SYSTEMS AND METHODS FOR AUTOMATED DETECTION IN MAGNETIC RESONANCE IMAGES

• US 20190033414A1
• Filed: 08/29/2018
• Published: 01/31/2019
• Est. Priority Date: 11/22/2016
• Status: Active Grant

• Current Assignee: Hyperfine Operations Incorporated
• Original Assignee: Hyperfine Research, Inc.
• Inventors: Sofka, Michal, Rothberg, Jonathan M., Charvat, Gregory L., Ralston, Tyler S.
• Granted Patent: US 10,534,058 B2
• Time in Patent Office: Days
• 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: A61B 5/7267 involving training the classification device
• CPC Class Codes: G01R 33/3806 Open magnet assemblies for improved access to the sample, e.g. C-type or U-type magnets
• CPC Class Codes: G01R 33/383 using permanent magnets
• CPC Class Codes: G01R 33/445 MR involving a non-standard magnetic field B0, e.g. of low magnitude as in the earth's magnetic field or in nanoTesla spectroscopy, comprising a polarizing magnetic field for pre-polarisation, B0 with a temporal variation of its magnitude or direction such as field cycling of B0 or rotation of the direction of B0, or spatially inhomogeneous B0 like in fringe-field MR or in stray-field imaging
• CPC Class Codes: G01R 33/4806 Functional imaging of brain activation
• CPC Class Codes: G01R 33/483 with selection of signals or spectra from particular regions of the volume, e.g. in vivo spectroscopy
• CPC Class Codes: G01R 33/5608 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 Class Codes: G06F 18/2411 based on the proximity to a decision surface, e.g. support vector machines
• CPC Class Codes: G06F 18/24143 Distances to neighbourhood prototypes, e.g. restricted Coulomb energy networks [RCEN]
• CPC Class Codes: G06N 3/044 Recurrent networks, e.g. Hopfield networks
• CPC Class Codes: G06N 3/045 Combinations of networks
• CPC Class Codes: G06T 2207/10088 Magnetic resonance imaging [MRI]
• CPC Class Codes: G06T 2207/20081 Training; Learning
• CPC Class Codes: G06T 2207/20084 Artificial neural networks [ANN]
• CPC Class Codes: G06T 2207/20092 Interactive image processing based on input by user
• CPC Class Codes: G06T 2207/20212 Image combination
• CPC Class Codes: G06T 2207/30016 : Brain
• CPC Class Codes: G06T 2207/30101 : Blood vessel; Artery; Vein; Vascular
• CPC Class Codes: G06T 2207/30172 : Centreline of tubular or elongated structure
• CPC Class Codes: G06T 7/0012 : Biomedical image inspection
• CPC Class Codes: G06T 7/0016 : involving temporal comparison
• CPC Class Codes: G06V 10/764 : using classification, e.g. of video objects
• CPC Class Codes: G06V 10/82 : using neural networks
• CPC Class Codes: G06V 2201/031 : of internal organs
• CPC Class Codes: G16H 30/40 : for processing medical images, e.g. editing
• CPC Class Codes: G16H 50/70 : for mining of medical data, e.g. analysing previous cases of other patients

Abstract

Some aspects include a method of determining change in size of an abnormality in a brain of a patient positioned within a low-field magnetic resonance imaging (MRI) device. The method comprises, while the patient remains positioned within the low-field MRI device, acquiring first and second magnetic resonance (MR) image data of the patient'"'"'s brain; providing the first and second MR image data as input to a trained statistical classifier to obtain corresponding first and second output; identifying, using the first output, at least one initial value of at least one feature indicative of a size of the abnormality; identifying, using the second output, at least one updated value of the at least one feature; determining the change in the size of the abnormality using the at least one initial value of the at least one feature and the at least one updated value of the at least one feature.

Claims

1. A method of determining change in size of an abnormality in a brain of a patient positioned within a low-field magnetic resonance imaging (MRI) device, the method comprising:

while the patient remains positioned within the low-field MRI device;

acquiring first magnetic resonance (MR) image data of a patient'"'"'s brain;

providing the first MR image data as input to a trained statistical classifier to obtain corresponding first output;

identifying, using the first output, at least one initial value of at least one feature indicative of a size of an abnormality in the patient'"'"'s brain;

acquiring second MR image data of the patient'"'"'s brain subsequent to acquiring the first MR image data;

providing the second MR image data as input to the trained statistical classifier to obtain corresponding second output;

identifying, using the second output, at least one updated value of the at least one feature indicative of the size of the abnormality in the patient'"'"'s brain;

determining the change in the size of the abnormality using the at least one initial value of the at least one feature and the at least one updated value of the at least one feature.

2. The method of claim 1, wherein identifying, using the first output, the at least one initial value of the at least one feature indicative of the size of the abnormality in the patient'"'"'s brain comprises:

identifying a region in the MR image data including the abnormality.

3. The method of claim 1, wherein identifying, using the first output, the at least one initial value of the at least one feature indicative of the size of the abnormality in the patient'"'"'s brain comprises:

identifying one or more first values indicative of a first diameter of the abnormality.

4. The method of claim 3, wherein the identifying further comprises identifying one or more second values indicative of a second diameter of the abnormality orthogonal to the first diameter.

5. The method of claim 1, wherein determining the change in the size of the abnormality using the at least one initial value of the at least one feature and the at least one updated value of the at least one feature comprises:

determining an initial size of the abnormality using the at least one value of the at least one feature;

determining an updated size of the abnormality using the at least one updated value of the at least one feature; and

determining the change in the size of the abnormality using the determined initial and updated sizes of the abnormality.

6. The method of claim 1, wherein the trained statistical classifier comprises a multi-layer neural network.

7. The method of claim 1, wherein the trained statistical classifier comprises a convolutional neural network.

8. The method of claim 1, wherein the trained statistical classifier comprises a fully convolutional neural network.

9. The method of claim 1, wherein the second MR image data is obtained within one hour of the first MR image data.

10. The method of claim 1, further comprising repeating acquiring MR image data to obtain a sequence of frames of MR image data.

11. The method of claim 1, wherein the sequence of frames is acquired over a period of time greater than an hour while the patient remains positioned within the low-field magnetic resonance imaging device.

12. The method of claim 11, wherein the sequence of frames is acquired over a period of time greater than two hours while the patient remains positioned within the low-field magnetic resonance imaging device.

13. The method of claim 11, wherein the sequence of frames is acquired over a period of time greater than five hours while the patient remains positioned within the low-field magnetic resonance imaging device.

14. The method of claim 1, wherein the abnormality comprises a hemorrhage.

15. The method of claim 1, wherein the abnormality comprises a hemorrhage, a lesion, an edema, a stroke core, a stroke penumbra, and/or swelling.

16. A low-field magnetic resonance imaging (MRI) device configured to determine change in size of an abnormality in a brain of a patient, the low-field MRI device comprising:

a plurality of magnetic components, including;

a B0 magnet configured to produce, at least in part, a B0 magnetic field;

at least one gradient magnet configured to spatially encode magnetic resonance data; and

at least one radio frequency coil configured to stimulate a magnetic resonance response and detect magnetic components configured to, when operated, acquire magnetic resonance image data; and

at least one controller configured to operate the plurality of magnet components to, while the patient remains positioned within the low-field magnetic resonance device, acquire first magnetic resonance (MR) image data of the patient'"'"'s brain, and acquire second MR image data of the patient'"'"'s brain subsequent to acquiring the first MR image data,

wherein the at least one controller further configured to perform;

providing the first and second MR image data as input to a trained statistical classifier to obtain corresponding first output and second output;

identifying, using the first output, at least one initial value of at least one feature indicative of a size of an abnormality in the patient'"'"'s brain;

acquiring second MR image data for the portion of the patient'"'"'s brain subsequent to acquiring the first MR image data;

identifying, using the second output, at least one updated value of the at least one feature indicative of the size of the abnormality in the patient'"'"'s brain;

determining the change in the size of the abnormality using the at least one initial value of the at least one feature and the at least one updated value of the at least one feature.

17. At least one non-transitory computer-readable storage medium storing processor-executable instructions that, when executed by at least one computer hardware processor, cause the at least one computer hardware processor, to perform method of determining change in size of an abnormality in a brain of a patient positioned within a low-field magnetic resonance imaging (MRI) device, the method comprising:

while the patient remains positioned within the low-field MRI device;

acquiring first magnetic resonance (MR) image data of the patient'"'"'s brain;

providing the first MR image data as input to a trained statistical classifier to obtain corresponding first output;

identifying, using the first output, at least one initial value of at least one feature indicative of a size of an abnormality in the patient'"'"'s brain;

acquiring second MR image data of the patient'"'"'s brain subsequent to acquiring the first MR image data;

providing the second MR image data as input to the trained statistical classifier to obtain corresponding second output;

identifying, using the second output, at least one updated value of the at least one feature indicative of the size of the abnormality in the patient'"'"'s brain;

determining the change in the size of the abnormality using the at least one initial value of the at least one feature and the at least one updated value of the at least one feature.

18. A system, comprising:

at least one computer hardware processor;

at least one non-transitory computer-readable storage medium storing processor-executable instructions that, when executed by the at least one computer hardware processor, cause the at least one computer hardware processor, to perform method of determining change in size of an abnormality in a brain of a patient positioned within a low-field magnetic resonance imaging (MRI) device, the method comprising;

while the patient remains positioned within the low-field MRI device;

acquiring first magnetic resonance (MR) image data of the patient'"'"'s brain;

providing the first MR image data as input to a trained statistical classifier to obtain corresponding first output;

identifying, using the first output, at least one initial value of at least one feature indicative of a size of an abnormality in the patient'"'"'s brain;

acquiring second MR image data of the patient'"'"'s brain subsequent to acquiring the first MR image data;

providing the second MR image data as input to the trained statistical classifier to obtain corresponding second output;

identifying, using the second output, at least one updated value of the at least one feature indicative of the size of the abnormality in the patient'"'"'s brain; and

determining the change in the size of the abnormality using the at least one initial value of the at least one feature and the at least one updated value of the at least one feature.