Magnetic resonance imaging of breast microcalcifications

Abstract

A medical imaging method is disclosed herein. The method includes receiving (202) gradient echo magnetic resonance imaging data (123) describing a breast region (319) of a subject (318). The method also includes receiving (202) DIXON magnetic resonance imaging data describing the breast region. The DIXON magnetic resonance imaging data and the gradient echo magnetic resonance imaging data are spatially matched. The method also includes generating (204) a breast microcalcification magnetic resonance image (128) by inputting the gradient echo magnetic resonance imaging data and the DIXON magnetic resonance imaging data to a breast microcalcification image reconstruction module (122). The breast microcalcification image reconstruction module is configured to output a breast microcalcification magnetic resonance image in response to inputting the gradient echo magnetic resonance imaging data and the DIXON magnetic resonance imaging data. The breast microcalcification magnetic resonance image describes a location of breast microcalcification, wherein the gradient echo magnetic resonance imaging data is a phase image.

Description

Technical Field

[0001] This invention relates to magnetic resonance imaging, and more particularly to magnetic resonance imaging of breast tissue. Background Technology

[0002] Magnetic resonance imaging (MRI) scanners use a large static magnetic field to align the nuclear spins of atoms as part of the process for generating images of a patient's body. This large static magnetic field is called the B0 field or main magnetic field. MRI enables the spatial measurement of various quantities or properties of an object.

[0003] US Patent Publication US1,0401,459B2 discloses a method for acquiring magnetic resonance imaging that accurately depicts vascular calcifications or other objects composed of magnetically displaced material within a subject. Images are acquired generally using pulse sequences designed to reduce artifacts caused by physiological motion and mitigate chemical shift artifacts from the water-fat boundary. Advantageously, the MRI technique described herein suppresses chemical shift artifacts without significantly reducing signal intensity from adipose tissue, thereby allowing for more reliable visualization of vascular calcifications.

[0004] Baheze et al. disclosed a method for detecting calcium deposits in “Detection of microcalcifications by characteristic magnetic susceptibility effects using MR phase image cross-correlation analysis” (Medical Physics, February 25, 2015, Vol. 42, No. 3, pp. 1436-1452), which uses the characteristic magnetization susceptibility effect on high-field magnetic resonance images to detect breast microcalcifications. Summary of the Invention

[0005] This invention provides a medical system, a computer program, and a method.

[0006] Mammograms are typically used to detect breast microcalcifications. However, this exposes the subject to X-rays. This embodiment provides a method for imaging breast microcalcifications without using ionizing radiation. For this purpose, two types of magnetic resonance imaging data are used. The first is gradient echo magnetic resonance imaging (MRI) data, and the second is DIXON MRI data. In DIXON MRI, measurements are taken in multiple phases, and the image components can be separated into a fat image imaging the fat or lipid content of voxels and a water image imaging non-fatty tissue. The DIXON MRI data and gradient echo MRI data are input into a breast microcalcification image reconstruction module, which outputs a breast microcalcification MRI image displaying the microcalcifications. The breast microcalcification MRI image can be further filtered using a high-pass spatial filter and used to generate a projection image. The generated projection image is comparable to a conventional mammogram used for detecting breast microcalcifications.

[0007] In one aspect, the present invention provides a medical system including a memory storing machine-executable instructions and a breast microcalcification image reconstruction module. The breast microcalcification image reconstruction module is configured to output a breast microcalcification magnetic resonance image in response to input gradient echo magnetic resonance imaging (MRI) data and DIXON magnetic resonance imaging (MRI) data. The DIXON MRI data may include both fat images and water images. The breast microcalcification phase MRI image describes the location of the breast microcalcifications. The gradient echo MRI data is a phase image. The medical system also includes a computing system configured to control the medical system.

[0008] The execution of machine-executable instructions causes the computing system to receive gradient echo magnetic resonance imaging (MRI) data describing a breast region of an object. The breast region of the object can be, for example, within a region of interest. The execution of the machine-executable instructions also causes the processor to receive DIXON MRI data describing the breast region. The DIXON MRI data and the gradient echo MRI data are spatially matched. Spatial matching means that there may be a one-to-one correspondence between the voxels of the gradient echo MRI data and the Dixon MRI image, or there may be a registration that provides a mapping between the voxels of the two.

[0009] In some examples, registration can be used to transform either gradient echo MRI data or DIXON MRI data so that they are matched to each other or spatially. However, in most cases, gradient echo MRI data and DIXON MRI data will be acquired so that they are spatially matched through acquisition. In some cases, the acquisition of gradient echo MRI data and DIXON MRI data can be combined, thereby automatically matching them spatially.

[0010] The execution of machine-executable instructions also enables the computing system to generate breast microcalcification MRI images by inputting gradient echo MRI data and DIXON MRI data into the breast microcalcification image reconstruction module. This embodiment can be advantageous because it provides a way to identify breast microcalcifications without using an X-ray machine.

[0011] Gradient echo MRI data can be acquired in several different ways. In some cases, it can be single-echo wavelength TE gradient echo MRI data. In other cases, it can be acquired using a multi-echo gradient echo MRI protocol. In this case, it can span several echo times. In the case of multi-echo gradients, DIXON MRI data and gradient echo MRI data can be acquired simultaneously.

[0012] In another embodiment, gradient echo MRI data is so-called long TE or echo time gradient echo MRI data. The TE or echo time value can be used to control how small calcifications can be detected. The term long TE also relates to DIXON MRI acquisitions using relatively short TE or echo times. One specification for defining long TE would be to make it equivalent to or less than the T2* of adipose tissue for a specific MRI protocol. T2* depends on the magnetic field strength and therefore will depend on the specific MRI system.

[0013] The phase of gradient echo MRI data is inherently selectable. However, there are practical choices that can simplify manipulation or image reconstruction. For example, gradient echo MRI data can be selected to be in phase or out of phase with the fat signal. Being in phase with fat can be advantageous because the image is more homogeneous and numerically easier to process. However, knowledge of DIXON MRI data allows for efficient mapping.

[0014] In another embodiment, the magnetic resonance image of breast microcalcifications is a phase image.

[0015] In another embodiment, the DIXON magnetic resonance imaging data is a water image.

[0016] In another embodiment, the execution of machine-executable instructions also enables the computing system to compute a filtered breast microcalcification MRI image by applying a high-pass spatial filter to the breast microcalcification MRI image. This embodiment can be advantageous because larger structures in the breast microcalcification MRI image can be removed. One way to implement the high-pass spatial filter is to compute an image that has already been low-pass filtered and has essentially blended out features with high spatial frequencies, and then remove it from the original image.

[0017] In another embodiment, the execution of machine-executable instructions also enables the computing system to generate a projected image of breast microcalcifications by calculating a projection image of the filtered breast microcalcification image. This embodiment may be advantageous because the breast microcalcification image can be, for example, a three-dimensional dataset or a stack of two-dimensional slices. The generation of the projected image can make the data easier to understand for physicians accustomed to examining X-rays. The projected image can be calculated, for example, in different ways. The projected image can be calculated for the projection of the minimum or maximum value in a specific voxel.

[0018] In another embodiment, the breast microcalcification image reconstruction module is implemented as an algorithm that includes the step of calculating fibroglandular tissue segmentation from DIXON MRI image data. This step involves identifying the location of fibroglandular tissue by inputting the DIXON MRI data into a fibroglandular tissue identification module. The DIXON MRI data may include water images and fat images. Since the breast is primarily composed of adipose or fat tissue, the water image will mainly display fibroglandular tissue.

[0019] Therefore, water images can be easily segmented using algorithms. The algorithm also includes a step of calculating a phase image of the fibroglandular tissue by inputting the segmented fibroglandular tissue into a phase image calculation module. The phase image calculation module takes the location of the tissue as input, and then, knowing the intensity of the B0 field, can calculate the possible perturbations of the field by the tissue. Several different methods are known for this. So-called kernel computation and trained neural networks can be used.

[0020] The algorithm also includes a step of calculating an adjusted phase image by subtracting the fibroglandular tissue phase image from the gradient echo MRI data. This step essentially eliminates the influence of fibroglandular tissue from the gradient echo MRI data. Finally, the algorithm includes a step of providing a breast microcalcification MRI image by performing phase unrolling of the adjusted phase image. For example, the phase in a normal phase image varies between - and +pi. When the phase transitions between one of these boundaries, a sudden change occurs in the image. However, automatically performing phase unrolling and avoiding this is relatively straightforward.

[0021] In another embodiment, the fibroglandular tissue identification module is configured to provide fibroglandular tissue segmentation by first thresholding the image and then performing region growing, undergoing processes (not necessarily in the order mentioned) such as erosion, dilation, or smoothing to provide smooth geometry for the segmentation.

[0022] In another embodiment, the fibroglandular tissue recognition module is configured to provide fibroglandular tissue segmentation by inputting DIXON MRI data into a trained fibroglandular tissue recognition neural network. For example, the DIXON MRI data may include images of water, and in some cases fat, which are input into the trained fibroglandular tissue recognition neural network, and then output the results. This can be trained, for example, by generating a specialized MRI phantom containing calcium particles placed in known locations within an MRI breast phantom.

[0023] In another embodiment, the phase image calculation module is implemented using a dipole kernel calculation module. This can be accomplished using known methods for calculating dipole kernels.

[0024] In another embodiment, the phase image calculation module is implemented as a trained phase image generating neural network. In this example, segmentation is input into the trained phase image generating neural network and outputs an adjusted phase image. Training data for the trained phase image generating neural network can be generated in several different ways. In one case, known tissue arrangements can be measured to calculate or provide training data. In other examples, a dipole kernel can be used to train the neural network.

[0025] In another embodiment, the breast microcalcification image reconstruction module is implemented as a trained breast microcalcification image reconstruction neural network. In this example, the entire step of reconstructing the breast microcalcification MRI image is performed by a single neural network. This neural network can be trained by providing a training dataset. Each training dataset includes gradient echo MRI data, DIXON MRI data, and control breast microcalcification MRI data. The breast microcalcification MRI image can be generated, for example, by performing measurements on a phantom. In other examples, the breast microcalcification MRI image can be generated by an algorithm through analysis of gradient echo MRI data and DIXON MRI data.

[0026] In another embodiment, the execution of machine-executable instructions also enables the computing system to receive k-space data acquired according to the DIXON MRI data protocol and the gradient echo MRI protocol with a single or multiple gradient echoes. In some examples, k-space data can be combined, i.e., DIXON MRI data and gradient echo MRI data can be combined. Therefore, the data can be acquired in a single acquisition, which ensures their spatial matching. In other examples, the k-space data can have two parts: one is DIXON MRI data, and the other is data acquired according to the gradient echo MRI protocol.

[0027] The execution of machine-executable instructions also enables the computing system to reconstruct gradient echo magnetic resonance imaging (MRI) data according to the gradient echo MRI protocol. The execution of machine-executable instructions further enables the processor to reconstruct DIXON MRI data according to the DIXON MRI protocol.

[0028] In another embodiment, the DIXON magnetic resonance imaging protocol is a compressed sensing protocol.

[0029] In another embodiment, the gradient echo magnetic resonance imaging protocol is a compressed sensing protocol.

[0030] In another embodiment, the DIXON MRI protocol is a multi-point DIXON MRI protocol. The use of a multi-point DIXON MRI protocol can be beneficial because the resulting water and fat images can be more accurate, which can lead to more accurate identification of the location of breast microcalcifications.

[0031] In another embodiment, the medical system also includes a magnetic resonance imaging (MRI) system configured to acquire k-space data from an imaging region. The memory also contains pulse sequence commands configured to acquire the measured k-space data according to the DIXON MRI protocol and the gradient echo MRI protocol. Execution of the machine-executable instructions further enables the computing system to control the MRI system to acquire k-space data using the pulse sequence commands. As previously described, if the DIXON MRI protocol and the gradient echo MRI protocol are combined, this can be a single k-space data acquisition. In other examples, if the protocols are executed individually, it can be two separate k-space data acquisitions.

[0032] In another embodiment, the pulse sequence command is configured to acquire k-space data for the region of interest. The pulse sequence command is also configured to suppress vascular structures by implementing at least one saturation band outside the region of interest. The use of a saturation band can be useful because it reduces or removes the MR signal from blood diffusing into the breast. Since blood is primarily water, this reduces the difficulty of correctly segmenting fibroglandular tissue. Removing blood from the MRI image also makes it easier for the neural network to generate MRI images of breast microcalcifications.

[0033] In another embodiment, gradient echo magnetic resonance imaging has an echo time. The echo time is a long echo time.

[0034] In another embodiment, the gradient echo MRI data has an echo time. In this embodiment, the echo time is between 100% and 90% of the T2* time for adipose tissue. In some examples, this echo time can provide a definition of a long echo time. This embodiment can be advantageous because it can provide a way to identify small breast microcalcifications.

[0035] In another aspect, the present invention provides a computer program comprising machine-executable instructions for execution by a computing system configured to control a medical system. Execution of the machine-executable instructions causes the computing system to receive gradient-echo magnetic resonance imaging (MRI) data describing a breast region of a subject. Execution of the machine-executable instructions also causes the computing system to receive DIXON MRI data describing the breast region. The DIXON MRI data and the gradient-echo MRI data are spatially matched.

[0036] The execution of machine-executable instructions also enables the computing system to generate breast microcalcification MRI images by inputting gradient echo MRI data and DIXON MRI data into the breast microcalcification image reconstruction module. The breast microcalcification image reconstruction module is configured to output breast microcalcification MRI images in response to the input gradient echo MRI data and DIXON MRI data. The breast microcalcification phase MRI image describes the location of the breast microcalcifications. The gradient echo MRI data is a phase image.

[0037] In another aspect, the present invention provides a method for medical imaging. The method includes receiving gradient echo magnetic resonance imaging (MRI) data describing a breast region of a subject. The method also includes receiving DIXON magnetic resonance imaging (MRI) data describing the breast region. The DIXON MRI data and the gradient echo MRI data are spatially matched. The method further includes generating a breast microcalcification MRI image by inputting the gradient echo MRI data and the DIXON MRI data into a breast microcalcification image reconstruction module. The breast microcalcification image reconstruction module is configured to output a breast microcalcification MRI image in response to the input gradient echo MRI data and DIXON MRI data. The breast microcalcification phase MRI image describes the location of breast microcalcifications. The gradient echo MRI data is a phase image.

[0038] It should be understood that one or more of the above embodiments of the present invention can be combined, as long as the combined embodiments are not mutually exclusive.

[0039] As will be understood by those skilled in the art, aspects of the present invention can be embodied as apparatus, method, or computer program product. Therefore, aspects of the present invention can take the form of a completely hardware embodiment, a completely software embodiment (including firmware, resident software, microcode, etc.), or an embodiment combining software and hardware aspects, all of which can be generally referred to herein as a “circuit,” “module,” or “system.” Furthermore, aspects of the present invention can take the form of a computer program product embodied in one or more computer-readable media having computer-executable code embodied thereon.

[0040] Any combination of one or more computer-readable media can be used. A computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. As used herein, "computer-readable storage medium" encompasses any tangible storage medium capable of storing instructions executable by a processor or computing system of a computing device. A computer-readable storage medium may be referred to as a computer-readable non-transitory storage medium. A computer-readable storage medium may also be referred to as a tangible computer-readable medium. In some embodiments, a computer-readable storage medium may also be capable of storing data accessible by a computing system of a computing device. Examples of computer-readable storage media include, but are not limited to: floppy disks, magnetic hard disk drives, solid-state drives, flash memory, USB thumb drives, random access memory (RAM), read-only memory (ROM), optical discs, magneto-optical discs, and register files of computing systems. Examples of optical discs include compact discs (CDs) and digital versatile discs (DVDs), such as CD-ROMs, CD-RWs, CD-Rs, DVD-ROMs, DVD-RWs, or DVD-R discs. The term computer-readable storage medium also refers to various types of recording media accessible by a computer device via a network or communication link. For example, data can be retrieved via a modem, the Internet, or a local area network. Computer-executable code embodied on a computer-readable medium may be transmitted using any suitable medium, including but not limited to wireless, wired, fiber optic cable, RF, or any suitable combination of the foregoing media.

[0041] Computer-readable signal media may include data signals propagated, for example, in baseband or as part of a carrier wave, having computer-executable code embodied therein. Such propagated signals may take any of a variety of forms, including, but not limited to, electromagnetic, optical, or any suitable combination thereof. A computer-readable signal medium may be any computer-readable medium that is not a computer-readable storage medium but is capable of communicating, propagating, or transmitting programs for use by or in conjunction with an instruction execution system, apparatus, or device.

[0042] "Computer memory" or "memory device" is an example of a computer-readable storage medium. Computer memory is any memory that a computing system can directly access. "Computer storage" or "memory device" is another example of a computer-readable storage medium. Computer storage is any non-volatile computer-readable storage medium. In some embodiments, computer memory can also be computer memory, and vice versa.

[0043] As used herein, "computing system" encompasses electronic components capable of executing programs or machine-executable instructions or computer-executable code. References to computing systems that include examples of "computing system" should be interpreted as potentially including more than one computing system or processing core. A computing system can be, for example, a multi-core processor. A computing system can also refer to a collection of computing systems within a single computer system or distributed across multiple computer systems. The term computing system should also be interpreted as potentially referring to a collection or network of computing devices, each including a processor or computing system. Machine-executable code or instructions can be executed by multiple computing systems or processors, which may be within the same computing device or even distributed across multiple computing devices.

[0044] Machine-executable instructions or computer-executable code may include instructions or programs that cause a processor or other computing system to perform aspects of the present invention. Computer-executable code for performing operations of aspects of the present invention may be written and compiled into machine-executable instructions using any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, Smalltalk, C++, etc., and traditional procedural programming languages ​​such as the "C" programming language, or similar programming languages. In some cases, the computer-executable code may be in the form of a high-level language or a pre-compiled form and may be used in conjunction with an interpreter that generates machine-executable instructions on the spot. In other cases, the machine-executable instructions or computer-executable code may be in the form of programming for programmable gate arrays.

[0045] Computer executable code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the latter case, the remote computer can be connected to the user's computer via any type of network (including a local area network (LAN) or a wide area network (WAN)), or it can be connected to an external computer (e.g., via the Internet through an Internet service provider).

[0046] Various aspects of the invention are described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It should be understood that each block or part of a block in a flowchart, illustration, and / or block diagram can be implemented, where applicable, by computer program instructions in the form of computer-executable code. It should also be understood that combinations of blocks from different flowcharts, illustrations, and / or block diagrams can be combined where they are not mutually exclusive. These computer program instructions can be provided to a computing system of a general-purpose computer, special-purpose computer, or other programmable data processing apparatus to produce a machine such that the instructions, executed via the computing system of the computer or other programmable data processing apparatus, create a manner for implementing the functions / actions specified in the blocks of the flowcharts and / or block diagrams.

[0047] These machine-executable instructions or computer program instructions may also be stored in a computer-readable medium that can direct a computer, other programmable data processing apparatus or other device to operate in a particular manner, such that the instructions stored in the computer-readable medium produce an article of writing which includes instructions that implement functions / actions specified in the frames of flowcharts and / or block diagrams.

[0048] Machine-executable instructions or computer program instructions may also be loaded onto a computer, other programmable data processing apparatus or other equipment to cause a series of operational steps to be performed on the computer, other programmable apparatus or other equipment to produce a computer-implemented process, such that the instructions, which execute on the computer or other programmable apparatus, provide for implementing the functions / actions specified in the frames of flowcharts and / or block diagrams.

[0049] As used herein, a "user interface" is an interface that allows a user or operator to interact with a computer or computer system. A "user interface" can also be called a "human-machine interface device." A user interface can provide information or data to and / or receive information or data from an operator. A user interface enables input from an operator to be received by the computer and output from the computer to the user. In other words, a user interface allows an operator to control or manipulate a computer, and the interface allows the computer to indicate the effects of the operator's control or manipulation. The display of data or information on a monitor or graphical user interface is an example of providing information to an operator. Receiving data via a keyboard, mouse, trackball, touchpad, pointing stick, graphics tablet, joystick, game controller, webcam, headset, pedal, wired gloves, remote control, and accelerometer are all examples of user interface components that can receive information or data from an operator.

[0050] As used herein, "hardware interface" encompasses the interface that enables a computer system to interact with and / or control external computing devices and / or devices. A hardware interface may allow a computing system to send control signals or instructions to external computing devices and / or devices. A hardware interface may also enable a computing system to exchange data with external computing devices and / or devices. Examples of hardware interfaces include, but are not limited to: Universal Serial Bus (USB), IEEE 1394 port, parallel port, IEEE 1284 port, serial port, RS-232 port, IEEE-488 port, Bluetooth connectivity, wireless LAN connectivity, TCP / IP connectivity, Ethernet connectivity, control voltage interface, MIDI interface, analog input interface, and digital input interface.

[0051] As used herein, "display" or "display device" encompasses an output device or user interface suitable for displaying images or data. Displays can output visual, audio, and / or tactile data. Examples of displays include, but are not limited to: computer monitors, television screens, touchscreens, tactile electronic displays, Braille screens, cathode ray tubes (CRTs), memory tubes, bistable displays, electronic paper, vector displays, flat panel displays, vacuum fluorescent displays (VFs), light-emitting diode (LED) displays, electroluminescent displays (ELDs), plasma display panels (PDPs), liquid crystal displays (LCDs), organic light-emitting diode (OLED) displays, projectors, and head-mounted displays.

[0052] K-space data are defined in this paper as recorded measurements of radio frequency signals emitted by atomic spins using the antenna of a magnetic resonance imaging (MRI) device during a magnetic resonance imaging (MRI) scan. MRI data is an example of computed tomography (CT) medical imaging data.

[0053] Magnetic resonance imaging (MRI) images, MR images, or MRI data are defined herein as reconstructed two-dimensional or three-dimensional visualizations of anatomical data contained within MRI data. Such visualizations can be performed using a computer. Attached Figure Description

[0054] Preferred embodiments of the invention will be described below by way of example only and with reference to the accompanying drawings, wherein:

[0055] Figure 1 This demonstrates an example of a medical system;

[0056] Figure 2 The instructions are shown. Figure 1 A flowchart of the methods used in medical systems;

[0057] Figure 3 This illustrates another paradigm of the medical system.

[0058] List of reference numerals

[0059] 100 Medical System

[0060] 102 Computer

[0061] 104 hardware interface

[0062] 106 Computing System

[0063] 108 User Interface

[0064] 110 memory

[0065] 120 Machine Executable Instructions

[0066] 122 Breast Microcalcification Image Reconstruction Module

[0067] 124 gradient echo magnetic resonance imaging data

[0068] 126 DIXON MRI data

[0069] 128 MRI images of breast microcalcifications

[0070] 130 high-pass spatial filter

[0071] 132 filtered images of breast microcalcifications

[0072] 134 projection images

[0073] 200 receives gradient echo magnetic resonance imaging data describing the breast region of the subject.

[0074] 202 Receives DIXON MRI data describing the breast region

[0075] 204. By inputting gradient echo magnetic resonance imaging data and DIXON magnetic resonance imaging data into the breast microcalcification image reconstruction module, a breast microcalcification magnetic resonance image is generated.

[0076] 300 Medical System

[0077] 302 Magnetic Resonance Imaging System

[0078] 304 magnet

[0079] 306 magnet hole

[0080] 308 imaging area

[0081] 309 Area of ​​Interest

[0082] 310 magnetic gradient coil

[0083] 312 magnetic field gradient coil power supply

[0084] 314 RF coil

[0085] 316 transceiver

[0086] 318 objects

[0087] 319 Breast Area

[0088] 320 object support

[0089] 330 Pulse Sequence Command

[0090] 332k-Spatial Data Detailed Implementation

[0091] Elements with the same number in these figures are either equivalent elements or perform the same function. If they have the same function, elements already discussed earlier need not be discussed in the following figures.

[0092] Figure 1 An example of a medical system 100 is shown. The medical system 100 is shown as including a computing system 106, in which, in this example, the computing system 106 is part of a computer 102. However, the computing system 106 can be embedded in or integrated into other devices or systems. Computer 102 is intended to represent one or more computers that can be networked or connected together. In different examples, the medical system 100 can take different forms. In one example, the medical system 100 can be a workstation or terminal controlling a magnetic resonance imaging system. In other examples, the medical system 100 can be a computer or workstation for viewing medical images. In yet another example, the medical system 100 can be a remote computer providing cloud computing or other image processing services.

[0093] Computer 102 is also shown as including a hardware interface 104 connected to computing system 106. Hardware interface 104 enables computing system 106 to communicate with other computer systems and computing networks. In some examples, medical system 100 may include additional components, such as a magnetic resonance imaging (MRI) system, and hardware interface 104 enables computing system 106 to control the operation and functions of the MRI system. Computer 10 is also shown as including a user interface 108 and memory 110 also connected to computing system 106. User interface 108 may be an optional component that enables an operator to use and control medical system 100. Memory 110 is intended to represent any combination of memory accessible to computing system 106. In some examples, memory 110 may be a non-transitory storage medium.

[0094] Memory 110 is shown to contain machine-executable instructions 120. The machine-executable instructions 120 enable the computing system 106 to control the operation and functions of the medical system 100. The machine-executable instructions 120 also enable the computing system 106 to perform various computational tasks, data processing tasks, and image processing tasks. For example, it enables the computing system 106 to perform magnetic resonance imaging reconstruction.

[0095] The memory 110 is also shown to include a breast microcalcification image reconstruction module 122. The breast microcalcification image reconstruction module 122 is configured to output a breast microcalcification magnetic resonance image 128 in response to receiving gradient echo magnetic resonance imaging data 124 and DIXON magnetic resonance imaging data 126. In various examples, the DIXON magnetic resonance imaging data 126 may be a purely water image. In other examples, the DIXON magnetic resonance imaging data 126 includes both water and fat images. The gradient echo magnetic resonance imaging data 124, the DIXON magnetic resonance imaging data 126, and the breast microcalcification magnetic resonance image 128 are shown to be stored in the memory 110.

[0096] In different paradigms, the breast microcalcification image reconstruction module 122 can take different forms. It can be a purely algorithmic module, a hybrid module that combines some algorithms with various artificial intelligence components such as neural networks, or it can be implemented simply as a neural network.

[0097] The memory 110 is also shown to include an optional high-pass spatial filter. The high-pass spatial filter is configured such that objects with smaller spatial features compared to the image size are preserved in the image after larger features have been removed. The high-pass spatial filter reduces the low spatial frequency content of the image. One way to implement this is by using a kernel of a specific size.

[0098] Memory 110 is also shown to contain a filtered breast microcalcification image 132 received in response to an input breast microcalcification MRI image 128 via a high-pass spatial filter 130. This embodiment can be advantageous because it may minimize or remove objects that are not microcalcifications. Furthermore, the use of a high-pass spatial filter helps in simulating mammograms. Memory 110 is also shown to contain a projection image 134. Projection image 134 is a projection of the filtered breast microcalcification image 132 onto a two-dimensional plane. The filtered breast microcalcification image 132 can be, for example, a three-dimensional MRI image, or it can be a collection of two-dimensional slices, which is also considered a three-dimensional dataset. Projection image 134 can be useful for radiologists who are accustomed to and trained to interpret mammograms. The combination of the high-pass spatial filter and the generation of projection image 134 produces an MRI image suitable as an alternative to a mammogram.

[0099] Figure 2 The instructions are shown. Figure 1 The flowchart of the method of the medical system 100 is as follows. First, in step 200, gradient echo magnetic resonance imaging (MRI) data 124 is received. Next, in step 202, DIXON magnetic resonance imaging (MRI) data 126 is received. This DIXON MRI data 126 also describes a breast region. The breast region can be, for example, a region of interest in the MRI image. The DIXON MRI data 126 and the gradient echo MRI data 124 are spatially matched. Finally, in step 204, a breast microcalcification MRI image 128 is generated by inputting the gradient echo MRI data 124 and the DIXON MRI data 126 into a breast microcalcification image reconstruction module 122.

[0100] Figure 3 This illustrates another example of medical system 300. Figure 3 The medical system depicted in the text is similar to 300 Figure 1 The medical system 100 in the system includes a magnetic resonance imaging system 302 in addition to the medical system.

[0101] The magnetic resonance imaging system 302 includes a magnet 304. Magnet 304 is a superconducting cylindrical magnet with a hole 306 passing through it. Different types of magnets may also be used; for example, split cylindrical magnets and so-called open magnets may also be used. A split cylindrical magnet is similar to a standard cylindrical magnet, except that the cryostat has been divided into two parts to allow access to the equiplanar plane of the magnet; such a magnet can be used, for example, in conjunction with charged particle beam therapy. An open magnet has two magnet sections, one on top of the other, with a sufficiently large space between them to receive the object: the arrangement of the two sections is similar to that of a Helmholtz coil. Open magnets are popular because the object is not restricted. An assembly of superconducting coils is located inside the cryostat of the cylindrical magnet.

[0102] An imaging region 308 is located within the aperture 306 of a cylindrical magnet 304, wherein the magnetic field is strong and uniform, sufficient to perform magnetic resonance imaging. A region of interest 309 is shown within the imaging region 308. The acquired magnetic resonance data is typically acquired for the region of interest. An object 318 is shown supported by an object support 320 such that at least a portion of the object 318 is within the imaging region 308 and the region of interest 309. The region of interest 309 is shown to have a breast region or breast tissue 319 therein.

[0103] Within the aperture 306 of the magnet, there is also a set of magnetic field gradient coils 310, which are used to acquire preliminary magnetic resonance data for spatial encoding of the magnetic spin within the imaging region 308 of the magnet 304. The magnetic field gradient coils 310 are connected to a magnetic field gradient coil power supply 312. The magnetic field gradient coils 310 are intended to be representative. Typically, the magnetic field gradient coils 310 contain three separate sets of coils for spatial encoding in three orthogonal spatial directions. The magnetic field gradient power supply provides current to the magnetic field gradient coils. The current supplied to the magnetic field gradient coils 310 is controlled as a function of time and can be tilted or pulsed.

[0104] Adjacent to the imaging region 308 is an RF coil 314, used to manipulate the direction of the magnetic spin within the imaging region 308 and to receive radio transmissions from the spin, also within the imaging region 308. The RF antenna may comprise multiple coil elements. The RF antenna may also be referred to as a channel or antenna. The RF coil 314 is connected to an RF transceiver 316. The RF coil 314 and the RF transceiver 316 may be replaced by separate transmit and receive coils, and separate transmitters and receivers. It should be understood that the RF coil 314 and the RF transceiver 316 are representative. The RF coil 314 is also intended to represent a dedicated transmit antenna and a dedicated receive antenna. Similarly, the transceiver 316 may also represent a separate transmitter and receiver. The RF coil 314 may also have multiple receive / transmit channels, and the RF transceiver 316 may have multiple receive / transmit channels. For example, if a parallel imaging technique such as SENSE is performed, the RF coil 314 will have multiple coil elements.

[0105] Transceiver 316 and gradient controller 312 are shown as a hardware interface 106 connected to computer system 102. Memory 110 is also shown as containing pulse sequence commands 330. The pulse sequence commands contain data or instructions that can be used to control the magnetic resonance imaging system 302 to acquire k-space data 332 according to the DIXON magnetic resonance imaging protocol and the gradient echo magnetic resonance imaging protocol. Memory 110 is also shown as containing k-space data 332 acquired by controlling the magnetic resonance imaging system 302 with pulse sequence commands 330.

[0106] Breast microcalcifications are one of the early precursors to breast cancer. Although mammography is currently the preferred diagnostic modality for microcalcifications, it uses ionizing X-ray radiation. Reliable detection of breast microcalcifications using MRI would provide a safer option for breast cancer screening. Calcifications found in vivo have a significantly higher susceptibility compared to surrounding tissue. The MRI signal phase is highly sensitive to changes in this sensitivity. The presence of fat can also contribute to the MRI signal phase, forming a confounding factor. In the proposed method, prior information from mDIXON or other types of DIXONS scans is used to remove the contribution of fat, creating a relatively clean phase image that is primarily sensitive to the presence of breast calcifications. This approach would be a powerful enabler for MRI-based breast cancer screening because it addresses one of the key limitations of MRI—the inability to image microcalcifications.

[0107] Breast cancer is the second most common form of cancer worldwide and a leading cause of cancer in women. Mammography is the standard first-line diagnostic imaging for breast cancer screening. The European Society for Medical Oncology, the American Cancer Society, and the Indian Council of Medical Research recommend annual mammography for women aged 45 and older and / or with a family history of breast cancer. Microcalcifications associated with ductal carcinoma in situ (DCIS) are observed in approximately 31% of patients undergoing mammography screening, a potential precursor to invasive breast cancer. Because not all patients with DCIS progress to the invasive cancer stage, the detection of microcalcifications on an initial screening mammogram usually prompts a follow-up mammogram or a magnetic resonance imaging (MRI)-based evaluation. The mammographic procedure utilizes ionizing radiation for imaging purposes, which is itself a significant carcinogenic risk factor. Annual mammography further increases exposure to ionizing radiation. MRI, on the other hand, provides an alternative imaging method for breast cancer assessment based on non-ionizing radiation. However, a major drawback of using MRI for routine breast cancer screening is that it cannot definitively detect microcalcifications, which are one of the early imaging markers of breast cancer.

[0108] Current MRI methods utilize phase-based pattern matching or quantitative susceptibility mapping (QSM) to detect breast calcifications. Due to long scan times, most of these methods have not yet been used in vivo or for detecting near-microcalcifications or microcalcifications. They have been used to visualize large breast calcifications. QSM-based methods have previously been used to detect large breast calcifications. QSM is a concise tool that can interpret phase effects between fibroadenomas and fat and can help visualize calcified nodules. However, QSM has not yet been shown to be useful in detecting microcalcifications. In this work, we propose an optimized MRI protocol and a post-processing algorithm that directly uses MRI for the detection of breast calcifications.

[0109] Due to relative size differences, breast calcifications appear nodular and are typically partial volumes relative to the imaging voxel size (microcalcifications are approximately 0.1–0.4 mm in diameter, and typical imaging voxel sizes are approximately 2 mm in slice thickness). The partial volume phase is detectable in gradient echo MRI phase images and can be identified as nodules using projection images because vascular and fibroglandular tissues are continuous structures. Fibroadenoid tissue also exhibits significant phase effects due to its own sensitivity differences compared to surrounding breast fat. This phase effect can be mitigated if the geometry of the fibroglandular tissue is known a priori. This information can be obtained from mDIXON or other DIXON technologies, the implementation of which is unique to Philips. The phase from the fibroglandular tissue can then be estimated using a universal dipole kernel. Using this simulated phase information, clean breast phase images sensitive only to calcified nodules can then be obtained.

[0110] Visualization of breast calcifications in phase images has been previously completed. Similarly, in the case of the brain, the use of tissue geometry to predict phase information, where the tissue geometry is extracted from different acquisitions, has been previously performed. The novelty of this invention / innovation lies in the unique combination of gradient echo acquisition for breast calcification detection applications with mDIXON or other DIXON acquisitions.

[0111] A method for detecting breast calcification may include one or more of the following steps:

[0112] 1) Acquire long echo time (TE) gradient echo data and mDIXON or other DIXON data with matched resolution and bandwidth—either in a single sequence or as separate acquisitions. Long echo time (TE) gradient echo data can be gradient echo MRI data 124, or simply echo time (TE) gradient echo data without the “long TE” label. mDIXON or other DIXON data can be DIXON MRI data 126.

[0113] 2) In some examples, compressed SENSE can be used to acquire high-resolution data more quickly, but it is not necessary. In some examples, the quantization in terms of voxels is on the order of 1:1:3 to 1:1:5, and the voxel size is on the order of 0.1 x 0.1 mm² to 0.4 x 0.4 mm².

[0114] 3) Long TE gradient echo data are acquired using TE in phase with fat. As mentioned above, being in phase with fat is advantageous, but it also allows for greater freedom in phase selection.

[0115] 4) Obtaining the geometry of fibroglandular tissue from mDIXON images using image processing methods (fibroglandular tissue segmentation) including but not limited to: thresholding, region growing, erosion, and dilation for smoothing geometry extraction. This is fibroglandular tissue segmentation from DIXON MRI data 126.

[0116] 5) If necessary, saturation bands during data acquisition or vascular tracking in data acquired in an unsaturated state can be used to remove vascular structures within mDIXON images.

[0117] 6) Use one of the dipole kernels or use machine learning methods to simulate the phase effect of fibroglandular tissue.

[0118] 7) Remove the phase effect of fibroglandular tissue from long TE phase images by phase subtraction.

[0119] 8) Expand the phase.

[0120] 9) Filter the phase image (using a high-pass spatial filter) to remove low-frequency background field information.

[0121] 10) Create projection image 134 (maximum or minimum intensity projection, depending on the asymmetry of the phase image) for visualization of nodular breast calcification.

[0122] 11) An alternative implementation may use a machine learning framework (e.g., a neural network) to take phase and geometry information as input and output a projected image showing only the calcified nodules.

[0123] Although the invention has been described and illustrated in detail in the accompanying drawings and the foregoing description, such description and illustration should be considered illustrative or exemplary rather than restrictive; the invention is not limited to the disclosed embodiments.

[0124] By studying the accompanying drawings, the disclosure, and the appended claims, those skilled in the art can understand and implement other variations to the disclosed embodiments in practicing the claimed invention. In the claims, the word "comprising" does not exclude other elements or steps, and the indefinite articles "a" or "an" do not exclude a plural. A single processor or other unit can perform the functions of several items listed in the claims. The fact that certain measures are listed in mutually different dependent claims does not indicate that a combination of these measures cannot be used advantageously. Computer programs can be stored / distributed on suitable media, such as optical storage media or solid-state media provided with or as part of other hardware, but can also be distributed in other forms, such as via the Internet or other wired or wireless telecommunications systems. Any reference numerals in the claims should not be construed as limiting the scope.

Claims

1. A medical system (100, 300), comprising: - A memory (110) storing machine-executable instructions (120) and a breast microcalcification image reconstruction module (122), wherein the breast microcalcification image reconstruction module is configured to output a breast microcalcification magnetic resonance image (128) in response to input gradient echo magnetic resonance imaging data (124) and DIXON magnetic resonance imaging data (126), wherein the breast microcalcification magnetic resonance image describes the location of breast microcalcifications, wherein the gradient echo magnetic resonance imaging data is a phase image; and - A computing system configured to control the medical system, wherein the execution of machine-executable instructions causes the computing system to: - Receive the gradient echo magnetic resonance imaging data of the breast region (319) of the object (318); - Receive the DIXON magnetic resonance imaging data describing the breast region, wherein the DIXON magnetic resonance imaging data and the gradient echo magnetic resonance imaging data are spatially matched; and - The breast microcalcification magnetic resonance image is generated (204) by inputting the gradient echo magnetic resonance imaging data and the DIXON magnetic resonance imaging data into the breast microcalcification image reconstruction module, wherein the breast microcalcification image reconstruction module is implemented as a trained breast microcalcification image reconstruction neural network.

2. The medical system according to claim 1, wherein, The execution of the machine-executable instructions also enables the computing system to compute a filtered breast microcalcification image (132) by applying a high-pass spatial filter (130) to the breast microcalcification magnetic resonance image.

3. The medical system according to claim 2, wherein, The execution of the machine-executable instructions also enables the computing system to generate a breast microcalcification projection image (134) by calculating the projection of the filtered breast microcalcification image.

4. The medical system according to claim 1, 2 or 3, wherein, The breast microcalcification image reconstruction module is implemented as an algorithm including the following steps: - Calculate the fibroglandular tissue segmentation of the DIXON magnetic resonance imaging data, wherein the fibroglandular tissue segmentation identifies the location of the fibroglandular tissue by inputting the DIXON magnetic resonance imaging data into the fibroglandular tissue recognition module; - The phase image of the fibroglandular tissue is calculated by inputting the segmented fibroglandular tissue into the phase image calculation module; - The adjusted phase image is calculated by subtracting the phase image of the fibroglandular tissue from the gradient echo magnetic resonance imaging data; and - Provide the breast microcalcification magnetic resonance image by performing phase unrolling on the adjusted phase image.

5. The medical system according to claim 4, wherein, The fibroglandular tissue identification module is configured to provide segmentation of the fibroglandular tissue in any of the following ways: - Thresholding, region growing, erosion, and dilation for smooth geometry extraction; and - Input the DIXON magnetic resonance imaging data into a trained neural network for recognizing fibroglandular tissue; as well as - Their combination.

6. The medical system according to claim 4, wherein, The phase image calculation module is implemented as any of the following: - Use the dipole kernel computing module; and - A trained phase image generation neural network.

7. The medical system according to claim 1, 2 or 3, wherein, The execution of the machine-executable instructions also enables the computing system to: - Receive k-space data acquired according to the DIXON magnetic resonance imaging protocol and the gradient echo magnetic resonance imaging protocol with a single gradient echo or multiple gradient echoes (332). - Reconstruct the gradient echo magnetic resonance imaging data according to the gradient echo magnetic resonance imaging protocol; and - Reconstruct the DIXON magnetic resonance imaging data according to the DIXON magnetic resonance imaging protocol.

8. The medical system according to claim 7, wherein, The DIXON magnetic resonance imaging protocol is a compressed sensing protocol and / or the gradient echo magnetic resonance imaging protocol is a compressed sensing protocol.

9. The medical system according to claim 7, wherein, The DIXON magnetic resonance imaging protocol is a multi-point DIXON magnetic resonance imaging protocol.

10. The medical system according to claim 7, wherein, The medical system further includes a magnetic resonance imaging system (302) configured to acquire k-space data from an imaging region, wherein the memory further contains pulse sequence commands (330) configured to acquire measured k-space data acquired according to the DIXON magnetic resonance imaging protocol and the gradient echo magnetic resonance imaging protocol, wherein execution of the machine-executable instructions further enables the computing system to acquire the k-space data by controlling the magnetic resonance imaging system using the pulse sequence commands.

11. The medical system according to claim 10, wherein, The pulse sequence command is configured to acquire k-space data for the region of interest, wherein the pulse sequence command is also configured to inhibit vascular structures by implementing at least one saturation tract outside the region of interest.

12. The medical system according to claim 7, wherein, The gradient echo magnetic resonance imaging data has an echo time, wherein the echo time is a long echo time, which is defined as the T2 of adipose tissue for a specific magnetic resonance imaging protocol. Equivalent to or smaller than the T2 of adipose tissue for a specific magnetic resonance imaging protocol The echo time.

13. The medical system according to claim 12, wherein, The echo time is at T2 for adipose tissue. Between 100% and 90% of the time.

14. A computer program comprising machine-executable instructions (120) for execution by a computing system (106) configured to control a medical system (100, 300), in, The execution of the machine-executable instructions enables the computing system to: - Receive gradient echo magnetic resonance imaging data (124) of the breast region (319) of the object (318). - Receive DIXON magnetic resonance imaging data (126) describing the breast region, wherein the DIXON magnetic resonance imaging data and the gradient echo magnetic resonance imaging data are spatially matched; and - A breast microcalcification magnetic resonance image (128) is generated (204) by inputting the gradient echo magnetic resonance imaging data and the DIXON magnetic resonance imaging data into a breast microcalcification image reconstruction module (122), wherein the breast microcalcification image reconstruction module is implemented as a trained breast microcalcification image reconstruction neural network and is configured to output a breast microcalcification magnetic resonance image in response to inputting the gradient echo magnetic resonance imaging data and the DIXON magnetic resonance imaging data, wherein the breast microcalcification magnetic resonance image describes the location of breast microcalcifications, and wherein the gradient echo magnetic resonance imaging data is a phase image.

15. A medical imaging method, wherein, The method includes: - Receive gradient echo magnetic resonance imaging data (124) of the breast region (319) of the object (318). - Receive DIXON magnetic resonance imaging data describing the breast region, wherein the DIXON magnetic resonance imaging data and the gradient echo magnetic resonance imaging data are spatially matched; and - A breast microcalcification magnetic resonance image (128) is generated (204) by inputting the gradient echo magnetic resonance imaging data and the DIXON magnetic resonance imaging data into a breast microcalcification image reconstruction module (122), wherein the breast microcalcification image reconstruction module is implemented as a trained breast microcalcification image reconstruction neural network and is configured to output a breast microcalcification magnetic resonance image in response to input of the gradient echo magnetic resonance imaging data and the DIXON magnetic resonance imaging data, wherein the breast microcalcification magnetic resonance image describes the location of breast microcalcifications, and wherein the gradient echo magnetic resonance imaging data is a phase image.

Images