# US 12,111,373 B2 · Determination of a further processing location in magnetic resonance imaging

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 12,111,373 B2
- Document ID: US12111373B2
- Status: Granted
- Kind: B2
- Application number: 17/294,698
- Filed: 2019-11-19
- Published: 2022-01-13
- Granted: 2024-10-08
- Earliest claimed: 2018-11-20
- Assignees: Koninklijke Philips N.V.
- Applicants: Koninklijke Philips N.V.
- Inventors: Karsten Sommer; Michael Gunter Helle
- Web page: https://patentgrep.com/patent/US12111373B2

## CPC

- G01R 33/543: {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)}
- G01R 33/546: {Interface between the MR system and the user, e.g. for controlling the operation of the MR system or for the design of pulse sequences}
- 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)}
- G01R 33/56308: {Characterization of motion or flow; Dynamic imaging}
- G01R 33/56366: {Perfusion imaging}
- G06N 3/045: Combinations of networks
- G06N 3/0464: Convolutional networks [CNN, ConvNet]
- G06N 3/084: Backpropagation, e.g. using gradient descent
- G06N 3/09: Supervised learning
- G06T 7/0012: {Biomedical image inspection}
- G06V 10/764: using classification, e.g. of video objects
- G06V 10/82: using neural networks
- G06T 2207/10096: Dynamic contrast-enhanced magnetic resonance imaging [DCE-MRI]
- G06T 2207/20081: Training; Learning
- G06T 2207/20084: Artificial neural networks [ANN]
- G06T 2207/20101: Interactive definition of point of interest, landmark or seed
- G06T 2207/30016: Brain
- G06V 2201/03: Recognition of patterns in medical or anatomical images
- G01R: MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES (indicating correct tuning of resonant circuits H03J3/12)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING

## Abstract

The invention provides for a method of training a neural network (322) configured for providing a further processing location (326). The method comprises providing (200) a labeled medical image (100), wherein the labeled medical image comprises multiple labels each indicating a truth processing location (102, 104, 106). The method further comprises inputting (202) the labeled medical image into the neural network to obtain one trial processing location. The one trial processing location comprises a most likely trial processing location (108). The method further comprises determine (204) the closest truth processing location (106) for the most likely trial processing location. The method further comprises calculating (206) an error vector (110) using the closest truth processing location and the most likely trial processing location. The method further comprises training (208) the neural network using the error vector.

## Claims

### Claim 1

A method of training a neural network configured for providing a further processing location, the method comprising:
providing a labeled medical image, wherein the labeled medical image comprises a plurality of labels indicating a plurality of truth processing locations, respectively;
inputting the labeled medical image into the neural network to obtain a most likely trial processing location as an output of the neural network;
determining a closest truth processing location of the plurality of truth processing locations for the most likely trial processing location, wherein the closest truth processing location is a closest of the plurality of truth processing locations to the output of the neural network;
calculating an error vector using the closest truth processing location and the most likely trial processing location, wherein the error vector is a position change between the closest truth processing location and the most likely trial processing location, wherein the error vector is calculated only using the closest truth processing location and the most likely trial processing location; and
training the neural network using the error vector.

### Claim 2 (depends on 1)

The method of claim 1, wherein the method is performed repeatedly using multiple additional labeled medical images.

### Claim 3

A medical imaging system comprising:
a memory for storing machine executable instructions and a neural network; and
a processor for controlling the medical imaging system, wherein execution of the machine executable instructions causes the processor to:
receive a labeled medical image for training the neural network, wherein the labeled medical image comprises a plurality of labels indicating a plurality of truth processing locations, respectively;
input the labeled medical image into the neural network to obtain a most likely trial processing location as an output of the neural network;
determine a closest truth processing location of the plurality of truth processing locations for the most likely trial processing location, wherein the closest truth processing location is a closest of the plurality of truth processing locations to the output of the neural network;
calculate an error vector using the closest truth processing location and the most likely trial processing location, wherein the error vector is a position change between the closest truth processing location and the most likely trial processing location, wherein the error vector is calculated only using the closest truth processing location and the most likely trial processing location; and
train the neural network using the error vector.

### Claim 4 (depends on 3)

The medical imaging system of claim 3, wherein execution of the machine executable instructions further causes the processor to:
receive a medical image; and
input the medical image into the trained neural network to obtain a further processing location from the trained neural network.

### Claim 5 (depends on 4)

The medical imaging system of claim 4, wherein the medical image is a diffusion-weighted magnetic resonance image of a brain, wherein the further processing region is a seed point and/or a region of interest size selection, wherein execution of the machine executable instructions further causes the processor to calculate a brain fiber tracking image by inputting the further processing region and the medical image into a brain fiber tracking algorithm.

### Claim 6 (depends on 4)

The medical imaging system of claim 4, wherein execution of the machine executable instructions further causes the processor to:
display the medical image on a user interface;
receive a plurality of addition labels indicating a plurality of further truth processing locations, respectively, to provide a further labeled medical image; and
further train the neural network by:
inputting the further labeled medical image into the neural network to obtain at least one further trial processing location, wherein the at least one further trial processing location comprises a further most likely trial processing location;
determining a further closest truth processing location for the further most likely trial processing location;
calculating a further error vector using the further closest truth processing location and the further most likely trial processing location; and
training the neural network using the further error vector.

### Claim 7 (depends on 4)

The medical imaging system of claim 4, wherein the medical imaging system further comprises a magnetic resonance imaging system, wherein the memory further comprises pulse sequence commands configured for controlling the magnetic resonance imaging system to acquire magnetic resonance imaging data according to a magnetic resonance imaging protocol, wherein execution of the machine executable instructions further causes the processor to:
modify the pulse sequence commands using the further processing location;
control the magnetic resonance imaging system with the modified pulse sequence commands to acquire the magnetic resonance imaging data; and
reconstruct a magnetic resonance image using the magnetic resonance imaging data.

### Claim 8 (depends on 7)

The medical imaging system of claim 7, wherein the memory further comprises initial pulse sequence commands configured for controlling the magnetic resonance imaging system to acquire initial magnetic resonance imaging data, wherein execution of the machine executable instructions further causes the processor to:
control the magnetic resonance imaging system to acquire the initial magnetic resonance imaging data using the initial pulse sequence commands; and
reconstruct the medical image using the initial magnetic resonance imaging data.

### Claim 9 (depends on 7)

The medical imaging system of claim 7, wherein the magnetic resonance imaging protocol is an arterial spin labelling protocol, wherein the further processing region is a labelling location.

### Claim 10 (depends on 7)

The medical imaging system of claim 7, wherein the magnetic resonance imaging protocol is a Dynamic Contrast Enhanced magnetic resonance imaging protocol or a Dynamic Susceptibility Contrast perfusion magnetic resonance imaging protocol, wherein the further processing region is a measurement location for an arterial input function.

### Claim 11 (depends on 7)

The medical imaging system of claim 7, wherein the magnetic resonance imaging protocol is single voxel magnetic resonance spectroscopy protocol, wherein the further processing region is a single voxel location for performing the single voxel magnetic resonance spectroscopy protocol.

### Claim 12 (depends on 7)

The medical imaging system of claim 7, wherein the magnetic resonance imaging protocol is a phase-contrast magnetic resonance imaging protocol, wherein the further processing region is a slice position for performing the phase-contrast magnetic resonance imaging protocol.

### Claim 13 (depends on 7)

The medical imaging system of claim 7, wherein the magnetic resonance imaging protocol is a contrast enhanced magnetic resonance angiography protocol, wherein the further processing region is a temporal region for triggering acquisition of the magnetic resonance imaging data.

### Claim 14

A non-transitory computer readable medium storing machine executable instructions for execution by a processor controlling a medical imaging system, wherein execution of the machine executable instructions causes the processor to:
receive a labeled medical image for training a neural network, wherein the labeled medical image comprises a plurality of labels indicating a plurality of truth processing locations, respectively;
input the labeled medical image into the neural network to obtain a most likely trial processing location as an output of the neural network;
determine the closest truth processing location of the plurality of truth processing locations for the most likely trial processing location, wherein the closest truth processing location is the closest of the truth processing locations to the output of the neural network;
calculate an error vector using the closest truth processing location and the most likely trial processing location, wherein the error vector is a position change between the closest truth processing location and the most likely trial processing location, wherein the error vector is calculated only using the closest truth processing location and the most likely trial processing location; and
train the neural network using the error vector.

### Claim 15 (depends on 14)

The non-transitory computer readable medium of claim 14, wherein execution of the machine executable instructions further causes the processor to:
receive a medical image; and
input the medical image into the trained neural network to obtain a further processing location from the trained neural network.

### Claim 16 (depends on 15)

The non-transitory computer readable medium of claim 15 further storing pulse sequence commands for controlling a magnetic resonance imaging system to acquire magnetic resonance imaging data according to a magnetic resonance imaging protocol, wherein execution of the machine executable instructions further causes the processor to:
modify the pulse sequence commands using the further processing location;
control the magnetic resonance imaging system with the modified pulse sequence commands to acquire the magnetic resonance imaging data; and
reconstruct a magnetic resonance image using the magnetic resonance imaging data.

### Claim 17 (depends on 16)

The non-transitory computer readable medium of claim 16 further storing initial pulse sequence commands for controlling the magnetic resonance imaging system to acquire initial magnetic resonance imaging data, wherein execution of the machine executable instructions further causes the processor to:
control the magnetic resonance imaging system to acquire the initial magnetic resonance imaging data using the initial pulse sequence commands; and
reconstruct the medical image using the initial magnetic resonance imaging data.

### Claim 18 (depends on 16)

The non-transitory computer readable medium of claim 16, wherein the magnetic resonance imaging protocol is an arterial spin labelling protocol, wherein the further processing region is a labelling location.

### Claim 19 (depends on 16)

The non-transitory computer readable medium of claim 16, wherein the magnetic resonance imaging protocol is a Dynamic Contrast Enhanced magnetic resonance imaging protocol or a Dynamic Susceptibility Contrast perfusion magnetic resonance imaging protocol, wherein the further processing region is a measurement location for an arterial input function.

### Claim 20 (depends on 16)

The non-transitory computer readable medium of claim 16, wherein the magnetic resonance imaging protocol is a single voxel magnetic resonance spectroscopy protocol, wherein the further processing region is a single voxel location for performing the single voxel magnetic resonance spectroscopy protocol.