane localizer images, wherein the localizer network is trained to select a respective 2D three-plane localizer image having the maximal coverage of the anatomy-of-interest;
processing the subset of 2D three-plane localizer images using a scan plane network trained to determine one or more image scan planes or image scan plane parameters for the anatomy-of-interest, wherein the localizer network and the scan plane networks are separate from each other;
generating one or more diagnostic images of the anatomic region of the patient using the MRI system based on the one or more image scan planes or image scan plane parameters; and
wherein the anatomy-of-interest is not segmented prior to determining the one or more image scan planes or image scan plane parameters.
2. The method of claim 1 , wherein the plurality of 2D three-plane localizer images are acquired as part of a pre-acquisition step prior to acquisition of one or more diagnostic images.
3. The method of claim 1 , wherein one or both of the localizer network or scan plane network are trained using pairs of 2D three-plane localizer images and corresponding diagnostic images acquired based on the plurality of 2D three-plane localizer images, wherein the diagnostic images include data specifying an image scan plane prescription with respect to the associated 2D three-plane localizer image.
4. The method of claim 1 , wherein the subset of 2D three-plane localizer images, prior to processing by the scan plane network, are processed by a coverage network trained to identify an imaging field-of-view associated with the anatomy-of-interest.
5. The method of claim 4 , wherein the coverage network generates a binary coverage mask as part of identifying the imaging field-of-view.
6. The method of claim 1 , wherein the scan plane network determines the one or more image scan planes or image scan plane parameters by fitting an analytic plane to a plane mask encompassing the anatomy-of-interest in the subset of 2D three-plane localizer images generated from the plurality of 2D three-plane localizer images.
7. The method of claim 1 , wherein the plurality of 2D three-plane localizer images include 2D axial plane localize images, 2D sagittal plane localizer images, and 2D coronal plane localizer images.
8. A magnetic resonance imaging (MRI) system comprising:
a memory encoding processor-executable routines for determining one or more imaging scan planes;
a processing component configured to access the memory and execute the processor-executable routines, wherein the routines, when executed by the processing component, cause the processing component to:
acquire a plurality of two-dimensional (2D) three-plane localizer images of a patient;
process the plurality of 2D three-plane localizer images using a localizer network trained to select a subset of the 2D three-plane localizer images for detection of an anatomy-of-interest based on the image contents of the plurality of 2D three-plane localizer images, wherein the localizer network is trained to select a respective 2D three-plane localizer image having the maximal coverage of the anatomy-of-interest;
process the subset of the 2D three-plane localizer images using a scan plane network trained to determine one or more image scan planes or image scan plane parameters that contain regions of the anatomy-of-interest, wherein the localizer network and the scan plane networks are separate from each other;
generate one or more diagnostic images of an anatomic region of the patient using the one or more image scan planes or image scan plane parameters; and
wherein the anatomy-of-interest is not segmented prior to determining the one or more image scan planes or image scan plane parameters.
9. The MRI system of claim 8 , wherein the subset of 2D three-plane localizer images, prior to processing by the scan plane network, are processed by a coverage network trained to identify an imaging field-of-view associated with the anatomy-of-interest or a related anatomic structure.
10. The MRI system of claim 8 , wherein the plurality of 2D three-plane localizer images are acquired as part of a pre-acquisition step prior to acquisition of one or more diagnostic images.
11. The MRI system of claim 8 , wherein the scan plane network determines the one or more image scan planes or image scan plane parameters by fitting an analytic plane to a plane mask encompassing the anatomy-of-interest in the subset of 2D three-plane localizer images generated from the plurality of 2D three-plane localizer images.