公开(公告)号：US20210181287A1

公开(公告)日：2021-06-17

申请号：US16759778

申请日：2018-10-22

Applicant: KONINKLIJKE PHILIPS N.V.

Inventor： KARSTEN SOMMER ,  TOM BROSCH ,  TIM PHILIPP HARDER ,  JOCHEN KEUPP ,  INGMAR GRAESSLIN ,  RAFAEL WIEMKER ,  AXEL SAALBACH

IPC: G01R33/565 ,  G01R33/56 ,  G06T5/00 ,  G06T7/00 ,  G06T5/50 ,  G06T5/20 ,  G06N3/08 ,  G06N3/04

Abstract: The invention relates to a magnetic resonance imaging data processing system (126) for processing motion artifacts in magnetic resonance imaging data sets using a deep learning network (146, 502, 702) trained for the processing of motion artifacts in magnetic resonance imaging data sets. The magnetic resonance imaging data processing system (126) comprises a memory (134, 136) storing machine executable instructions (161, 164) and the trained deep learning network (146, 502, 702). Furthermore, the magnetic resonance imaging data processing system (126) comprises a processor (130) for controlling the magnetic resonance imaging data processing system. Execution of the machine executable instructions (161, 164) causes the processor (130) to control the magnetic resonance imaging data processing system (126) to: receive a magnetic resonance imaging data set (144, 500, 800), apply the received magnetic resonance imaging data set (144, 500, 800) as an input to the trained deep learning network (146, 502, 702), process one or more motion artifacts present in the received magnetic resonance imaging data set (144, 500, 800) using the trained deep learning network (146, 502, 702).

公开(公告)号：US20200041594A1

公开(公告)日：2020-02-06

申请号：US16498555

申请日：2018-03-30

Applicant: KONINKLIJKE PHILIPS N.V.

Inventor： THOMAS ERIK AMTHOR ,  MARIYA IVANOVA DONEVA ,  KARSTEN SOMMER ,  PETER KOKEN

IPC: G01R33/483 ,  G01R33/48 ,  G01R33/24 ,  G01R33/44 ,  G01R33/56 ,  G01R33/565

Abstract: The invention provides for a magnetic resonance imaging (MRI) system (100) that comprises a memory (134) for storing machine executable instructions (140) and MRF pulse sequence commands (142). The MRF pulse sequence commands cause the MRI system to acquire MRF magnetic resonance data (144) according to a magnetic resonance (MR) fingerprinting protocol. The pulse sequence commands are configured for acquiring the MRF magnetic resonance data in two dimensional slices (410, 412, 414, 416, 418, 420), wherein the two dimensional slices have a slice selection direction, wherein the pulse sequence commands comprises a train of pulse sequence repetitions. The train of pulse sequence repetitions comprises a sampling event where the MRF magnetic resonance data is repeatedly sampled. The MRI system further comprises a processor for controlling the magnetic resonance imaging system. Execution of the machine executable instructions causes the processor to: acquire (200) the MRF magnetic resonance data by controlling the magnetic resonance imaging system with the MRF pulse sequence commands; and construct (202) a series (148) of at least one magnetic resonance parameter value for each voxel of the two dimensional slices using the MRF magnetic resonance data, wherein each of the series corresponds to the sampling event of each pulse sequence repetition; and calculate (204) a composition (502, 504, 506, 508) of each of a set of predetermined substances within two or more sub voxels (306, 308) for each voxel of the two dimensional slices using a sub-voxel magnetic resonance fingerprinting dictionary (150) for each of the two or more sub voxels and the series of the at least one magnetic resonance parameter value, wherein sub voxels divide each voxel in the slice selection direction.

公开(公告)号：US20190242961A1

公开(公告)日：2019-08-08

申请号：US16339405

申请日：2017-09-22

Applicant: KONINKLIJKE PHILIPS N.V.

Inventor： JAN JAKOB MEINEKE ,  THOMAS ERIK AMTHOR ,  PETER KOKEN ,  KARSTEN SOMMER

IPC: G01R33/44 ,  G01R33/24 ,  G01R33/56

CPC classification number: G01R33/443 ,  G01R33/243 ,  G01R33/246 ,  G01R33/50 ,  G01R33/5608

Abstract: The invention provides for a magnetic resonance imaging system (100). Machine executable instructions cause a processor controlling the MRI system to control (200) the magnetic resonance imaging system with the pulse sequence commands to acquire the magnetic resonance data. The pulse sequence commands are configured for controlling the magnetic resonance imaging system to acquire the magnetic resonance data according to a magnetic resonance fingerprinting protocol. The pulse sequence commands are configured for controlling the magnetic resonance imaging system to acquire the magnetic resonance data during multiple pulse repetitions (302). The pulse sequence commands are configured for controlling the magnetic resonance imaging system to cause gradient induced spin rephasing at least twice during each of the multiple pulse repetitions using a gradient magnetic field generating system (110, 112). The pulse sequence commands are configured for controlling the magnetic resonance imaging system to acquire at least two magnetic resonance signals during each of the multiple pulse repetitions. Each of the at least two magnetic resonance signals is measured during a separate one of the gradient induced spin rephasing. The magnetic resonance data comprises the at least two magnetic resonance signals acquired during each of the multiple pulse repetitions. Execution of the machine executable instructions further cause the processor to calculate (202) a B0-off-resonance map (158) using the magnetic resonance data, wherein the B0-off-resonance map is descriptive of a B0-off-resonance magnetic field of the magnetic resonance imaging system when the subject is within the imaging zone, wherein the B0-off-resonance map is at least partially calculated using at least two magnetic resonance signals measured during a single pulse repetition. Execution of the machine executable instructions further cause the processor to generate (204) at least one magnetic resonance parametric map by comparing the magnetic resonance data with a magnetic resonance fingerprinting dictionary (152).

公开(公告)号：US20220229134A1

公开(公告)日：2022-07-21

申请号：US17614595

申请日：2020-06-03

Applicant: KONINKLIJKE PHILIPS N.V.

Inventor： PETER BOERNERT ,  KARSTEN SOMMER ,  CHRISTOPHE MICHAEL JEAN SCHULKE ,  JOHAN SAMUEL VAN DEN BRINK

IPC: G01R33/48 ,  G01R33/561 ,  G01R33/565

Abstract: The invention relates to a method of MR imaging of an object (10) positioned in an examination volume of a MR device (1). It is an object of the invention to enable efficient and high-quality non-Cartesian MR imaging, even in situations of strong B0 inhomogeneity. In accordance with the invention, the method comprises: —subjecting the object to an imaging sequence comprising at least one RF excitation pulse and modulated magnetic field gradients, —acquiring MR signals along at least one non-Cartesian k-space trajectory, —reconstructing an MR image from the acquired MR signals, and —detecting one or more mal-sampling artefacts caused inhomogeneity induced insufficient k-space sampling in the MR image using a deep learning network. Moreover, the invention relates to a MR device (1) and to a computer program.

公开(公告)号：US20210312659A1

公开(公告)日：2021-10-07

申请号：US17263985

申请日：2019-12-12

Applicant: KONINKLIJKE PHILIPS N.V.

Inventor： KARSTEN SOMMER ,  SASCHA KRUEGER ,  JULIAN SENEGAS

IPC: G06T7/73 ,  G01T1/29 ,  G01R33/56 ,  A61B5/00

Abstract: The invention provides for a medical instrument (100, 400) comprising a medical imaging system (102, 402) configured for acquiring medical imaging data (432) from a subject (108); a subject support (110) configured for supporting the subject during acquisition of the medical imaging data; and an optical imaging system (114, 114′) configured for acquiring optical imaging data (134) of the subject on the subject support. The execution of the machine executable instructions causes a processor (122) controlling the medical instrument to: control (200) the optical imaging system to acquire the optical imaging data; generate (202) the initial vector (136) using the optical imaging data; generate (204) the synthetic image by inputting the initial vector into a generator neural network; calculate (206) a difference (140) between the synthetic image and the optical imaging data; and provide (208) a warning signal (142) if the difference differs by a predetermined threshold. The generator neural network is trained to generate a synthetic image (138) of the subject on the subject support in response to inputting an initial vector.

公开(公告)号：US20190285711A1

公开(公告)日：2019-09-19

申请号：US16463922

申请日：2016-11-23

Applicant: KONINKLIJKE PHILIPS N.V.

Inventor： JOCHEN KEUPP ,  JAN JAKOB MEINEKE ,  KARSTEN SOMMER

IPC: G01R33/54 ,  G01R33/58 ,  G01R33/56 ,  G01R33/50

Abstract: The invention provides for a magnetic resonance imaging system (100) comprising a memory (134) for storing machine executable instructions (140) and pulse sequence commands (142). The pulse sequence commands are configured for controlling the magnetic resonance imaging system according to a DCE Magnetic Resonance Imaging protocol. The magnetic resonance imaging system further comprises a user interface (200) and a processor (130) for controlling the magnetic resonance imaging system. Execution of the machine executable instructions causes the processor to: control (500) the magnetic resonance imaging system using the pulse sequence commands to acquire calibration magnetic resonance data (144) two or more times for varying flip angles; reconstruct (502) each acquisition of the calibration magnetic resonance data into a calibration image (146) to create a set of variable flip angle images (148); calculate (504) a T1 mapping (150) using the set of variable flip angle images; calculate (506) a contrast agent calibration (152) for a predetermined magnetic resonance imaging contrast agent using at least partially the T1 mapping; calculate (508) an estimated calibration error (154) that is descriptive of an estimated error in the contrast agent calibration and/or the T1 mapping using a calibration accuracy model, wherein the calibration accuracy model is configured for calculating the estimated calibration error using the set of variable flip angle images; and display (510) a calibration warning message (202) on the user interface if the estimated calibration error is outside of a predetermined calibration error range.

公开(公告)号：US20210177296A1

公开(公告)日：2021-06-17

申请号：US16759755

申请日：2018-10-26

Applicant: KONINKLIJKE PHILIPS N.V.

Inventor： AXEL SAALBACH ,  STEFFEN WEISS ,  KARSTEN SOMMER ,  CHRISTOPHE SCHUELKE ,  MICHAEL HELLE

IPC: A61B5/055 ,  A61B5/00 ,  G16H30/40

Abstract: The invention relates to a magnetic resonance imaging system, the magnetic resonance imaging system (100) comprising: —a memory (134, 136) storing machine executable instructions (160, 162, 164), pulse sequence commands (140) and a first machine learning model (146) comprising a first deep learning network (502), wherein the pulse sequence commands (140) are configured for controlling the magnetic resonance imaging system (100) to acquire a set of magnetic resonance imaging data, wherein the first machine learning model (146) comprises a first input and a first output, —a processor, wherein an execution of the machine executable instructions (160, 162, 164) causes the processor (130) to control the magnetic resonance imaging system (100) to repeatedly perform an acquisition and analysis process comprising: —acquiring a dataset (142.1, . . . , 142.N) comprising a subset of the set of magnetic resonance imaging data from an imaging zone (108) of the magnetic resonance imaging system (100) according to the pulse sequence commands (140), —providing the dataset (142.1, . . . , 142.N) to the first input of the first machine learning model (146), I-n response of the providing, receiving a prediction (148, 502) of a motion artifact level of the acquired magnetic resonance imaging data from the first output of the first machine learning model (146), the motion artifact level characterizing a number and/or extent of motion artifacts present in the acquired magnetic resonance imaging data.20

公开(公告)号：US20200320705A1

公开(公告)日：2020-10-08

申请号：US16955959

申请日：2018-12-14

Applicant: KONINKLIJKE PHILIPS N.V.

Inventor： RAFAEL WIEMKER ,  TANJA NORDHOFF ,  THOMAS BUELOW ,  AXEL SAALBACH ,  TOBIAS KLINDER ,  TOM BROSCH ,  TIM PHILIPP HARDER ,  KARSTEN SOMMER

IPC: G06T7/00 ,  G06T7/11 ,  G06T5/00 ,  A61N5/10

Abstract: The invention relates to a system for assessing a pulmonary image which allows for an improved assessment with respect to lung nodules detectability. The pulmonary image is smoothed for providing different pulmonary images (20, 21, 22) with different degrees of smoothing, wherein signal values and noise values, which are indicative of the lung vessel detectability and the noise in these images, are determined and used for determining an image quality being indicative of the usability of the pulmonary image to be assessed for detecting lung nodules. Since a pulmonary image shows lung vessels with many different vessel sizes and with many different image values, which cover the respective ranges of potential lung nodules generally very well, the image quality determination based on the different pulmonary images with different degrees of smoothing allows for a reliable assessment of the pulmonary image's usability for detecting lung nodules. The image quality is used to determine a radiation dose level to be applied for generating a next pulmonary image.

公开(公告)号：US20200096589A1

公开(公告)日：2020-03-26

申请号：US16468849

申请日：2017-12-06

Applicant: KONINKLIJKE PHILIPS N.V.

Inventor： KARSTEN SOMMER ,  MARIYA IVANOVA DONEVA ,  THOMAS ERIK AMTHOR ,  PETER KOKEN ,  JAN JAKOB MEINEKE

IPC: G01R33/561 ,  G01R33/50 ,  G01R33/56 ,  G01R33/54 ,  G06T11/00 ,  G06T7/00 ,  G06K9/00 ,  G16H30/20

Abstract: The invention provides for a magnetic resonance imaging system (100) for acquiring magnetic resonance data (142) from a subject (118) within a measurement zone (108). The magnetic resonance imaging system (100) comprises: a processor (130) for controlling the magnetic resonance imaging system (100) and a memory (136) storing machine executable instructions (150, 152, 154), pulse sequence commands (140) and a dictionary (144). The pulse sequence commands (140) are configured for controlling the magnetic resonance imaging system (100) to acquire the magnetic resonance data (142) of multiple steady state free precession (SSFP) states per repetition time. The pulse sequence commands (140) are further configured for controlling the magnetic resonance imaging system (100) to acquire the magnetic resonance data (142) of the multiple steady state free precession (SSFP) states according to a magnetic resonance fingerprinting protocol. The dictionary (144) comprises a plurality of tissue parameter sets. Each tissue parameter set is assigned with signal evolution data pre-calculated for multiple SSFP states.

公开(公告)号：US20250031994A1

公开(公告)日：2025-01-30

申请号：US18710650

申请日：2022-10-27

Applicant: KONINKLIJKE PHILIPS N.V.

Inventor： KARSTEN SOMMER ,  SASCHA KRUEGER ,  JULIEN THOMAS SENEGAS ,  PETER KOKEN

IPC: A61B5/107 ,  A61B5/00 ,  G06T7/00 ,  G06T7/62 ,  G06T7/70

Abstract: A computer-implemented method is provided for estimating a weight of a patient when supported on a patient table. Optical image data and depth image data are obtained and patient body keypoints are extracted from the optical image data. A first frame is selected which comprises an image of the patient table, by selecting a frame in which no body keypoints are present in the optical image data. A second frame is selected of the patient on the table, by selecting a frame with patient body keypoints and with little or no movement. A patient volume is obtained based on a difference between the depth image data for the first and second frame (or depth data at those times) and the patient weight is estimated from the determined patient volume.