FIELD OF THE INVENTION

The invention relates to scanning imaging systems, in particular to a method for planning a scan geometry. BACKGROUND OF THE INVENTION

Planning scan geometries like single slices or multi-slices of 3D volumes in scanning imaging systems such as magnetic resonance imaging (MRI) systems is a time- consuming task for a technician. Systems for automatic scan planning exist, but adaptation to a user's preferences still requires manual interaction such as collecting and identifying a number of training scans and annotating a number of dedicated landmarks per scan.

WO 2008/129506 disloses an apparatus for generating a feature probability map arranged to be employed during automatic geometry planning for a medical diagnostic modality. The apparatus comprises a storage for a plurality of training images, each training image at least representing a same type of anatomical structure and having a scan geometry associated with it. The apparatus according to D2 may comprise a database storing user preferences regarding modifications of scan geometries. D2 does not disclose clinical context information comprising at least a series description describing the scan.

SUMMARY OF THE INVENTION

Various embodiments provide for a method for planning a scan geometry, scanning imaging system, and computer program product, as described by the subject matter of the independent claims. Advantageous embodiments are described in the dependent claims.

In one aspect, the invention relates to method for planning a scan geometry for imaging a predefined anatomy (or anatomical structure) using a scanning imaging system. The method comprises: building a database using data mining during a learning phase comprises determining clinical context information, anatomy information and scan geometries of previously performed scans; planning the scan geometry during a planning phase comprising performing a survey scan (e.g. 3D survey scan) of the anatomy using the scanning imaging system, thereby generating scan information indicative of the scan and anatomy; matching the generated scan information with the content of the database; based on the matching result, providing one or more scan geometries.

The learning phase of the method may automatically be performed e.g. on a predefined periodic basis. In another example, the learning and the planning phases may automatically be performed upon receiving a trigger signal e.g. from a user of the scanning imaging system.

The determined clinical context information (meta-information), anatomy information and scan geometries may be stored in the database. For that, the determined information may for example be clustered and structured and linked to other sources of available data. For example, the database may be populated with entries for each scan relating meta-information with parameters about the field of view and presence or location of anatomical landmarks.

The clinical context information refers to information specifying a "context" (e.g., a type of scanning imaging system, patient, intervenors, type of anatomy and/or type of disease, medical history, clinical questions) in which a scan planning, scanning and/or imaging process is performed. The piece of information (e.g. obtained from a DICOM database) of the clinical context information that relates to the anatomy may not be enough detailed. For that the present method may generate the anatomy information which may comprise certain relevant information such as a more precise description of the anatomy imaged using analysis of image data. DICOM stands for Digital Imaging and

Communications in Medicine.

The providing of the scan geometries may further comprise providing the scan geometries to other users and/or other scanning imaging systems (not only to same site/machine users but also to other hospitals or specialists). This may enable to exchange the learned scan geometries between different scanning imaging systems.

The term "scan geometry" refers to positional information that for example describes the field of view of a medical image relative to an anatomy or relative to the patient's coordinate system. The positional information may be expressed relative to the geometry of the survey scan or relative to anatomical landmarks that are automatically detected in the survey scan.

The term "scan" includes both scans including only a single 2D image frame acquisition pass as well as 3D scanning techniques wherein each individual scan is performed as a time series of individual acquisition passes which are equal in terms of parameters and contrasts. For example, in case the scanning imaging system is a magnetic resonance (MR) imaging system, the term "scan" may refer to a data acquisition sequence including applying a static magnetic field and a gradient magnetic field, transmitting an RF pulse, receiving an MRI signal, performing a predetermined processing on the NMR signal, and storing the processed MRI signal.

"Data mining" refers to the process of selecting, exploiting, modeling, etc., large amounts of data to detect, reveal, and quantify trends, patterns, and relationships within and among various data points and data attributes. The data points, for example, might be grouped or clustered on the basis of one or more common attributes. For example, the scan geometries and/or anatomy information may be obtained by image analysis of image data that are accessible by the data mining of the present method. In another example, the scan geometries and/or anatomy information may be readily obtained from databases.

The present method enables to provide a system which does not interfere with a technician's workflow and nevertheless is able to identify preferred scan geometries for specific situations. This may be done by automatic generation and population of the built database via retrospective analysis of available scan data from previously performed scans with respect to image content as well as meta-information (e.g. DICOM tags). Clinical image data is frequently transferred as DICOM format. The DICOM image format includes the image, and also a header containing meta-information that relates to the subject and the acquired images.

The automatic scan planning of the present method may enable a systematic approach for scanning anatomical structures. The present method may thus reduce to a minimum the user intervention in the scan planning process. This may provide an accurate and consistent scan planning across different scanning imaging systems although they are used or operated by different users. Using the data mining an optimal planning may be achieved since information from different data sources is considered.

The present method may be advantageous as it may enable a fully automated scan planning without interfering with a user's workflow.

The scan planning of the present method may be performed on the fly or dynamically without relying on preexisting defined scan plans for the anatomy to be imaged.

By means of the present method, scan geometry planning can be mad available with a limited burden to the user, as the standard workflow can be followed without a need for a separate training session of the user. According to one embodiment, the method further comprises: in response to receiving a modification of the provided scan geometry repeating the data mining using the modification, thereby updating the database. In other terms, the learning phase can be automatically or manually restarted as soon as a user manually adjusts the proposed planning or provided scan geometries.

This embodiment may provide an accurate and up to date content of the built database and may thus increase the performance of the scanning imaging system. This embodiment may be performed in a "Feedback mode" of the scanning imaging system where the database is updated by incorporating feedback of the user when adjusting proposed or provided scanned geometries. Another mode of the scanning imaging system may comprise "Bootstrap mode" according to which the system might inspect retrospectively available scans in a hospitals' picture archiving and communication system (PACS) and learn specific scan plans from existing data. The two modes may for example be part of the learning phase.

According to one embodiment, the repeating is performed only if the modification introduces a change in a position of the scan geometry that is higher than a predefined amount of change. In other terms, starting a new learning phase can be made dependent on the amount of change applied to the planning or scan geometry. If only a small change is applied, the learning phase might be skipped as the user is probably happy with the result, but just wants some minor refinement. This may save processing resources that would otherwise be required for updating the built database with data that may not be relevant as it comes from slightly changed data with respect to the content of the built database and which may have no impact on the scan planning.

According to one embodiment, the anatomy information comprises an anatomy field of view, wherein determining the scan geometry of a given scan of the previously performed scans during the learning phase comprises: identifying anatomy landmarks in the scan images associated with the given scan, and comparing the field of view with the identified landmarks for determining the scan geometry of the given scan.

The anatomy field of view may automatically be extracted by automated image analysis techniques. The anatomy field of view may for example be used for determining the scan geometry as exemplified in the following sequence of analysis steps:

A fully automatic detection of the scanned anatomy by probabilistic atlases, by the identification of anatomy-specific landmarks, by systems utilizing deep learning technology, potentially guided by the clinical context information. For example, the clinical context information may comprise a series description string that might be indicative of the scanned anatomy;

A detection of the scan geometry by means of the

variability of detected anatomical landmark positions with respect to the anatomy field-of-view, wherein the anatomical landmarks match the detected scanned anatomy, and/or

variability of the corners of the bounding box defining the anatomy field of view with respect to an existing 3D survey scan

The results of the detection of the scan geometry (point 2 above) might be clustered into a set of possible scan geometries per anatomy.

According to one embodiment, determining the scan geometry of a given scan comprises: identifying multiple anatomy landmarks in the scan images associated with respective multiple scans of the previously performed scans; determining which of the multiple anatomy landmarks match each other; and combining scan geometries of the multiple scans having matching landmarks for determining the scan geometry of the given scan. In other words, the results of the detection of the scan geometry (point 2 above) might be clustered into a set of possible scan geometries per anatomy which are combined if anatomical landmark positions of two or more scans do not differ more than a spatial threshold.

However, outliers may result from misplaced landmarks or from inconsistent input of the user. For that the present method may optionally further comprise: performing the learning or training with the identification of outlier geometries (e.g. by utilizing the Random sample consensus (RANSAC) algorithm) which may be removed to improve the consistency of the content of the database. Thus, the negative influence of outliers can be suppressed.

According to one embodiment, the building of the database further comprises determining scan parameters of the previously performed scans. The method further comprises based on the matching result providing the scan parameters in association with the scan geometries. Having the scan parameters may further increase the accuracy and consistency of the scan planning across different scanning imaging systems although they are used or operated by different users.

According to one embodiment, the clinical context information of each scan comprises at least one of: identifiers of the technician, the radiologist and the requesting physician of the scanning imaging system, identifier of a patient, indication of an anatomy, series description describing the scan, geometry information of the scan with respect to DICOM patient coordinates. The clinical context information may for example be used as part of the data mining to find and compare required patient information quickly and easily, supporting a better scan planning. Series Description string may be a DICOM tag, which defines the type of scan (e.g. Tl weighted vs T2 weighted MRI scan).

According to one embodiment, the scan information comprises multiple information elements, wherein the matching further comprises: comparing each element of the scan information with the database content, and scoring that element based on the comparison result; combining the scores for determining an overall score indicative of the level of matching of the scan information with scan entries of the database, ranking the scan entries of the database based on the overall score, wherein the providing comprises providing the one or more scan geometries based on the ranking. For example, the scan geometries may be provided in association with the respective ranking. This may be advantageous as it may enable an accurate selection of the scan geometry for the scan planning.

In one example, when planning the scan geometry of a new scan, the method may further comprise automatically computing a score for each potentially relevant scan geometry as available in the built database. The scan geometry having the highest score is chosen and is provided to the user for potential manual refinement.

According to one embodiment, the building of the database further comprises determining clinical and/or diagnostic information from Electronic Medical Records, EMR, associated with patients. This may support the built database by clinical and/or diagnostic information from a subject's EMR. The EMR may improve the accuracy of the matching by for example performing the matching based on the extended patient's information obtained from the EMR. This may further increase the accuracy and consistency of the scan planning across different scanning imaging systems although they are used or operated by different users. Diagnostic information from an EMR can also be applied to the planning phase, e.g. the system might suggest a scan geometry for other patients which suffer from disease ABC (as being mentioned in the EMR). In one example, the integration of recommendations of scan geometries from clinical guidelines may be performed along with the diagnostic information.

According to one embodiment, the matching comprises: matching the indicated anatomy with the anatomy information of existing scans in the database and/or matching a series description string associated with the survey scan to existing ones in the database and/or matching a patient identifier associated with the survey scan with patient identifiers in the database.

In another aspect, the invention relates to a computer program product comprising machine executable instructions for execution by a processor, wherein execution of the machine executable instructions causes the processor to the methods of any of the preceding embodiments.

In another aspect, the invention relates to a scanning imaging system for planning a scan geometry for imaging a predefined anatomy. The scanning imaging system comprises a memory containing machine executable instructions; and a processor for controlling the scanning imaging system, wherein execution of the machine executable instructions causes the processor to:

build a database using data mining during a learning phase comprising determining clinical context information, anatomy information and scan geometries of previously performed scans;

- lan the scan geometry during a planning phase comprising:

system.

It is understood that one or more of the aforementioned embodiments of the invention may be combined as long as the combined embodiments are not mutually exclusive.

BRIEF DESCRIPTION OF THE DRAWINGS

In the following preferred embodiments of the invention will be described, by way of example only, and with reference to the drawings in which:

Fig. 1 is a schematic diagram of a medical system,

Fig. 2 is a flowchart of a method for planning a scan geometry,

Fig. 3 shows a cross-sectional and functional view of an MRI system.

DETAILED DESCRIPTION OF THE EMBODIMENTS In the following, like numbered elements in the figures are either similar elements or perform an equivalent function. Elements which have been discussed previously will not necessarily be discussed in later figures if the function is equivalent.

Various structures, systems and devices are schematically depicted in the figures for purposes of explanation only and so as to not obscure the present invention with details that are well known to those skilled in the art. Nevertheless, the attached figures are included to describe and explain illustrative examples of the disclosed subject matter.

It is an object of the present disclosure to provide for a system for scan planning which is customized to the user, while easily fitting in the standard workflow. To achieve this object, the system makes use of two phases: a learning phase and a fully automatic planning phase. During the learning phase, the scanning imaging system may already have been used by the technician in his standard workflow. During the learning phase relevant data are collected for each planned scan (e.g. patient name, technician name, scan parameters, anatomy scanned). In addition, the scanned anatomy and the scan geometry may be determined and analyzed. In the fully automatic planning phase, the results of the learning phase are used to assist the user in planning a scan and proposing a (preferred) scan geometry to him.

The purpose of the present method is to learn user preferences not only according to the geometry (-adjustments) of scanned anatomy, but to relate it with meta- information. Therefore, the self-learning system is able to not only detect an anatomy in the survey image and propose scan geometry according to purely anatomical information, but also to identify the matching clinical context from other information, for which preferred scan plan might be specific.

For example, a QSM MR scan requires a tight field of view around the brain in order to reduce scan time. The scan geometry is tilted, axial, and parts of the skull and neck are excluded. In another example, a standard Tl -weighted brain scan might be required in sagittal orientation. In another example, an ASL scan requires untitled axial geometry but the field of view even excludes the top of the brain. In these examples, the context (e.g. or the clinical context information) might be identified or determined by the protocol description or sequence parameters that the user needs to enter, but there might be more parameters (clinical question or application such as Oncology, Neuro-Degeneration, Stroke etc. as available in the patient's EMR, among other information such as the physician, referrer etc.) which might have an effect on the (user-specific, customized) preferred view. The system or the method might work in two modes during the learning phase, both utilizing anatomical image content as well as meta-information:

"Feedback mode": Here, the system updates its internal knowledge base by incorporating feedback of the user when adjusting proposed scanned geometries.

- "Bootstrap mode": If not vendor-supplied knowledge base should be used, the system might also inspect retrospectively available scans in a hospitals PACS and learn specific scan plans from existing data.

Fig. 1 is a schematic diagram of a medical system 100. The medical system 100 comprises a control system 111 that is connected to a scanning imaging system 101. The control system 111 comprises a processor 103, a memory 107 each capable of

communicating with one or more components of the medical system 100. For example, components of the control system 111 are coupled to a bidirectional system bus 109.

It will be appreciated that the methods described herein are at least partly non- interactive, and automated by way of computerized systems. These methods can further be implemented in software 121, (including firmware), hardware, or a combination thereof. In exemplary embodiments, the methods described herein are implemented in software, as an executable program, and is executed by a special or general-purpose digital computer, such as a personal computer, workstation, minicomputer, or mainframe computer.

The processor 103 is a hardware device for executing software, particularly that stored in memory 107. The processor 103 can be any custom made or commercially available processor, a central processing unit (CPU), an auxiliary processor among several processors associated with the control system 111 , a semiconductor based microprocessor (in the form of a microchip or chip set), a macroprocessor, or generally any device for executing software instructions. The processor 103 may control the operation of the scanning imaging system 101.

The memory 107 can include any one or combination of volatile memory elements (e.g., random access memory (RAM, such as DRAM, SRAM, SDRAM, etc.)) and nonvolatile memory elements (e.g., ROM, erasable programmable read only memory

(EPROM), electronically erasable programmable read only memory (EEPROM),

programmable read only memory (PROM). Note that the memory 107 can have a distributed architecture, where various components are situated remote from one another, but can be accessed by the processor 103. Memory 107 may store an instruction or data related to at least one other constituent element of the medical system 100. The control system 11 1 may further comprise a display device 125 which displays characters and images and the like e.g. on a user interface 129. The display device 125 may be a touch screen display device.

The medical system 100 may further comprise a power supply 108 for powering the medical system 100. The power supply 108 may for example be a battery or an external source of power, such as electricity supplied by a standard AC outlet.

The scanning imaging system 101 may comprise at least one of MRI, CT and PET-CT imagers. The control system 111 and the scanning imaging system 101 may or may not be an integral part. In other terms, the control system 111 may or may not be external to the scanning imaging system 101.

The scanning imaging system 101 comprises components that may be controlled by the processor 103 in order to configure the scanning imaging system 101. The configuration of the scanning imaging system 101 may enable the operation of the scanning imaging system 101. The operation of the scanning imaging system 101 may for example be automatic. Fig. 3 shows example of components of the scanning imaging system 101 being an MRI system.

The connection between the control system 111 and the scanning imaging system 101 may for example comprise a BUS Ethernet connection, WAN connection, Internet connection etc.

In one example, the scanning imaging system 101 may be configured to provide output data such as images in response to a specified measurement.

The processor 103 may be adapted to receive information from the scanning imaging system 101 in a compatible digital form so that such information may be displayed on the display device 125. Such information may include operating parameters, alarm notifications, and other information related to the use, operation and function of the scanning imaging system 101.

The medical system 100 may be in communication via a network 130 with other scanning imaging systems 131 and/or databases 133. The network 130 comprises for example a wireless local area network (WLAN) connection, WAN (Wide Area Network) connection LAN (Local Area Network) connection or a combination thereof. The databases 133 may comprise information relates to patients, scanning imaging systems, anatomies, scan geometries, scan parameters, scans etc. The databases 133 may for example comprise an EMR database comprising patients' EMR, Radiology Information System database, medical image database, PACS, Hospital Information System database and/or other databases comparing data that can be used for planning a scan geometry.

Fig. 2 is a flowchart of a method for planning a scan geometry for imaging a predefined anatomy using a scanning imaging system e.g. 101. The anatomy may be part of a subject (e.g. 318 of Fig. 3) to be imaged as shown in Fig. 4. The anatomy may for example be a heart, brain, lunge etc. The imaging of the anatomy may result in images that can be used to perform further actions such as treatment delivery.

In step 201, a database BDB may be built using data mining. The building of the database BDB may be performed during a learning phase. The data mining may comprise determining clinical context information, anatomy information, scan parameters and scan geometries of previously performed scans. The determining may be performed by collecting and/or analysing data of multiple sources such as databases 133 and scanning imaging systems 131. The determined clinical context information may comprise meta-information such as DICOM tags.

The clinical context information may for example comprise identifiers of the technician, the radiologist and the requesting physician of the scanning imaging system 101. The clinical context information may further comprise identifiers of patients, scan

parameters, indications of an anatomy, series description strings describing scans, geometry information of a scan with respect to DICOM patient coordinates. An identifier may for example be a name or an ID.

The anatomy information may for example be obtained by automatically analysing the image data obtained by each scan. The image data may for example be part the information other than anatomical information obtained in step 201. The anatomy

information may comprise an anatomy field of view and anatomy- specific landmarks. For example, a fully automatic detection of a scanned anatomy may be performed by

probabilistic atlases (such as techniques for CT), by the identification of the anatomy-specific landmarks, and by emerging systems e.g. utilizing deep learning technologies.

The scan geometries may be read or retrieved from the data sources 131, 133 that are mined. In another example, the scan geometries may be derived from the images obtained by each scan and are stored in the data sources 131, 133. In this case, the scan geometry of a given scan of the previously performed scans during the learning phase may be determined by identifying anatomy landmarks in the scan images associated with the given scan and by comparing the field of view with the identified landmarks for determining the scan geometry of the given scan. The determined clinical context information, anatomy information, scan parameters and scan geometries of previously performed scans may be linked to each other via a common feature or common attribute such as an attribute indicating the scan or the patient or the anatomy.

The building of the database comprises the storing of the mined data in the database. The built database BDB stores the clinical context information, anatomy information, scan parameters and scan geometries of previously performed scans as records or entries of the database. For example, each record of the database BDB may represent an entity being indicative of a scan, patient and/or anatomy. For example, the database BDB may comprise groups of records representing respective common attribute such as an attribute indicating a scan, patient and anatomy. In another example, the database BDB may comprise records representing a single entity only such as the scan, patient and/or anatomy.

A record representing a scan may for example comprise values of attributes or properties of the scan. The attributes may for example comprise in a given record the scan geometry and scan parameters used for the scan represented by the given record, the identifier of the patient that has been scanned during the scan represented by the given record, anatomy information indicative of the anatomy scanned with the scan represented by the given record.

A record representing a patient may for example comprise values of attributes or properties of the patient. The attributes may for example comprise in a given record the identifier of the patient such as name, weight, age of the patient, scan geometries and scan parameters used for scans involving the patient represented by the given record, anatomy information indicative of the anatomy of the patient represented by the given record, clinical and/or diagnostic information from EMRs associated with the patient represented by the given record.

The database BDB may for example be part of the medical system 100 or may be separated from the medical system 100, where the medical system 100 has access to the database BDB.

In step 203, a survey scan of the anatomy may be performed using the scanning imaging system 101. The survey scan may be used to generate scan information indicative of the scan and the anatomy. The survey scan may for example be a three dimensional (3D) scan.

For example, the scanning imaging system 101 may comprise a MRI system. In this case, performing the survey scan may comprise acquiring survey magnetic resonance data from the anatomy by controlling the MRI system with survey pulse sequence data. The survey pulse sequence data comprises instructions for controlling the MRI system to acquire magnetic resonance data descriptive of a three-dimensional volume of the anatomy according to a survey scan geometry.

As described above with reference to step 201, the determined information may comprise multiple types of information e.g. patient type information, scan type information etc. The scan information of step 203 may for example comprise at least part of the multiple types of information. This may enable to match the scan information with the content of the database BDB.

In step 205, the generated scan information may be matched or compared with the content of the database. For example, the matching may comprise matching the indicated anatomy with the anatomy information of existing scans in the database BDB and/or matching a series description string associated with the survey scan to existing ones in the database BDB and/or matching a patient identifier associated with the survey scan with patient identifiers in the database BDB.

In another example, the scan information comprises multiple information elements. An information element may for example indicate the new anatomy to be scanned. In another example, an information element may comprise a series description of the scan to be performed. In another example, an information element may comprise patient data. Using the multiple information elements, the matching of step 205 be performed by comparing each element of the scan information with the entries or records of the database BDB. The comparison may comprise: for each record of the database BDB comparing the record attribute values with the information elements. Each comparison of each element of information may be scored based on the comparison result e.g. the score may indicate the level of similarity between the compared data. For example, for a given element compared with each record of the database BDB, as a score may be assigned to each record indicating the level of similarity or matching of the given element with the record. The scores may be combined for determining an overall score indicative of the level of matching of the scan information with each entry or record of the database BDB. The entries of the database BDB may be ranked based on the overall scores.

In step 207, one or more scan geometries may be provided based on the matching result of step 205. For example, if at least one of the performed comparisons is successful in that the compared objects are matching each other, the scan geometries may be provided. For example, the scan geometries may be displayed or indicated on the user interface 129.

The provision of the one or more scan geometries may be provided using the ranking of the above example. For example, for each record a scan geometry may be associated with or determined for the record. The scan geometry of the record may have a ranking equal to the ranking of the record. Step 207 may for example comprise providing the scan geometry having the highest ranking or providing the scan geometries in association with respective ranking such that the user may select among them.

Steps 203-207 may be performed during a planning phase. This planning phase may be automatically performed or executed.

Fig. 3 illustrates a magnetic resonance imaging system 300 as an example of the medical system 100. The magnetic resonance imaging system 300 comprises a magnet 304. The magnet 304 is a superconducting cylindrical type magnet with a bore 306 in it. The use of different types of magnets is also possible; for instance, it is also possible to use both a split cylindrical magnet and a so called open magnet. A split cylindrical magnet is similar to a standard cylindrical magnet, except that the cryostat has been split into two sections to allow access to the iso-plane of the magnet. Such magnets may for instance be used in conjunction with charged particle beam therapy. An open magnet has two magnet sections, one above the other with a space in-between that is large enough to receive a subject 318 to be imaged, the arrangement of the two sections area similar to that of a Helmholtz coil. Inside the cryostat of the cylindrical magnet there is a collection of superconducting coils. Within the bore 306 of the cylindrical magnet 304 there is an imaging zone or volume or anatomy 308 where the magnetic field is strong and uniform enough to perform magnetic resonance imaging.

Within the bore 306 of the magnet there is also a set of magnetic field gradient coils 310 which is used during acquisition of magnetic resonance data to spatially encode magnetic spins of a target volume within the imaging volume or examination volume 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, magnetic field gradient coils 310 contain three separate sets of coils for the encoding in three orthogonal spatial directions. A magnetic field gradient power supply supplies 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 may be ramped or pulsed. MRI system 300 further comprises an RF coil 314 at the subject 318 and adjacent to the examination volume 308 for generating RF excitation pulses. The RF coil 314 may include for example a set of surface coils or other specialized RF coils. The RF coil 314 may be used alternately for transmission of RF pulses as well as for reception of magnetic resonance signals e.g., the RF coil 314 may be implemented as a transmit array coil comprising a plurality of RF transmit coils. The RF coil 314 is connected to one or more RF amplifiers 315.

The magnetic field gradient coil power supply 312 and the RF amplifier 315 are connected to a hardware interface of control system 111. The memory 107 of control system 11 1 may for example comprise a control module. The control module contains computer-executable code which enables the processor 103 to control the operation and function of the magnetic resonance imaging system 300. It also enables the basic operations of the magnetic resonance imaging system 300 such as the acquisition of magnetic resonance data.

Planning MRI scan geometries (e.g. in 3D) is a time-consuming task for a technician which requires manual interaction and is subject to errors. The present disclosure discloses a fully automated system supporting planning of user-preferred scan geometries by retrospective analysis of existing scans and by generating a database of scan geometry presets. The analysis may consist of meta-information e.g. contained in a DICOM image (name of technician, patient, acquisition parameters, series description / scan name, contrast) harvested via big data mining technology as well as anatomical information (or anatomy information) which is extracted from the image's pixel data by automatic analysis algorithms. This database is then used for a new 3D survey (using MRI system 300) with a few available data terms such as patient name, technician's name, series description in order to

automatically infer a list of potential scan geometries previously used in similar situations, and presenting them to the user. This approach might optionally be supplemented by data from an EMR.

In the following further examples describing the planning phase are provided.

For a new to-be-acquired scan, series description, patient / technician's name, the 3D survey scan as well as sequence parameters may be available. Based on this information, a matching score with respect to previously acquired (or performed) scans can be extracted from the database BDB, consisting of individual terms, such as: an indication of the similarity of the new anatomy to be scanned (by automated analysis of the 3D survey scan) to the anatomy information of existing scans in the database BDB.

a similarity of a series description string to existing ones.

- an indication of previous scans of the same patient name and/or patient ID.

By combining different similarity terms (of which three specific examples are given above), a score might be computed for previously acquired scans with similar sequence parameters in the database BDB. These geometries might be presented to the user (e.g. as a bounding box overlaid on the 3D survey of the new scan) in order to let him choose a preferred scan geometry for final interactive modification. Alternatively, only the highest ranked scan geometry is presented as default, and the user might be enabled to switch to geometries which have a slightly lower rank.

In one example, the scan geometries of all scans comprised in an exam card, are proposed and refined in a batch mode, so that all of these scans can be carried out without interruption. Besides a text notice on the screen or display 125, the scan type to be planned can be indicated by application of a gray value intensity look-up-table, which roughly mimics the appearance of the scan to be planned. The exam card may list the activities in their order to be carried out in the course of the diagnostic imaging of a patient to be examined.

In one example, the estimated scan time of the automatic scan planning can be directly displayed on screen or display 125. When the user adjusts geometry, the scan time is adjusted simultaneously and possibly compared to the automatic-planning scan time.

In another example, the plan time reduction (if any) can be shown to the user after a certain time of utilization using a message such as "Congrats! You have increased your planning time efficiency by 35%".

In another example, the sequence parameters for a new scan may not have been made available by the technician, but may be suggested by the present method, similar to scored scan geometries, based on matching the anatomy information and the series description string for the same subject with content of the build database. LIST OF REFERENCE NUMERALS

100 medical system

101 scanning imaging system

103 processor

107 memory

108 power supply

109 bus

111 control system

121 software

125 display

129 user interface

201-207 method steps

300 magnetic resonance imaging system

304 magnet

306 bore of magnet

308 imaging zone

310 magnetic field gradient coils

312 magnetic field gradient coil power supply

314 radio-frequency coil

315 RF amplifier

318 subject.