DE102013202768A1 - Method for automatically determining adjustment volume in magnetic resonance system, involves identifying anatomical structure in magnetic resonance-image and automatically determining adjustment volume - Google Patents

Abstract

The method involves receiving (34) a magnetic resonance (MR)-image of an object, where a measurement volume and the adjustment volume are depicted in the MR-image. An anatomical structure is identified (35) in the MR-image, where the adjustment volume is automatically determined (36) depending on the identified anatomical structure. The MR-image is segmented and is compared with predetermined atlases, which each represent different areas of the object with the associated anatomical structures. An independent claim is included for a magnetic resonance system.

Description

The present invention relates to a method for the automatic determination of an adaptation volume in an MR system and an MR system therefor.

In order to achieve a good image quality and a high resolution in the MR images in MR systems, it is sometimes necessary to carry out adaptation steps before the actual imaging. In these adaptation steps effects can be compensated which result from the anatomy of the subject / patient and the examined tissue characteristics. For example, the homogeneity of the polarization field B0 can be adjusted by a so-called B0 shim. The resonance frequency can be adapted to the resonance frequency of the measuring volume. Furthermore, the RF transmission power radiated into the examination subject may be adjusted to excite the spins of a target region with a desired tilt angle. Likewise, in MR systems with a plurality of RF transmission channels, the excitation profile of the so-called B1 field of the RF excitation can also be adapted by so-called B1 shim methods.

It is also possible to optimize other parameters of the MR system for a given measurement volume, for example the compensation parameters for compensation of eddy currents or the Maxwell compensation parameters could be optimized at a different location in the MR system than for the ISO center of the magnet , The adaptation of the MR system to the examination object is of particular importance in spectroscopic imaging methods, but these adaptation steps are not limited to spectroscopic imaging methods, but may also be of importance in conventional MR imaging.

One way of determining such an adaptation volume is to define an area via an interface in the MR images by drawing a box, which area is to be used in order to adapt the MR system to the upcoming examination, in particular in the area within this box , Furthermore, there is theoretically the possibility that the operator of the MR system on manually created MR images by hand draws an area to which the adjustment steps should be particularly aimed. However, this is very time consuming for the operator. Often this area contains volumes in which an adaptation can not be optimally performed, or not the entire area is of importance for the clinical question.

It is therefore an object of the present invention to further improve a determination of an adaptation volume, in particular to minimize the operating steps of the operator and to optimize the adaptation specifically in the volume from which a finding is to be made later. This object is solved by the features of the independent claims. Further embodiments are described in the dependent claims.

According to a first aspect of the invention, a method for the automatic determination of an adaptation volume in an MR system is provided, wherein MR signals are recorded in the MR system from a measurement volume of an examination subject. Before the acquisition of the MR signals from the measurement volume, an adaptation of components of the MR system to the examination object located in the MR system is carried out specifically in the adaptation volume. The method comprises the step of recording at least one first MR image of the examination object, in which the measurement volume and the adaptation volume are mapped. Furthermore, at least one anatomical structure is identified in the at least one first MR image. Taking into account the indexed anatomical structure, the adaptation volume is then automatically determined. The measurement volume is the volume from which an MR image is later taken, which is used to answer a clinical question. However, in the recorded MR image for the clinical diagnosis, not the entire area of the MR image is usually of interest. In most cases, interest is limited to a section of this MR image. By separating the measurement volume and the adaptation volume, it is possible to determine the adaptation volume, if necessary, such that it is not identical to the measurement volume. The adaptation volume can be limited to areas of particular interest and is no longer automatically limited to the measurement volume. Furthermore, the determination of the adaptation volume takes place automatically as a function of anatomical structures that were identified in the first MR image. This considerably accelerates the examination procedure and allows the determination of the adaptation volume as a function of the examined body region.

For the identification of the anatomical structures can proceed as follows. For example, it is possible for the recorded at least one first MR image to be segmented for the purpose of detecting predetermined patterns in this first MR image. The segmented first MR image can then be compared with predetermined atlases, each representing different regions of the examination subject with the associated anatomical structures. By comparison of the segmetierten Structures with the atlases can identify the body region in the subject being examined, such as the abdomen or the head. If the body region shown in the first MR image is known, then it can be deduced which clinical questions should be answered and the adaptation volume can then be set to a desired range, ie a desired volume.

This achieves a particularly good image quality in this adaptation volume.

To further accelerate the measurement method, it is possible that the identified anatomical structures are further used to determine the measurement volume, such as imaging planes for the acquisition of MR images or MR signals. For example, if a particular part of the head has been identified in the first MR image, such as certain anatomical structures, then these anatomical structures can be used to define imaging planes for actual or further MR imaging. The actual MR images are the images that are created with the MR signals from the measurement volume and for which the adaptation was performed. In many areas of the body, standard imaging levels exist that can be used to better compare images between different patients or images taken by a patient over time during treatment. The identified anatomical structures can be used in this case for the automatic determination of the measurement volume and the automatic determination of the adaptation volume.

It is possible that the adaptation volume is smaller than the measurement volume. As a result, portions of the measurement volume in which a particularly good fit is either not possible or not important may be excluded from the fit, and the fit steps may be restricted to a particular range. As a result, the adaptation can be carried out better in this subarea than in the case of adaptation over the entire measurement volume, since subareas of the measurement volume can be excluded during the adaptation.

One possible application of the method is in the imaging of the head. An application on the head is here in the production of spectroscopic MR images, in which spatially resolved in individual voxels, the associated spectra from the MR signals are determined. However, the method is not limited to imaging in the head, but can be applied anywhere else in the body. Also, the method is not limited to spectroscopic imaging. Even with "conventional" MR imaging, it may be necessary to adapt the measurement to a specific volume, the adaptation volume.

If an examiner is positioned in the MR system, a so-called MR overview image can be recorded. One possibility now is to automatically determine the position of the at least one first MR image in the examination object from the recorded MR overview image. If, for example, an overview image of the head is taken, then the midsagittal plane of the head can automatically be determined, which is then the image plane in which the first MR image is recorded. In this first MR image, i. in the mid-sagittal plane, certain anatomical structures can then be identified, with the help of which the adaptation volume can be automatically determined.

Furthermore, it is possible that the position of the at least one first MR image in the examination subject is automatically determined as a function of the selection of an imaging sequence with which MR images of the measurement volume are to be recorded. For example, if an overview image of the abdomen has been taken and an operator has selected an imaging sequence that is specifically used for some protocol of imaging sequences, such as for creating MR images of the heart, it can be concluded that the adaptation volume is in the range of Heart lies and not elsewhere in the abdomen.

The invention further relates to an MR system which carries out the method described above, having a control unit which is designed to record at least a first MR image of the examination subject in which the measurement volume and the adaptation volume are imaged. Furthermore, a computing unit is provided, which is designed to identify anatomical structures in the at least one MR image and to automatically determine the adaptation volume as a function of the identified anatomical structure.

In particular, the arithmetic unit can be designed as described in more detail in the above method in order to reliably identify the desired adaptation volume.

The invention will be explained in more detail with reference to the accompanying drawings. Show here

1 schematically an MR system with which an adaptation volume can be determined automatically,

2 by way of example the position of a measuring volume and an adaptation volume when taking MR images of a head,

3 FIG. 3 is a flowchart schematically showing the steps for automatically determining the adaptation volume. FIG.

4 how the adaptation volume can be automatically determined from identified anatomical structures in the example of the head.

1 schematically shows an MR system 1 , which is designed to adapt the components of the MR system specifically and specifically to an adaptation volume. The MR system has a magnet 2 to generate a polarization field B0. An examiner 3 is located in the magnet, resulting in a resulting magnetization in the B0 field in the direction of the B0 field. Via a gradient coil 4 For example, linear or non-linear magnetic field gradients can be generated for local coding of the MR signals. A radio frequency coil 5 can be used to detect the transverse magnetization. The sink 5 may be a pure receiving coil for receiving the MR signals, or a transmitting and receiving coil for emitting the RF pulse and for receiving the signal. If the coil 5 is only a receiving coil, the RF excitation pulse or the RF excitation pulses can be radiated through a body coil, not shown. Of course, a plurality of transmitting and receiving coils may be provided for a multi-channel transmitting unit or a multi-channel receiving unit.

How MR images can be generated by irradiation of RF pulses and switching of gradients is known to the person skilled in the art and will not be explained in more detail here.

The system also has an RF unit 6 for controlling the irradiated RF pulses and a gradient unit 7 to control the necessary magnetic field gradients. A central control unit 8th is provided, which controls the time course of the irradiated RF pulses and gradient fields as a function of a selected imaging sequence and thus also the RF unit 6 and the gradient unit 7 controls. Via an input unit 9 a user can use the MR system 1 operate, select imaging sequences, select image layers, etc. On a display unit 10 The generated MR images can be displayed and it can with the input unit 9 on the display 10 Measurements to be planned on the subject, for example by localization of the measurement volumes, from which then MR images are to be generated.

Furthermore, there is a database 11 are provided in which are stored by different areas of the body atlases that show the anatomical structure for the respective areas of the body by way of example. An arithmetic unit 12 is provided, which is adapted to apply in recorded MR images segmentation algorithms, such as algorithms for edge detection, which can identify the structures shown in the MR images.

As explained in detail below, the arithmetic unit compares 12 the recorded and segmented MR images with the atlases in the database 11 to identify which anatomical structure is currently displayed in a first MR image.

In 2 Shown schematically is a coronary, transverse and sagittal section of the subject's head. Generally, in MR images, it is desirable to examine certain predefined slice levels to allow for comparison between different subjects, and a prior and subsequent comparison to a single examiner. In 2 are in the sagittal cut / MR image 20 for example, two measuring volumes 21 and 22 represented, wherein meaningful MR images are to be generated from this measurement volume, which help the doctor in the diagnosis. Similarly, in the coronary view / MR image 23 two possible measurement volumes 24 and 25 shown, as well as in transverse MR image 26 the different measuring volumes 27 . 28 and 29 , Furthermore, in the images shown an adaptation volume 30 shown. This adjustment volume 30 is a special volume that is used to perform an adaptation of the components of the MR system to the recording situation before recording the actual MR images for the diagnosis. For example, in this adaptation volume, in particular the B0 field inhomogeneity can be minimized by so-called shim coils. Care is taken to ensure that the magnetic field inhomogeneity is minimal, especially in the adaptation volume. Furthermore, the RF frequency of the irradiation pulse can be adapted to the resonance frequency of the B0 field, especially in the adaptation volume. Likewise, an adaptation of the RF radiation power is possible in order to achieve a desired tilt angle or a desired tilt angle distribution in multi-channel systems in a target region. Furthermore, eddy currents can be compensated specifically for the adaptation volume or Maxwell compensation parameters can not be set in the isocenter of the magnet but outside the center. The correction of Maxwell parameters is in DE 198 21 780 described.

As in 2 can be seen, the invention distinguishes the measurement volume and the adaptation volume. In the invention, the MR system is optimized specifically for the adaptation volume and not for the entire measurement volume. The areas of interest for the clinical issue are usually limited to a smaller portion of the measurement volume. In the illustrated example of the head, it is sometimes difficult to achieve a homogeneous B0 field in all areas of the head, especially in areas where there are tissue junctions. According to the invention, it is now possible to exclude partial areas from the measuring volume which are then not taken into account in the adaptation steps. For example, in the frontal sinuses and the upper and lower jaw, the magnetic field is often very inhomogeneous, for example, by cavities in the mouth and throat area or by teeth and dental fillings. These subareas can then be disregarded when determining the adjustment volume.

In 3 is now described in more detail how the in 2 shown adjustment volume 30 can be automatically determined.

The procedure begins in step 31 , By the investigator can in a first step 32 an overview picture or several overview pictures are created. In a next step 33 then the position of the first MR images can be determined automatically. For example, if the overview image is a head image, then the midsagittal level can be determined automatically. The determination also does not have to be automatic, it is also possible to determine the position of the first MR image by the user. After taking the first MR images in the specific location in step 34 , then in step 35 automatically identify anatomical structures in the at least one first MR image. This allows the arithmetic unit 12 segment the MR images with known segmentation algorithms to identify the structures in the first MR images. By comparing the identified structures with predetermined reference structures stored in atlases, the anatomical structures represented in the first MR image can be identified (step 35 ). The identified anatomical structures can be used to automatically determine the adaptation volume in step 36 ,

This is schematically illustrated by an example 4 shown. In 4 is a sagittal cut in MR image 40 a head shown schematically. By segmentation algorithms and comparison with atlases of the head, for example, the following points could be identified. The tip of the occipital bone, with reference numerals 41 marked, and the structure Crista Galli, here with 42 shown. This line, the so-called AC-PC line, is an important anatomical feature for the positioning of measurement and adaptation volumes. In functional imaging (fMRI), the regions of the brain are imaged in a Talairach atlas. When recording with 2d techniques, the layer orientation is often placed parallel to an "AC-PC line".

For clarity, is in 4 the structure of the head shown in the MR image has been omitted. Join now the points 41 and 42 by a straight line and falls halfway along a lot, so for example a constant distance D1 can be covered on this Lot and in the head direction a distance D2. These constants D1 and D2 are known, so you have to point 43 which represents the center of the brain. By making the horizontal line 44 and by using the connecting line 45 between the points 41 and 42 Now the adjustment volume 46 be determined automatically, in the case shown, a circle, the lines 44 and 45 touched. Subareas of the measuring volume 49 can thus be excluded in the optimization of the signal in the adaptation volume, for example, the sub-area 47 or 48 , In the case shown, the subregions are with many tissue-air transitions, in which an adaptation such as a homogeneous B0 field is difficult to achieve. After determining the adaptation volume, the adaptation of the individual components, namely one or more of the following components, can take place: the adaptation of the B0 field, the resonance frequency, the B1 field, etc. With reference again to FIG 3 can in an optional step 37 the particular anatomical structure can also be used to automatically determine further image planes in which further MR images of the examination subject are to be recorded, for example from the measurement volume 45 , The procedure for automatically determining the adaptation volume ends in step 38 , Finally, the adaptation of the MR system can be carried out specifically to the adaptation volume. This in 4 described example is to describe only one way for the automatic determination of the adaptation volume. Of course, other anatomical areas such as abdomen or extremities are usable.

In summary, the present invention enables a simplified and automated determination of an adaptation volume.

QUOTES INCLUDE IN THE DESCRIPTION

This list of the documents listed by the applicant has been generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Cited patent literature

• DE 19821780 [0026]

Claims (10)

Method for automatically determining an adaptation volume ( 30 . 46 ) in an MR system ( 1 ), in the MR signals from a measuring volume ( 21 . 22 . 24 . 25 . 27 - 29 . 49 ) of an examination object ( 3 ), wherein, before the acquisition of the MR signals from the measurement volume, an adaptation of components of the MR system to the examination object located in the MR system is targeted in the adaptation volume (FIG. 30 . 46 ), comprising the following steps: recording at least one first MR image ( 20 . 23 . 26 . 40 ) of the examination subject ( 3 ), in which the measuring volume ( 21 . 22 . 24 . 25 . 27 - 29 . 49 ) and the adjustment volume ( 30 . 46 ), identifying at least one anatomical structure ( 41 . 42 ) in the at least one first MR image ( 20 . 23 . 26 . 40 ), - automatic determination of the adaptation volume ( 30 . 46 ) depending on the identified anatomical structure ( 41 . 42 ). A method according to claim 1, characterized in that the identification of anatomical structures comprises the following steps: - segmenting the at least one first MR image ( 20 . 23 . 26 . 40 ), - comparing the segmented at least one first MR image with predetermined atlases, each representing different areas of the examination subject with the associated anatomical structures, and - identifying the anatomical structures using the comparison. Method according to claim 1 or 2, characterized in that the identified anatomical structures ( 41 . 42 ) can be used to automatically determine the measurement volume for recording the MR signals. Method according to one of the preceding claims, characterized in that for determining the adaptation volume ( 30 . 46 ) a subarea ( 47 ) of the measuring volume is excluded, so that this subarea in the adaptation of components of the MR system to the in the MR system located examination object ( 3 ) is not considered. Method according to one of the preceding claims, characterized in that the at least one MR image is picked up by the head. Method according to one of the preceding claims, characterized in that the measuring volume ( 21 . 22 . 24 . 25 . 27 - 29 . 49 ) greater than the adjustment volume ( 30 . 46 ). Method according to one of the preceding claims, characterized in that the MR signals from the measurement volume are used for spectroscopic MR imaging. Method according to one of the preceding claims, characterized in that the position of the at least one first MR image ( 20 . 23 . 26 . 40 ) is determined automatically in the examination object from at least one recorded MR overview image. Method according to one of the preceding claims, characterized in that the position of the at least one first MR image ( 20 . 23 . 26 . 40 ) is determined automatically in the examination object as a function of the selection of an imaging sequence with which MR images of the measurement volume are to be recorded. MR system ( 1 ), in the MR signals from a measuring volume ( 21 . 22 . 24 . 25 . 27 - 29 . 49 ) of an examination object ( 3 ), wherein an adaptation of components of the MR system to the examination object located in the MR system is targeted in an adaptation volume (FIG. 3) before the MR signals are recorded from the measurement volume (FIG. 30 . 46 ), the MR system comprising: - a control unit ( 8th ), which is designed to record at least a first MR image of the examination subject in which the measurement volume and the adaptation volume are mapped, - a computing unit ( 12 ) configured to identify anatomical structures in the at least one first MR image and to automatically determine the adaptation volume in dependence on the identified anatomical structure.

Images