# US 2019/0167203 A1 · Method and apparatus for characterizing an obstacle within an examination object using a medical image data set

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

- Number: US 2019/0167203 A1
- Document ID: US20190167203A1
- Status: Published application
- Kind: A1
- Application number: 16/207,355
- Filed: 2018-12-03
- Published: 2019-06-06
- Earliest claimed: 2017-12-04
- Assignees: Siemens Healthcare GmbH
- Applicants: Siemens Healthcare GmbH
- Inventors: Dominik Paul; Mario Zeller
- Web page: https://patentgrep.com/patent/US20190167203A1

## CPC

- A61B 5/7267: {involving training the classification device}
- A61B 5/055: involving electronic [EMR] or nuclear [NMR] magnetic resonance, e.g. magnetic resonance imaging
- 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 33/5608: {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)}
- G16H 50/70: for mining of medical data, e.g. analysing previous cases of other patients
- A61B: DIAGNOSIS; SURGERY; IDENTIFICATION (usefulness limited to only animals A61D)
- G01R: MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES (indicating correct tuning of resonant circuits H03J3/12)
- G16H: HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA

## Abstract

In a method and apparatus for characterizing an obstacle within an examination object using a medical image data set by the use of a database containing at least one data class, a trained artificial neural network defines and develops relationships between different obstacles, features and medical imaging data sets. A data entry of a data class is assigned by the neural network to the obstacle within the examination object and the obstacle within the examination object is characterized in an electronic output by this data entry.

## Claims

### Claim 1

A method for characterizing an obstacle within an examination object, comprising:
providing a computer with a database comprising a data class for a feature that characterizes a plurality of different obstacles, said data class comprising at least one data entry respectively for each of said different obstacles;
providing said computer with a medical image data set acquired from an examination object comprising an obstacle therein;
providing said computer with a trained artificial neural network that defines and develops relationships between said different obstacles, said feature, and contents of an input medical image data set;
in said computer, applying said trained artificial network to said acquired medical image data set by using the acquired medical image data set as said input medical image data set, and thereby determining a property of said obstacle in said examination object and assigning a data entry from said data class to said obstacle in the examination object, so that said obstacle within the examination object is characterized by the assigned data entry; and
providing an electronic signal as an output from said computer that represents the obstacle characterized by the assigned data entry.

### Claim 2 (depends on 1)

A method as claimed in claim 1 comprising selecting said feature from the group consisting of a material of said different obstacles, a shape of said different obstacles, a size of said different obstacles, a produce of said different obstacles, an effect of the respective different obstacles on a medical image data set with regard to at least one of a medical imaging apparatus and data acquisition technique used to generate the medical image data set, and a position or orientation or environment of the different obstacles within an examination object.

### Claim 3 (depends on 1)

A method as claimed in claim 1 comprising in said computer, updating said trained artificial neural network while determining said property.

### Claim 4 (depends on 1)

A method as claimed in claim 1 comprising, in said computer, selecting a method for acquiring a further medical image data set from the examination object, based on the determined property of the obstacle within the examination object.

### Claim 5 (depends on 4)

A method as claimed in claim 4 comprising selecting said method for acquiring said further medical image data set as a method that is best suited for acquiring said further medical image data set from the examination object in view of said property of said obstacle.

### Claim 6 (depends on 4)

A method as claimed in claim 4 comprising operating a medical data acquisition scanner in order to acquire said further medical imaging data set according to the selected method.

### Claim 7 (depends on 6)

A method as claimed in claim 6 comprising acquiring said further medical image data set with a same medical imaging scanner as was used to acquire said acquired medical image data set.

### Claim 8 (depends on 6)

A method as claimed in claim 6 comprising acquiring said further medical image data set using a magnetic resonance data acquisition scanner, and selecting said method for acquiring said further medical image data set using a magnetic resonance data acquisition technique selected from the group consisting of TSE, SEMAC, VAT, MAVRIC, STIR, SPAR, DIXON, and a technique using saturation pulses having different resonance frequencies.

### Claim 9 (depends on 1)

A method as claimed in claim 1 comprising providing said trained artificial neural network to said computer by creating said trained artificial neural network in a training computer selected from the group consisting of said computer and another, different computer, by:
providing said training computer with a training data class comprising data entries for a feature that characterizes different obstacles, said training data class comprising at least one training data entry respectively for each of said different obstacles;
providing said training computer with a plurality of training medical image data sets;
in said training computer, assigning data entries to the plurality of training medical image data sets;
training an artificial network in said training computer using an image content of the plurality of training medical image data sets and the training data entries assigned to the plurality of training medical image data sets so as to assign one of said training data entries to a medical image data set provided to the neural network as an input medical image data set, thereby producing said trained artificial neural network; and
providing the trained artificial neural network from said training computer to said computer.

### Claim 10

A computer for characterizing an obstacle within an examination object, said computer comprising:
a processor configured to access a database comprising a data class for a feature that characterizes a plurality of different obstacles, said data class comprising at least one data entry respectively for each of said different obstacles;
an input that receives into said processor a medical image data set acquired from an examination object comprising an obstacle therein;
a trained artificial neural network that defines and develops relationships between said different obstacles, said feature, and contents of an input medical image data set;
said processor configured to apply said trained artificial network to said acquired medical image data set by using the acquired medical image data set as said input medical image data set, and thereby determine a property of said obstacle in said examination object and assign a data entry from said data class to said obstacle in the examination object, so that said obstacle within the examination object is characterized by the assigned data entry; and
an output which said processor provides an electronic signal as an output that represents the obstacle characterized by the assigned data entry.

### Claim 11 (depends on 10)

A computer as claimed in claim 10 wherein said processor is configured to select a method for acquiring a further medical image data set from the examination object, based on the determined property of the obstacle within the examination object, and to emit a further electronic signal at said output that represents the selected method.

### Claim 12 (depends on 11)

A computer as claimed in claim 11 wherein said processor is configured to operate a medical data acquisition scanner in order to acquire said further medical imaging data set according to the selected method.

### Claim 13

A medical imaging apparatus comprising:
a medical image data acquisition scanner;
a computer provided with a database comprising a data class for a feature that characterizes a plurality of different obstacles, said data class comprising at least one data entry respectively for each of said different obstacles;
said computer being configured to operate the medical image data acquisition scanner so as to acquire a medical image data set from an examination object comprising an obstacle therein;
said computer comprising a trained artificial neural network that defines and develops relationships between said different obstacles, said feature, and contents of an input medical image data set;
said computer being configured to apply said trained artificial network to said acquired medical image data set by using the acquired medical image data set as said input medical image data set, and thereby determine a property of said obstacle in said examination object and assign a data entry from said data class to said obstacle in the examination object, so that said obstacle within the examination object is characterized by the assigned data entry;
said computer being configured to provide an electronic signal as an output from said computer that represents the obstacle characterized by the assigned data entry; and
said computer being configured to select, based on the determined property of the obstacle, a method for acquiring a further medical image data set from said examination object containing said obstacle, and to operate said medical image data acquisition scanner according to the selected method, in order to acquire said further medical image data set from said examination object containing said obstacle.

### Claim 14 (depends on 13)

A medical imaging apparatus as claimed in claim 13 wherein said medical image data acquisition scanner is a magnetic resonance data acquisition scanner.

### Claim 15

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:
access a database comprising a data class for a feature that characterizes a plurality of different obstacles, said data class comprising at least one data entry respectively for each of said different obstacles;
receive a medical image data set acquired from an examination object comprising an obstacle therein;
access a trained artificial neural network that defines and develops relationships between said different obstacles, said feature, and contents of an input medical image data set;
apply said trained artificial network to said acquired medical image data set by using the acquired medical image data set as said input medical image data set, and thereby determine a property of said obstacle in said examination object and assign a data entry from said data class to said obstacle in the examination object, so that said obstacle within the examination object is characterized by the assigned data entry; and
provide an electronic signal as an output from said computer that represents the obstacle characterized by the assigned data entry.

### Claim 16 (depends on 15)

A non-transitory, computer-readable data storage medium as claimed in claim 15 wherein said programming instructions cause said computer to select a method for acquiring a further medical image data set from the examination object, based on the determined property of the obstacle within the examination object.

### Claim 17 (depends on 16)

A non-transitory, computer-readable data storage medium as claimed in claim 16 wherein said programming instructions cause said computer to operate a medical data acquisition scanner in order to acquire said further medical imaging data set according to the selected method.