Autonomous magnetic resonance scanning for a given medical test

Patent metadata

Field	Value
Publication number	US-12102423-B2
Application number	US-202217807218-A
Country	US
Kind code	B2
Filing date	Jun 16, 2022
Priority date	Feb 10, 2022
Publication date	Oct 1, 2024
Grant date	Oct 1, 2024

Abstract

For autonomous MR scanning for a given medical test, a simplified MR scanner may be used without or will little input or control by a technologist (e.g., by a physician, radiologist, or person trained in MR scanner operation). The MR scanner autonomously positions, scans, checks quality, analyzes, and/or outputs an answer to a diagnostic question with or without an MR image. Scan analysis, based on artificial intelligence, allows for on-going or on-the-fly alteration of the scanning configuration to acquire the data desired to answer the diagnostic question. By using a simplified MR scanner, both position of the patient relative to the MR scanner and localization of the scan by the MR scanner are jointly solved. Sensors may sense a patient in a scan position where the reduced radio frequency requirements allow for a more open bore.

First claim

What is claimed is:

1. A method of data analytics for magnetic resonance (MR) scanning, the method comprising:
MR scanning a patient using a first configuration of an MR scanner based on a medical test during a MR examination and resulting in first raw data;
analyzing, by a first machine-learned model, the first raw data, the analyzing resulting in a change of the first configuration;
controlling the MR scanner with a second configuration based on the change from the first configuration;
MR scanning the patient using the second configuration of the MR scanner for the medical test as part of the same MR examination and resulting in second raw data; and
generating a diagnostic output of the medical test of the MR examination from the first and second raw data.

2. The method of claim 1 wherein MR scanning using the first and second configurations comprises MR scanning with a non-uniform main magnetic field, non-homogeneous first pulses, and/or non-linear gradients.

3. The method of claim 1 wherein analyzing comprises determining a value for a diagnosis for the medical test by the first machine-learned model and an uncertainty for the value, and wherein controlling comprises altering from the first configuration to the second configuration where the uncertainty is above a threshold.

4. The method of claim 3 wherein controlling further comprises proscribing next measurements based on back-propagation of the first machine-learned model and setting parameters of the second configuration for the next measurements based on missing information identified by the backpropagation.

5. The method of claim 1 wherein analyzing comprises determining that insufficient information has been gathered by the MR scanning using the first configuration and wherein controlling comprises determining the second configuration to gain the information.

6. The method of claim 1 wherein analyzing comprises inputting the first raw data and the first configuration to the first machine-learned model and outputting by the first machine-learned model the diagnostic output as an answer to a diagnostic question.

7. The method of claim 1 further comprising positioning a patient relative to the MR scanner and localizing the MR scanning by a second machine-learned model solving for both the positioning and the localizing.

8. The method of claim 1 further comprising checking quality of the first raw data by a second machine-learned model, wherein controlling comprises controlling to increase the quality without input from a human.

9. The method of claim 1 wherein controlling comprises controlling by a second machine-learned model, the second machine-learned model comprising a reinforcement learned model.

10. The method of claim 9 wherein MR scanning using the first configuration continues until the reinforcement learned model generates the change.

11. A magnetic resonance (MR) system comprising:
a MR scanner configured by settings of controls to scan a region of a patient, the scan providing scan data;
a patient support for the patient, the patient support moveable relative to the MR scanner;
a sensor configured to sense the patient on the patient support;
a processor configured to jointly (1) position the patient by movement of the patient support and/or the MR scanner and (2) localize the scan of the region by the MR scanner, wherein the processor comprises a machine-learned detector configured to detect the position and localization from input of an output of the sensor and the scan data and comprises a machine-learned actor configured to move the patient support and change the localization when the detected position and localization is incorrect.

12. The MR system of claim 11 wherein the processor comprises a machine-learned model configured to jointly position and localize as a single solution for both the position and the localization.

13. The MR system of claim 11 wherein the processor is further configured to analyze the scan data once the patient is positioned and the scan localized, the analysis by a machine-learned model configured to output a diagnosis in response to input of the scan data, and further configured to alter the settings based on the analysis.

14. The MR system of claim 13 wherein the processor is further configured to monitor artifacts of the scan data and alter the settings based on artifact level.

15. The MR system of claim 11 wherein the sensor comprises a camera and the MR scanner comprises an open bore MR scanner where a field of view of the camera extends to the patient while localized within an open bore of the open bore MR scanner.

16. A method for autonomous magnetic resonance (MR) scanning, the method comprising:
positioning a patient in an MR scanner;
localizing a region of the patient by the MR scanner for scanning the region;
MR scanning the patient by the MR scanner, the MR scanning resulting in scan data;
analyzing the scan data by checking a quality of the scan data by a machine-learned model; and
generating a diagnostic answer from the scan data;
wherein positioning, localizing, MR scanning, analyzing, and generating are performed without human control.

17. The method of claim 16 further comprising altering, by a processor, a configuration of the MR scanner based on the quality.

18. The method of claim 17 wherein checking comprises checking by the machine-learned model comprising a machine-learned generator of quality used to identify deviation from the quality regardless of a type of artifact.

19. The method of claim 17 wherein checking comprises checking for a type of artifact, and wherein altering comprises altering the configuration to reduce the type of artifact.

Assignees

• Siemens Healthineers Ag

Inventors

• Mailhe Boris
• Comaniciu Dorin
• Kamen Ali
• Nadar Mariappan S
• Lou Bin
• Greiser Andreas
• Chebrolu Venkata Veerendranadh

Classifications

• 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

• A61B5/7264

  Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems · CPC title

• A61B5/7221

  Determining signal validity, reliability or quality (preventing, reducing or removing noise induced by motion artefacts A61B5/7207; noise originating from a therapeutic or surgical apparatus A61B5/7217) · CPC title

• A61B5/704

  Tables · CPC title

• A61B5/0077

  Devices for viewing the surface of the body, e.g. camera, magnifying lens · CPC title