Articulated structure registration in magnetic resonance images of the brain

• US 10,049,448 B2
• Filed: 01/06/2015
• Issued: 08/14/2018
• Est. Priority Date: 01/06/2014
• Status: Active Grant

• Current Assignee: Koninklijke Philips N.V.
• Original Assignee: Koninklijke Philips N.V.
• Inventors: Santiago Flores, Gerardo, Soldea, Octavian, Jasinschi, Radu Serban
• Primary Examiner(s): Osifade, Idowu O
• Application Number:
  US15/109,469
• Publication Number:
  US 20160328847A1
• Time in Patent Office: 1,316 Days
• Field of Search: None
• CPC Class Codes: A61B 2576/026 for the brain
• CPC Class Codes: A61B 5/0042 for the brain
• CPC Class Codes: A61B 5/055 involving electronic [EMR] or nuclear [NMR] magnetic resonance, e.g. magnetic resonance imaging
• CPC Class Codes: A61B 5/4064 Evaluating the brain A61B5/031, A61B5/369, A61B5/14553 take precedence
• CPC Class Codes: G06F 18/21355 nonlinear criteria, e.g. embedding a manifold in a Euclidean space
• CPC Class Codes: G06F 18/22 Matching criteria, e.g. proximity measures
• CPC Class Codes: G06T 2207/10088 Magnetic resonance imaging [MRI]
• CPC Class Codes: G06T 2207/20016 Hierarchical, coarse-to-fine, multiscale or multiresolution image processing; Pyramid transform
• CPC Class Codes: G06T 2207/30016 Brain
• CPC Class Codes: G06T 3/60 Rotation of whole images or parts thereof
• CPC Class Codes: G06T 7/0012 Biomedical image inspection
• CPC Class Codes: G06T 7/33 using feature-based methods
• CPC Class Codes: G06T 7/37 using transform domain methods

Abstract

A registration processor (74) is configured to obtain articulated brain substructures using acquired brain image data and template brain image data. The registration processor (74) annotates the brain image data; registers the brain image data with template image data using global brain registration; and registers at least one brain structure of the brain image data a corresponding brain structure of the template image data using a local brain substructure registration. The registration processor (74) articulates articulated substructures of the registered brain structures to improve registration using articulated substructure registration.

Claims

1. A brain registration system, comprising:

a main magnet configured to create a strong, static magnetic field through an imaging volume;

magnetic field gradient coils configured to generate magnetic field gradients in the imaging volume;

one or more RF coils configured to transmit radio frequency pulses into the imaging volume and receive magnetic resonance echo signals from the imaging volume;

a transmitter configured to control the one or more RF coils to transmit magnetic resonance excitation and manipulation pulses into the imaging volume;

one more receivers configured to demodulate the magnetic resonance echo signals from the one or more RF coils;

a back end system including a memory configured to store brain templates, a user input device, a display device, and one or more processors configured to;

reconstruct the demodulated magnetic resonance echo signals from the receiver into a brain image,

segment the brain image to annotate brain image data,

globally register the brain image with a template image retrieved from the storage memory,

identify an articulated structure in the segmented brain image, the articulated structure including a pair of sub-structures connected by a linking joint,

rotate the articulated sub-structures relative to each other about the linking joint to improve registration to maximize image similarity of the articulated structure in the brain image and the brain template, and

control the display device to display the brain image including the articulated structure.

2. The system according to claim 1, wherein the one or more processors are further configured to:

iteratively rotate the at least one of the sub-structures of the articulated structure in the brain image and the brain template about the linking joint,

compute a similarity metric between the brain image articulated structure and the brain template articulated structure for each iteration, and

select one of the iterations that maximizes the similarity metric.

3. The system according to claim 1, wherein the one or more processors are further configured to apply a transform to the brain image to maximize overlap between the articulated structure in the brain image and the brain template.

4. The system according to claim 1, wherein the one or more processors are further configured to identify the linking joint by:

calculating a Euclidian distance for all pairs of pixels/voxels between the sub-structures of the articulated structure;

selecting a pair of pixels/voxels having a minimum Euclidian distance therebetween; and

computing a middle point between the selected pair of pixels/voxels as the linking joint between the sub-structures.

5. The system according to claim 1, wherein the globally registering of the brain image with the brain template includes an isotropic moments-based global registration.

6. A brain registration method, comprising:

with a main magnet, creating a strong, static magnetic field through an imaging volume;

with magnetic field gradient coils, generating magnetic field gradients in the imaging volume;

with one or more RF coils, transmitting radio frequency pulses into the imaging volume and receiving magnetic resonance echo signals from the imaging volume;

with a transmitter, controlling the one or more RF coils to transmit magnetic resonance excitation and manipulation pulses into the imaging volume;

with one more receivers, demodulating the magnetic resonance echo signals from the one or more RF coils;

with one or more processors of a back end system including a memory configured to store brain templates, a user input device, a display device, and the one or more processors;

reconstructing the demodulated magnetic resonance echo signals from the receiver into a brain image,

segmenting the brain image to annotate brain image data,

globally registering the brain image with a template image retrieved from the storage memory,

identifying an articulated structure in the segmented brain image, the articular structure including a pair of sub-structures connected by a linking joint,

rotating the articulated sub-structures relative to each other about the linking joint to improve registration to maximize image similarity of the articulated structure in the brain image and the brain template, and

controlling the display device to display the brain image including the articulated structure.

7. The method according to claim 6, further including with the one or more processors:

iteratively rotating the at least one of the sub-structures of the articulated structure in the brain image and the brain template about the linking joint; and

computing a similarity metric between the brain image articulated structure and the template image articulated structure for each iteration; and

selecting the iteration that maximizes the similarity metric.

8. The method according to claim 6, wherein identifying the linking joint includes with the one or more processors:

calculating a Euclidian distance for all pairs of pixels/voxels between the articulated sub-structures of the articulated structure;

selecting a pair of pixels/voxels having a minimum Euclidian distance from each other; and

computing a middle point between the selected pair of pixels/voxels as the linking joint between the sub-structures.

9. A non-transitory computer readable medium having instructions configured to control the one or more processors to perform the method of claim 6.

10. A brain registration system comprising:

a registration processor having a processor configured to;

annotate brain image data, thereby obtaining a segmentation of brain structures in the brain image data;

register the brain image data with template image data using global brain registration, thereby obtaining registered brain image data comprising registered brain structures;

compensate for incorrect registration of at least one structure of the registered brain structure, which structure includes a pair of sub-structures, said compensation comprising;

identifying a linking joint between a pair of sub-structures of the registered brain structures;

rotating at least one of the pair of sub-structures in the registered brain image data about the linking joint to maximize a similarity to the template image data in accordance with a similarity metric which is computed between the registered brain image data comprising said rotated sub-structure and the template image data.

11. The system according to claim 10, further including:

a reconstruction processor configured to reconstruct brain image data into brain images; and

a display device configured to display the brain images including the articulated structure with the sub-structures rotated to maximize the similarity.