Magnetic resonance fingerprinting (MRF) serial artificial neural network (ANN) sequence design

Patent metadata

Field	Value
Publication number	US-10614359-B2
Application number	US-201514682220-A
Country	US
Kind code	B2
Filing date	Apr 9, 2015
Priority date	Apr 19, 2014
Publication date	Apr 7, 2020
Grant date	Apr 7, 2020

Abstract

Example apparatus and methods employ an artificial neural network (ANN) to automatically design magnetic resonance (MR) pulse sequences. The ANN is trained using transverse magnetization signal evolutions having arbitrary initial magnetizations. The trained up ANN may then produce an array of signal evolutions associated with a pulse sequence having user selectable pulse sequence parameters that vary in degrees of freedom associated with magnetic resonance fingerprinting (MRF). Efficient and accurate approaches are provided for predicting user controllable MR pulse sequence settings including, but not limited to, acquisition period and flip angle (FA). The acquisition period and FA may be different in different sequence blocks in the pulse sequence produced by the ANN. Predicting user controllable MR pulse sequence settings for both conventional MR and MRF facilitates achieving desired signal characteristics from a signal evolution produced in response to an automatically generated pulse sequence.

First claim

What is claimed is:

1. A method for training an artificial neural network (ANN) for use in automatically producing a magnetic resonance fingerprinting (MRF) pulse sequence, the method comprising:
accessing, by a computer system, an initial signal associated with a known pulse sequence;
accessing, by the computer system, a set of initial magnetization values associated with the initial signal;
accessing, by the computer system, a set of target magnetization values;
controlling, by the computer system, the ANN to produce a set of flip angle (FA) values for the MRF pulse sequence and a set of acquisition period (AP) values for the MRF pulse sequence based, at least in part, on the set of initial magnetization values and the set of target magnetization values;
selecting, by the computer system, an FA value from the set of FA values;
selecting, by the computer system, an AP value from the set of AP values;
updating, by the computer system, the ANN based, at least in part, on the FA value or the AP value;
controlling, by the computer system, a Bloch simulator to update the set of initial magnetization values based, at least in part, on the FA value or the AP value;
providing, by the computer system, a set of production inputs to the ANN; and
controlling, by the computer system, the ANN to produce a new pulse sequence based on the set of production inputs, where the new pulse sequence varies echo time and flip angle between at least two sequence blocks.

2. The method of claim 1 , where the known pulse sequence is a TrueFISP signal.

3. The method of claim 1 , where the set of initial magnetization values include arbitrary initial magnetizations.

4. The method of claim 1 , where the known pulse sequence is described by:

S
⁢

⁢
E

=

∑

s
=
1

N
s

⁢

⁢

∏

i
=
1

N
A

⁢

⁢

∑

j
=
1

N

R
⁢

⁢
F

⁢

⁢

R
i

⁡

(
α
)

⁢

R

R
⁢

⁢

F
ij

⁡

(

α
,
ϕ

)

⁢

R
⁡

(
G
)

⁢

E
i

⁡

(

T
⁢

⁢
1

,

T
⁢

⁢
2

,
D

)

⁢

M
0

or

S
⁢

⁢
E

=

∑

s
=
1

N
s

⁢

⁢

∏

i
=
1

N
A

⁢

⁢

∏

j
=
1

N

R
⁢

⁢

Assignees

• Univ Case Western Reserve

Inventors

• Griswold Mark
• Geshnizjani Nahal
• Ma Dan
• Loparo Kenneth
• Sunshine Jeffrey

Classifications

• G01R33/448

  Relaxometry, i.e. quantification of relaxation times or spin density (G01R33/50 takes precedence) · CPC title

• G01R33/5608

  Data processing and visualization specially adapted for MR, e.g. for feature analysis and pattern recognition on the basis of measured MR data, segmentation of measured MR data, edge contour detection on the basis of measured MR data, for enhancing measured MR data in terms of signal-to-noise ratio by means of noise filtering or apodization, for enhancing measured MR data in terms of resolution by means for deblurring, windowing, zero filling, or generation of gray-scaled images, colour-coded images or images displaying vectors instead of pixels (image data processing or generation, in general G06T) · CPC title

• G01R33/543

  Control of the operation of the MR system, e.g. setting of acquisition parameters prior to or during MR data acquisition, dynamic shimming, use of one or more scout images for scan plane prescription (G01R33/546 takes precedence) · CPC title

• G06N3/08 Primary

  Learning methods · CPC title

• G01R33/56341

  Diffusion imaging · CPC title