Determination of the energy consumption or load profile of medical devices

Abstract

Energy data and load profiles of a medical device should be obtained in a simple manner. For this purpose, a method for determining an energy consumption or a load profile (63) of a medical device is provided. The method is based on a load or energy consumption model (61). A virtual parameter (62) of the load or energy consumption model (61) is set, wherein the virtual parameter (62) corresponds to at least one real parameter of the medical device. The energy consumption or the load profile (63) of the medical device is determined as a function of the virtual parameter (62).

Description

The present invention relates to a method for determining an energy consumption or load profile of a medical device. In addition, the present invention relates to a corresponding calculation device and a computer program.Medical devices or devices can be different imaging or therapeutic devices. In particular, the invention relates to medical devices that have a high demand for energy or power.The increased energy costs, in particular in Europe, have focused attention on the MRI apparatuses (magnetic resonance tomography) used in the radiologic departments. Because of their high energy consumption, which may be about 7.5% of the total energy consumption of a hospital, the radiologic departments are responsible for a substantial portion of the greenhouse gas emissions caused by the facility. In radiology, MRT devices with an annual consumption of up to 100 MWh consume significantly more power than all other devices. This high energy consumption is primarily due to magnet cooling followed by gradient generation, both of which make up 75% of the total consumption. The reduction in the energy consumption of MRTs therefore largely depends on the optimization of the cooling process and the shortening of the active phases of the gradient. However, when optimizing a program step for energy efficiency, further components such as the RFPA (radio-frequency power amplifier) must also be taken into account.Magnetic resonance tomography apparatuses are imaging devices which, for imaging an object under examination, align nuclear spins of the object under examination with a strong external magnetic field and excite them by means of an alternating magnetic field for precession about this alignment. The precession or return of the spins from this excited to a state of lower energy in turn generates an alternating magnetic field in response which is received via antennas.With the aid of magnetic gradient fields, spatial coding is imposed on the signals, which then enables an assignment of the received signal to a volume element. The received signal is then evaluated and a three-dimensional imaging representation of the examination object is provided.High powers in the range of kilowatts are required for generating the magnetic fields and high-frequency signals, and a corresponding power must also be dissipated again by a cooling system.In X-ray systems such as e.g. computer tomography systems, high electrical powers are likewise required for generating the X-rays, and also cooling of the targets in the X-ray sources. The same applies to radiation therapy devices with corresponding accelerators.Due to the complexity of the technology, it is not transparent to the user how much energy is consumed at different parameter settings. This may result in waste of energy in suboptimal parameter setting.At present, there is no solution to this problem because the power consumption of an MRI scanner cannot be optimized. Neither for a user nor for the OEMs.Not only the energy consumption but also the load profile can present a problem. There are certainly MRT operators which cannot use their device fully on account of limited connection power and must correspondingly adapt parameters in order not to exceed specific power limits.The object of the present invention is to more comprehensively assist the user of a medical device in reducing the energy consumption or optimizing the load profile.According to the invention, this object is achieved by a method and a calculation device according to the independent claims. In addition, a corresponding computer program or computer program product is provided.Advantageous further developments of the invention are evident from the dependent claims.According to the present invention, a method for determining an energy consumption or a load profile of a medical device is accordingly provided. As mentioned at the beginning, the medical device is, for example, an MRI system, a CT system or an irradiation system, each of which exhibits a relatively high energy consumption or peak load. Determining the energy consumption or load profile does not include measuring the energy consumption or load profile. Rather, the determination is intended to include, for example, an estimation, whereby, for example, a virtual measurement method can be realized. Under certain circumstances, both the energy consumption and the load profile can be determined.In a method step, a load model and / or energy consumption model is provided for the medical device. The load or energy consumption model can be used to model or simulate the energy consumption or the load profile of the medical device. For example, a specific energy consumption or the load profile can be deduced with the aid of specific input variables of the load or energy consumption model. The energy consumption model or load profile may be based on analytic algorithms, look-up tables, neural networks, and the like.In a further step, a virtual parameter of the load or energy consumption model is set, wherein the virtual parameter corresponds to at least one real parameter of the medical device. The load or energy consumption model can thus be, for example, a digital twin of the medical device, which is supplemented with a virtual power and / or energy measurement device. A real parameter of the medical device can be, for example, the gradient strength of an MRI system. The corresponding virtual parameter then also relates to the gradient strength in the model. The effect of the change in the gradient strength in the real MRI system on the energy consumption is that the change in the virtual gradient strength in the load or energy consumption model also has an effect on the estimated energy consumption or the load profile. Depending on how the virtual parameter is thus set, a corresponding virtual or estimated energy consumption or load profile for the respective procedure or operation of the medical device concludes therefrom. The energy consumption or the load profile of the medical device is therefore determined as a function of the virtual parameter. In other words, the load or energy consumption model simulates the medical device's operation set with the virtual parameter and the resulting energy consumption or load profile.The energy consumption or load profile may be determined as a value (e.g., total or peak or average load for a completed operation, such as acquisition of an image sequence) or as a time profile. This curve represents the energy consumption per unit time and thus a power. The load profile can be, in particular, a power profile or an electrical current profile.Advantageously, it is thus possible to estimate or predict the energy consumption or the load profile of a medical device without a specific measuring device. Based on such prediction, power and / or energy optimization may be initiated.In one exemplary embodiment, it is provided that the virtual parameter is one of a plurality of virtual parameters of the energy consumption model, each of the virtual parameters is set and the power and / or energy consumption of the medical device is simulated as a function of all virtual parameters. This means that the energy consumption model has a plurality of input parameters, on the basis of which the energy consumption of the medical device is calculated. If the medical device is, for example, an MRI system, the parameters can relate, for example, to the field of view (field of view), such as gradient strength, saturation, radio-frequency energy and the like. All these parameters ultimately influence the energy consumption and / or the load profile of the MRI installation.The plurality of virtual parameters can be determined from a header of an image data set. If the image data record has, for example, the standardized DICOM format, the recording parameters are stored in the respective header of an image. The header data are real parameters, which however can be used as virtual parameters for the energy consumption model. Thus, the real parameters can either serve to subsequently estimate the energy consumption or the load profile for the recording of the respective image with the aid of the load or energy consumption model. Alternatively, the real parameters can also be used to generate predictions about the energy consumption or the load profile of new images. For example, a header parameter set can be varied slightly for prediction in order to determine a corresponding energy consumption or the load profile.According to an alternative exemplary embodiment, the plurality of virtual parameters are determined from a logfile of the medical device. This means that, when the image acquisition has taken place, the parameters are logged in a log file. The log data can then in turn be the starting point for a prediction of the energy consumption or the load profile for a virtual recording. Values of a real image acquisition are therefore also used here for setting the virtual parameters of the load or energy consumption model.In a further exemplary embodiment, it is provided that the energy consumption and / or the load profile of the medical device is graphically represented on a graphical user interface together with the virtual parameter and / or as a function of the virtual parameter. The graphical user interface can be, for example, a dashboard, with which, for example, functional relationships between one or more parameters and the energy consumption or the load profile can be graphically displayed. For example, the energy consumption and the load profile can be reproduced as a function or bar graph depending on a specific parameter.In another exemplary embodiment, it is provided that the medical device is designed to carry out a plurality of examination steps and the load or energy consumption model is designed to determine a respective energy consumption or a load profile for corresponding virtual examination steps, each virtual examination step being configured with a respective virtual parameter. If, for example, an examination sequence with several examination steps is to be carried out, the load or energy consumption model can be able to simulate these several steps in succession with respect to the energy consumption or the load profile. If necessary, a sum of the total energy consumption of all examination steps is determined by the energy consumption model. Alternatively, several examination steps could also be carried out in parallel. In this case, too, the energy consumption model could be designed to determine the energy consumption of the parallel examination steps and, if appropriate, to calculate a total energy consumption. This allows the user to directly influence the individual examination steps with regard to the energy consumption.In general, the load or energy consumption model can also be used to automatically optimize the energy consumption or the load profile. For this purpose, for example, an optimization algorithm could determine a minimum energy consumption and / or a maximum possible load profile by correspondingly varying the virtual parameters.In a further exemplary embodiment, the medical device contains an imaging device with which a recording sequence can be detected, wherein individual pieces of a corresponding virtual recording sequence are set with the plurality of virtual parameters, and a respective individual energy consumption or the load profile of each individual piece of the recording sequence and / or a total energy consumption of the recording sequence is calculated with the load or energy consumption model. It is thus intended, for example, to obtain a recording sequence with a plurality of images. The load or energy consumption model is in this case capable of providing for each recording at least one separate virtual parameter which has an influence on the recording quality of the respective image. With the load or energy consumption model, a corresponding energy consumption or the load profile can then be established for each individual recording of the virtual recording sequence. Thus, individual recordings, but also the entire recording sequence, can be optimized with regard to energy consumption and load profile.According to one exemplary embodiment, it is provided that the setting of the virtual parameter of the load or energy consumption model takes place with the aid of an input interface. For example, the input interface is an HMI (Human Machine Interface) interface. A user can thus set or change one or more virtual parameters of the load or energy consumption model, for example manually. For this purpose, buttons for numerical inputs and / or one or more adjusting wheels for setting a respective virtual parameter can be provided on the input interface.A further exemplary embodiment can be based on the energy consumption and / or the load profile being graphically depicted and interactively changed by inputting a value of the virtual parameter via the input interface. If, for example, the energy consumption or the load profile is thus represented by means of a bar diagram, the bar height can vary according to the set virtual parameter. Interactive display means, in particular, that the change of the graphic takes place in real time with the change of the virtual parameter. The user thus sees directly in time how a change in the virtual parameter has an effect on the energy consumption or the load profile.According to a further exemplary embodiment, provision is made for radio-frequency pulses and gradient waveforms of a sequence protocol of an MRI apparatus to be set using the virtual parameters. Radio frequency pulses and gradient waveforms significantly affect the energy consumption and / or load profile of an MRI plant. It is therefore expedient to model and optionally optimize these parameters in particular with regard to the energy consumption and / or the load profile.According to another embodiment, the virtual parameter is readjusted depending on the energy consumption or the load profile. If necessary, a plurality of virtual parameters are also readjusted as a function of the (respective) energy consumption and / or the load profile. This means that a feedback loop is provided, because first the energy consumption or the load profile is calculated on the basis of an output value of the virtual parameter and this energy consumption and / or the load profile are then used to change the virtual parameter again. Such feedback may be used to optimize power consumption and / or load profile. This optimization can be carried out automatically by an algorithm. If necessary, the energy consumption and / or the load profile is also determined as a function of two different virtual parameters, and a compromise is found between the two parameters with regard to the energy consumption and / or the load profile. Thus, for example, gradient strength and saturation can be balanced with respect to one another in order to optimize the energy consumption.According to the invention, a method for training a load model or energy consumption model for the medical device is also provided byproviding setting data (headers, logfiles) for setting the medical device,providing load or energy consumption data, which are measured on the medical device as a function of the setting data,training a neural network with the setting data as input data and the load or energy consumption data as output data.A neural network is thus used for the load or energy consumption model, which neural network can learn the energy consumption and the load profile at respective settings from older headers and / or logfiles. Thus, load or energy data can be obtained very reliably from the parameter settings. In particular, the load or energy consumption model can be trained in such a way that it carries out one of the aforementioned methods.The above-mentioned object is also achieved according to the invention by a calculation device for determining an energy consumption or the load profile of a medical device witha data processing device in which a load or energy consumption model for the medical device is implemented, andan interface for setting a virtual parameter of the load or energy consumption model, wherein the virtual parameter corresponds to at least one real parameter of the medical device, whereinthe data processing device is designed to determine (simulate) the energy consumption or the load profile (as values (average, absolute value, variance, standard deviation, etc. or curve) of the medical device as a function of the virtual parameter.The data processing device can be, for example, a computer or another computing unit having one or more processors and one or more storage elements. The interface may be an automatic data input data interface or a human machine interface for manual data input.In addition, the calculation device can be part of a control unit of the medical device. Thus, one or more set virtual parameters can be used to control the real medical device.Furthermore, according to the invention, a computer program or computer program product can also be provided which has instructions which, when executed in the above calculation device, cause the latter to execute a method as has been presented above.For use cases or application situations which can arise in the method and which are not explicitly described here, provision can be made for an error message and / or a request for inputting a user feedback to be output and / or for a default setting and / or a predetermined initial state to be set according to the method.Regardless of the grammatical sex of a certain term, individuals with male, female or other sex identity are included.The present invention will now be explained in more detail with reference to the accompanying drawings, in which: FIG. 1 is a schematic illustration of an exemplary medical device according to the invention; and FIG. 2 shows a schematic diagram of the sequence of an exemplary embodiment of a method according to the invention.The exemplary embodiments described in more detail below represent preferred embodiments of the present invention.FIG. 1 shows a schematic representation of an embodiment of a magnetic resonance tomography system 1 as an example medical device. In addition, other devices with considerable requirements for electrical energy or power, cooling or other operating means are also conceivable as a medical device, such as, for example, linear accelerators for radiotherapy.The magnet unit 10 has a field magnet 11 which generates a static magnetic field B0for aligning nuclear spins of samples or of the patient 100 in a recording region. The receiving region is characterized by an extremely homogeneous static magnetic field B0, wherein the homogeneity relates in particular to the magnetic field strength or the amount. The receiving region is virtually spherical and is arranged in a patient tunnel 16 which extends through the magnet unit 10 in a longitudinal direction 2. A patient bed 30 is movable in the patient tunnel 16 by the displacement unit 36. Typically, the field magnet 11 is a superconducting magnet that can provide magnetic fields having a magnetic flux density of up to 3T, even higher in recent devices. However, permanent magnets or electromagnets with normally conducting coils can also be used for lower magnetic field strengths.The superconducting magnet requires a cooling unit with high power consumption and equally high waste heat and thus cooling requirement for maintaining the low temperatures of the superconducting magnet coils.Furthermore, the magnet unit 10 has gradient coils 12 which are designed to superimpose temporally and spatially variable magnetic fields in three spatial directions on the magnetic field B0for the spatial differentiation of the captured imaging regions in the examination volume. The gradient coils 12 are usually coils made of normally conducting wires which can generate fields in the examination volume that are orthogonal to one another.The resistive gradient coils 12 are also controlled by a gradient controller 21 with very high currents and have a corresponding requirement for electrical energy, power and cooling requirements for the waste heat.The magnet unit 10 further comprises a body coil 14 which is designed to emit a radio-frequency signal fed via a signal line into the examination volume and to receive resonance signals emitted by the patient 100 and to emit them via a signal line.A control unit 20 supplies the magnet unit 10 with the various signals for the gradient coils 12 and the body coil 14 and evaluates the received signals.Thus, the control unit 20 has the gradient control 21 which is designed to supply the gradient coils 12 via supply lines with variable currents which provide the desired gradient fields in the examination volume in a temporally coordinated manner.Furthermore, the control unit 20 has a radio-frequency unit 22, which is designed to generate a radio-frequency pulse with a predefined temporal profile, amplitude and spectral power distribution for exciting a magnetic resonance of the nuclear spins in the patient 100. In this case, pulse powers in the range of kilowatts can be achieved. The excitation signals can be emitted into the patient 100 via the body coil 14 or also via a local transmitting antenna.A device controller 23 communicates via a signal bus 25 with the gradient controller 21 and the radio-frequency unit 22.For receiving the magnetic resonance signal, a local coil 50 according to the invention is arranged on the patient 100 in the patient tunnel 16 in order to acquire magnetic resonance signals from an examination region in the immediate vicinity with the largest possible signal-to-noise ratio. The local coil 50 is in signal connection via a connecting line 33 with a receiver in the high-frequency unit 22.Furthermore, the magnetic resonance tomography system 1 optionally has an interface, for example to a data network, via which the device controller 23 can communicate with a supply controller of a supply device, for example send messages and receive instructions.According to the invention, a virtual measuring device is now made available, with which the energy consumption of the medical device can be determined or estimated. This can serve, for example, to optimize the energy consumption, for example by finding a compromise between the quality of the images to be recorded and the associated energy consumption. If necessary, the type of image recording can also be varied in order to optimize the energy consumption and / or the load profile. Thus, for example, weights, relaxation times (T1or T2) et ceteracan be changed as virtual parameters for the energy consumption estimate.A load model and / or energy consumption model can therefore be created, for example, on the basis of a statistical analysis. This load or energy consumption model may be based on look-up tables, analytic functions, machine learning, and so forth. The data basis of the load or energy consumption model can be, for example, DICOM (standard format for medical images), log files, measured cooling temperature, parameter settings, et cetera.The load or energy consumption model can be used to determine the energy consumption or the load profile of program steps with characteristic parameter settings. The corresponding energy consumption result can be presented to the user in different ways.For example, a dashboard may be used for retrospective display. This means that the energy consumption and the load profile of an already performed study is calculated or estimated. Input variables of the load and / or energy consumption model can be log files (log files) of the scanner or, for example, DICOM headers (headers).Alternatively, the load or energy consumption model can also be used for a predictive or prospective representation of the expected energy consumption or the load profile of a program step and / or a complete examination strategy. For example, the energy consumption to be expected and / or the load profile can be displayed in an operator interface for the comfortable creation of MRI examination sequences. This consumption could be represented specifically for individual examination steps or as a total consumption or a statistical description of the load profile. In addition, the energy consumption to be expected and / or the load profile can also be represented in a parameter list of an examination sequence. A user may modify his examination strategy to achieve reduced energy consumption and lower peak load.For this purpose, the load or energy consumption model can use, for example, a sequence protocol with planned high-frequency pulses and gradient waveforms in order to represent, for example, energy contents and / or the load profile of individual examination steps.In a further alternative, the calculated energy consumption and / or the load profile of a program step can be used for retrospective display. Similar to the current calculation of SED (Specific Energy Dose) and SAR (Specific Absorption Rate), the current energy consumption and / or the load profile in an MRT scan can also be calculated. With such a retroactive display, in any case for the next examination step, a compromise can be made between information content in the image on the one hand and energy consumption and / or load profile on the other hand. The energy consumption calculated with the aid of the load or energy consumption model and displayed retrospectively and / or the load profile can, however, also be used, for example, for corrections on the system.Alternatively to calculating the energy consumption or load profile based on the scheduled parameters, the energy consumption and to a limited extent the load profile could also be calculated from the cooling performance of the system (cooling performance based load or energy consumption model). For this purpose, a system-specific conversion factor between cooling power and electrical power must be determined by measurements. The cooling power can be calculated on the basis of the temperature difference between water inlet and outlet.Visualization of the energy consumption or load profile may be done with absolute numbers (e.g., kWh) as well as relative numbers (e.g., percentage relative to a reference).FIG. 2 schematically shows the sequence of an exemplary embodiment of the method according to the invention in order to determine or estimate the energy consumption and / or the load profile of a medical device. For this purpose, a load and / or energy consumption model 61 is provided for the medical device, e.g. the magnetic resonance tomography system 1, and one or more virtual parameters 62 of the load or energy consumption model 61 are set. The virtual parameters preferably correspond to real parameters for setting the real medical device. Under certain circumstances, however, a virtual parameter also corresponds to a real parameter set. This means that a real parameter set, such as stored in the header of an MRT image, may correspond to a single virtual parameter. In a further exemplary embodiment, it is also possible, for example, to see a subset of an image header, that is to say, for example, two or three parameters, as a single virtual parameter.In a further step of the method according to the invention, the energy consumption and / or the load profile 63 of the medical device is determined as a function of the one or more virtual parameters 62 or the plurality of virtual parameters 62. The energy consumption 63 is determined, for example, as a single value or value profile over time. As a value profile, it corresponds to a power. The load profile can be determined, for example, as a current or power profile (time profile).The energy consumption 63 is preferably displayed on an operating surface 64. The user interface 64 may be part of an input interface. Via the user interface 64, one or more of the virtual parameters 62 can be changed or reset. This may be necessary within the scope of an optimization or a compromise.With the virtual measuring device according to the invention, which is based on the load and / or energy consumption model, a conventional electricity meter on the scanner can be dispensed with. This makes it possible to save corresponding investment costs for electricity meters. In addition, it is possible with the virtual measuring device not only to carry out analyses in succession, but also to represent an estimated energy consumption or the load profile prospectively. This allows the user to optimize individual program steps and ultimately the entire scanner fleet with regard to energy consumption and load characteristics. The estimated power consumption and load profile could also be used, for example, to control an infrastructure such that any load spikes are avoided.

Claims

Method for determining an energy consumption or load profile (63) of a medical device by - providing a load or energy consumption model (61) for the medical device, - setting a virtual parameter (62) of the load or energy consumption model (61), wherein the virtual parameter (62) corresponds to at least one real parameter of the medical device, - determining the energy consumption or the load profile (63) of the medical device as a function of the virtual parameter (62).The method of claim 1, wherein the virtual parameter (62) is one of a plurality of virtual parameters (62) of the load or energy consumption model (61), each of the virtual parameters (62) is set, and the energy consumption or load profile (63) of the medical device is simulated depending on all the virtual parameters (62).The method of claim 2, wherein the plurality of virtual parameters (62) are determined from a header of an image data set.The method of claim 2, wherein the plurality of virtual parameters (62) are determined from a logfile of the medical device.Method according to any of the preceding claims, wherein the energy consumption or the load profile (63) of the medical device is graphically represented on a graphical user interface together with the virtual parameter (62) and / or depending on the virtual parameter (62).The method according to any of the preceding claims, wherein the medical device is configured to perform a plurality of examination steps, and the load or energy consumption model (61) is configured to determine a respective energy consumption or load profile (63) for corresponding virtual examination steps, each virtual examination step being configured with a respective virtual parameter (62).Method according to Claims 2 and 6, wherein the medical device contains an imaging device with which a recording sequence can be detected, individual pieces of a corresponding virtual recording sequence are set using the plurality of virtual parameters (62), and a respective individual energy consumption of each individual piece of the recording sequence and / or a total energy consumption of the recording sequence is calculated using the load or energy consumption model (61).Method according to one of the preceding claims, wherein the setting of the virtual parameter (62) of the load or energy consumption model (61) is effected with the aid of an input interface.The method of claim 8, wherein the energy consumption and / or load profile (63) is graphically depicted and interactively altered by entering a value of the virtual parameter (62) via the input interface.Method according to one of Claims 2 to 9, wherein the virtual parameters (62) are used to set planned radio-frequency pulses and gradient waveforms of a sequence protocol of an MRI apparatus (1).Method according to one of the preceding claims, wherein the virtual parameter (62) is readjusted as a function of the energy consumption and / or the load profile (63).Method for training a load or energy consumption model (61) for the medical device by - providing setting data for setting the medical device, - providing energy consumption data or load profiles which are measured on the medical device as a function of the setting data, - training a neural network with the setting data and the energy consumption data or load profiles.The method of any one of claims 1 to 11, wherein the load or energy consumption model (61) is trained with the method of claim 12.Calculation device for determining an energy consumption or a load profile (63) of a medical device, having - a data processing device in which a load model or energy consumption model (61) for the medical device is implemented, and - an interface for setting a virtual parameter (62) of the load or energy consumption model (61), wherein the virtual parameter (62) corresponds to at least one real parameter of the medical device, wherein - the data processing device is designed to determine the energy consumption or the load profile (63) of the medical device as a function of the virtual parameter (62).A computer program or computer program product comprising instructions which, when executed in an apparatus according to claim 14, cause the apparatus to carry out a method according to any one of claims 1 to 11.

Images