IP Library › Granted Patent US 10,664,683
Granted Patent B2
US 10,664,683 · App. 15/875,086 · Granted May 26, 2020

Method and apparatus for classifying magnetic resonance measurement data acquired from an examination object using a magnetic resonance fingerprinting method

Inventors: Thorsten Feiweier (Poxdorf, DE); Stefan Huwer (Erlangen, DE); Mathias Nittka (Baiersdorf, DE)
Assignee: Siemens Healthcare GmbH
G06K9/00114A61B5/055A61B5/7267G01R33/5608G06K9/00087G16H50/20A61B2576/00G01R33/24G01R33/4818G01R33/50G01R33/561
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 10,664,683
App. No.
15/875,086
Granted
May 26, 2020
Kind
B2
Abstract

In a method and computer and magnetic resonance (MR) apparatus for classifying MR measurement data acquired from an object under examination by execution of an MR fingerprinting method, wherein the MR measurement data include multiple MR signal profiles acquired by the MR fingerprinting method. At least one texture parameter is derived from the MR measurement data. The MR measurement data are classified into at least one tissue class using the at least one texture parameter. The classified MR measurement data are provided as an output.

Claims (31)

1. A method for classifying magnetic resonance (MR) measurement data acquired from an object by execution of an MR fingerprinting method, said method for classifying comprising:

with a computer, controlling an MR data acquisition scanner to execute an MR fingerprint method to acquire MR measurement data of an object, said MR measurement data comprising a plurality of MR signal profiles acquired by execution of said MR fingerprinting method and a plurality of spatially resolved tissue parameter maps reconstructed from said plurality of MR signal profiles;

in said computer, deriving a plurality of texture parameter maps from the plurality of spatially resolved tissue parameter maps to derive at least one texture parameter from said MR measurement data;

in said computer, classifying the MR measurement data into at least one tissue class using said at least one texture parameter and the plurality of texture parameter maps; and

in said computer, generating an electronic signal corresponding to the classified MR measurement data and providing the electronic signal as an output from said computer.

2. A method as claimed in claim 1 comprising, in said computer:

deriving said at least one texture parameter by implementing a reconstruction algorithm in order to reconstruct at least one texture parameter map of the plurality of texture parameter maps, with said plurality of MR signal profiles being directly introduced into said reconstruction algorithm; and

classifying said MR measurement data using said at least one texture parameter map.

3. A method as claimed in claim 2 comprising, in said reconstruction algorithm, using an MR fingerprinting model in which a spatial context of said plurality of MR signal profiles is modeled.

4. A method as claimed in claim 1 comprising, in said computer, deriving said at least one texture parameter with a time-resolution over a time interval during which said plurality of MR signal profiles were acquired.

5. A method as claimed in claim 4 comprising classifying said MR measurement data from said at least one time-resolved texture parameter using a database comprising information representing a temporal evolution of said at least one texture parameter over said time interval, as a function of at least one tissue class.

6. A method as claimed in claim 4 comprising using at least one filter operation to classify said MR measurement data from said at least one time-resolved texture parameter.

7. A method as claimed in claim 1 comprising deriving said at least one texture parameter in the spatial domain in which a length scale is determined based on at least one element selected from the group consisting of an anatomical region from which said MR measurement data were acquired, a clinical problem underlying the acquisition of said MR measurement data, a disease process that is suspected in said examination object, an image data previously acquired from said examination object.

8. A classifying computer comprising:

a processor configured to control an MR data acquisition scanner to execute an MR fingerprint method to acquire MR measurement data of an object;

an input interface configured to receive said MR measurement data of said object from said MR data acquisition scanner and into said processor, said MR measurement data comprising a plurality of MR signal profiles acquired by execution of said MR fingerprinting method and a plurality of spatially resolved tissue parameter maps reconstructed from said plurality of MR signal profiles;

said processor being configured to derive a plurality of texture parameter maps from the plurality of spatially resolved tissue parameter maps to derive at least one texture parameter from said MR measurement data;

said processor being configured to classify the MR measurement data into at least one tissue class using said at least one texture parameter and the plurality of texture parameter maps; and

an output interface at which said processor is configured to provide the classified MR measurement data in electronic form as an output from said processor.

9. A magnetic resonance (MR) apparatus comprising:

an MR data acquisition scanner;

a control computer configured to control the MR data acquisition scanner to execute an MR fingerprinting method to acquire MR measurement data from an examination object, so that the MR measurement data comprises a plurality of MR signal profiles acquired with said MR fingerprinting method and a plurality of spatially resolved tissue parameter maps reconstructed from said plurality of MR signal profiles;

said control computer being configured to derive a plurality of texture parameter maps from the plurality of spatially resolved tissue parameter maps to derive at least one texture parameter from said MR measurement data;

said control computer being configured to classify the MR measurement data into at least one tissue class using said at least one texture parameter and the plurality of texture parameter maps; and

said control computer being configured to provide the classified MR measurement data in electronic form as an output from said computer.

10. A non-transitory, computer-readable data storage medium encoded with programming instructions, said storage medium being loaded into a computer and said programming instructions causing said computer to:

control an MR data acquisition scanner to execute an MR fingerprint method to acquire MR measurement data of an object;

receive said MR measurement data of said object, said MR measurement data comprising a plurality of MR signal profiles acquired by execution of said MR fingerprinting method and a plurality of spatially resolved tissue parameter maps reconstructed from said plurality of MR signal profiles;

derive a plurality of texture parameter maps from the plurality of spatially resolved tissue parameter maps to derive at least one texture parameter from said MR measurement data;

classify the MR measurement data into at least one tissue class using said at least one texture parameter and the plurality of texture parameter maps; and

provide the classified MR measurement data in electronic form as an output from said computer.

Assignments (2)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Dec 20, 2023
From: SIEMENS HEALTHCARE GMBH
To: SIEMENS HEALTHINEERS AG
Reel/Frame 066267/0346 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded May 10, 2018
From: FEIWEIER, THORSTEN, MR.; HUWER, STEFAN, MR.; NITTKA, MATHIAS, MR.
To: SIEMENS HEALTHCARE GMBH
Reel/Frame 045769/0612 →
Priority Claims (1)
EP 17152234 · Jan 19, 2017 · regional
Continuity (1)
Related Publication 20180204045A1 · Jul 19, 2018