Support device for issuing a recommendation for action to a surgeon

Abstract

Support device (70) for issuing a recommendation for action (110) to an operator (64) who is performing a surgical procedure on an object (12), with - an object identification unit (66) for identifying the object, - an operator identification unit (68) for identifying the operator (64), - a first storage unit (72) for storing object property data describing the identified object (12), - a second storage unit (74) for storing operator property data describing the identified operator (64), - a specification unit (76) for specifying the intervention to be carried out on the object (12), - a preparation unit (78) in communication with the specification unit (76), which is trained to obtain and store intervention-related operational data (104), and - is trained in an evaluation unit (80) to evaluate at least the operator property data, the object property data and the intervention-related operation data (104) and, depending on the evaluation, to determine and issue the recommendation for action (110), characterized by that the support device includes an operator detection unit (90) for detecting operator characteristics, wherein the operator detection unit (90) is designed to detect at least one position or orientation of the operator (64) in relation to an intervention site on the object (12) during the intervention.

Description

[0001] The invention relates to a support device for providing a recommendation for action to a surgeon performing a surgical procedure on an object. Furthermore, the invention also relates to a method for providing a recommendation for action to a surgeon performing a surgical procedure on an object. Finally, the invention also relates to a computer program product.

[0002] Support devices, methods for their use, and computer programs of this type, as well as uses of such methods to assist a surgeon during a procedure on an object, are extensively known in the prior art, so that no separate written reference is required. Such support devices, methods, and computer programs serve, among other things, to assist a surgeon performing a procedure on an object in order to at least partially determine and / or at least partially modify the object's structure in the area of ​​the procedure. Such procedures are frequently used in materials testing, but also in the field of medical procedures involving biological material, living beings, especially patients, or the like.The object can therefore be, for example, an item resulting from an industrial manufacturing process, but also a mining product, the body of a living being, especially a patient, or the like.

[0003] Such procedures are now frequently performed using diagnostic equipment, particularly imaging devices such as magnetic resonance imaging (MRI) scanners, computed tomography (CT) scanners, or similar instruments. This allows for the preparation, immediate monitoring, and optimization of the procedure, minimizing its duration and / or ensuring its reliability. Typically, at least one area of ​​the object where the procedure is to take place is scanned. However, this can result in the area being difficult for the surgeon to access, making the procedure more challenging and time-consuming. Furthermore, it increases the risk of unintentionally intervening in an area where intervention is not intended.In living beings, this can lead to dangerous injuries, for example.

[0004] In medical technology, for example, magnetic resonance imaging (MRI), also known as magnetic resonance tomography (MRT), is used. In this process, magnetic field pulses are generated by a magnetic field source and applied to the object. The magnetic field can have magnetic flux densities in the range of several Tesla. Due to interactions between the object and the magnetic field at the atomic level, so-called MR signals can be triggered in the object, primarily by aligning proton spins. The MR signals are generated by the application of high-frequency magnetic pulses. These signals can be detected with a local coil, preferably positioned on the object within the area being examined, for example, on a surface of the object, and then evaluated by a processing unit of the magnetic resonance imaging scanner.The evaluation may result in, for example, an image relating to the structure of the object.

[0005] The magnetic field strengths occurring during normal operation, that is, in particular during the execution of the investigation on the object, usually result in very high values ​​with respect to the magnetic flux density, for example in the range of about 0.2 T to about 7 T or the like.

[0006] Even though generic support devices, procedures, and computer programs have proven effective, disadvantages remain. Currently, the operator is only supported during the procedure insofar as their actions are guided solely from the object's perspective. The state of the art does not consider that the operator, as an individual, can also influence the procedure. This can lead to the provision of instructions or recommendations that the operator cannot implement, or can only implement inadequately. Consequently, the reliability and / or duration of the procedure can be negatively impacted.

[0007] From publication EP 4104787 A1, a method for generating and updating a three-dimensional representation of a surgical site based on image data from an imaging system is known. Furthermore, a surgical hub for use with a surgical system during a surgical procedure performed in an operating room is known, wherein the surgical hub comprises a control circuit configured to provide communicative coupling with at least one surgical device in the operating room; to receive visualization data from a camera; and to determine possible surgical sites based on the visualization data. to obtain user properties and to determine a device placement that is selected from the potential surgical sites, based on at least the user characteristics.

[0008] German patent application DE 102022104138 A1 relates to a medical imaging device comprising a spatially and spectrally resolving image acquisition unit, which includes at least one optical system and at least one image acquisition sensor coupled to the optical system, configured to generate image data of an object area containing spatial and spectral information. An image analysis unit is configured to perform an analysis of the image data based on spatial and spectral information, the analysis comprising at least one evaluation parameter that relates to a sub-area of ​​the object area.An evaluation unit is configured to generate an evaluation of an attribute of the object sub-area relevant to the diagnostic and / or therapeutic action, based on the evaluation parameter and in accordance with the analysis and information regarding a diagnostic and / or therapeutic action to be performed and / or already performed. An output generation unit is configured to generate an output based on the evaluation. The publication also pertains to a medical system with a medical imaging device and related procedures.

[0009] The invention proposes a support device, a method and a computer program product according to the independent claims as a solution.

[0010] With regard to a support device for issuing a recommendation for action to an operator performing a surgical procedure on an object, the invention particularly proposes that the support device comprise an object identification unit for identifying the object, an operator identification unit for identifying the operator, a first storage unit for storing object property data describing the identified project, a second storage unit for storing operator property data describing the identified operator, a specification unit for specifying the procedure to be performed on the object, a preparation unit communicating with the specification unit and configured to obtain and store procedure-related operational data, and an evaluation unit configured to process at least the operator property data.to evaluate the object property data and the intervention-related operation data and, depending on the evaluation, to determine and issue a recommendation for action.

[0011] With regard to a method for issuing a recommendation for action to an operator performing a surgical procedure on an object, the invention particularly proposes that the object is identified by means of an object identification unit, the operator is identified by means of an operator identification unit, object property data describing the identified object is stored in a first storage unit, operator property data describing the identified operator is stored in a second storage unit, the procedure to be performed on the object is specified by means of a specification unit, procedure-related operational data is obtained and stored by means of a preparation unit communicating with the specification unit, and at least the operator property data is retrieved by means of an evaluation unit.The object property data and the intervention-related operation data are evaluated, and depending on the evaluation, a recommendation for action is determined and issued.

[0012] With regard to a computer program product, the invention particularly proposes that the computer program product comprises a program for a computing unit of a support device, wherein the program includes program code sections for executing at least one of the following steps of a method for issuing a recommendation for action to an operator performing an operational intervention on an object, in which an object identification unit of the support device identifies the object, an operator identification unit of the support device identifies the operator, object property data describing the identified object are stored in a first storage unit of the support device, and operator property data describing the identified operator are stored in a second storage unit of the support device.A control unit of the support device specifies the intervention to be performed on the object, a preparation unit of the support device, which communicates with the control unit, procures and stores intervention-related operation data, and an evaluation unit of the support device evaluates at least the operator property data, the object property data, and the intervention-related operation data, and determines and issues the action recommendation based on the evaluation.

[0013] The invention is based, among other things, on the idea that—in contrast to the prior art—the support device considers not only object property data for determining the recommended course of action, but also, and in particular, operator property data. This makes it possible to determine the recommended course of action for the operator during the procedure in a significantly more specific and accurate manner, thereby improving the overall reliability of the procedure and / or reducing its duration. By additionally considering the operator's property data, the recommended course of action can be determined and issued in a way that is tailored to the individual operator. This prevents the operator from receiving a recommendation that is difficult or even impossible to implement, or that is, for example, ergonomically unfavorable for the individual operator.

[0014] In this context, operator attribute data can include, for example, the operator's physical data, such as data relating to height or limb dimensions, the operator's level of training, procedure-related quality data relating to the operator, and / or the like. Object attribute data can include, for example, object dimensions, dimensions of the object's limbs, dimensions of the object's organs, and other physical object data, such as those relating to living beings, their overall physical condition, blood pressure, oxygen saturation, body temperature, and / or the like.

[0015] The object identification unit (OID) enables the identification of an object based on its properties, a unique identifier, and / or similar characteristics. For this purpose, imaging can be used as a supplementary method, where the object is first captured by a camera and a corresponding digitized image is stored. The OID can then use its own camera to create a digital image of the object and compare it to the stored image to identify the object. Alternatively, an identification element can be attached to the object, such as an RFID tag, which provides a unique identifier that can be read wirelessly.The object identification unit may include a corresponding reading unit capable of reading the stored identification or similar. Depending on the identified object, the support device may access further databases to obtain object property data, particularly for intervention-related purposes. This object property data may include, for example, geometry, dimensions, and / or other attribute data of the object.

[0016] Accordingly, the operator identification unit is designed to identify the operator performing the intervention on the object. The operator identification unit is therefore positioned at least partially near the object during the intervention and can detect the operator for identification purposes, preferably even during the intervention itself. For this purpose, the operator can be identified, for example, by means of a camera, or by entering an identification code into an input unit of the operator identification unit, and / or similar means. Depending on the identified operator, the support device can retrieve and provide relevant operator attribute data from other storage units or databases.A particularly advantageous feature is that the operator identification unit allows for continuous monitoring of the operator's identification during the procedure. This ensures that the identified operator is indeed performing the intervention. For example, if the operator changes during the procedure, this can be recorded by the operator identification unit and taken into account when considering further action recommendations.

[0017] The predefined unit allows the user to specify the intervention to be performed on the object. This intervention might involve, for example, using an instrument to manipulate the object, specifically inserting or moving the instrument within a specific area of ​​the object. The instrument can be used to manipulate the object, at least partially, either within or on its surface, for instance, removing or inserting material, applying heat to the material, and / or performing similar actions. The predefined unit can include an input unit for specifying the intervention. For example, during surgery on a patient, the unit might specify the organ to be operated on. Alternatively, it might specify that a particular chamber of a multi-chambered hollow body should be filled with a substance. Numerous other interventions are conceivable.

[0018] The preparation unit, which communicates with the input unit, can acquire and store procedure-related operational data. The preparation unit can access additional storage units and / or databases to obtain procedure-specific data. For example, the preparation unit can be configured to acquire specific data for the procedure specified by the input unit. This data might already be collected from other objects where similar procedures have been performed. Databases relating to the object's anatomy can also be used. The support device can communicate with the internet or a data cloud for this purpose.

[0019] The evaluation unit may include a computing unit and / or at least partially utilize a computing unit of the support device to provide the desired functionality. The evaluation unit may also, at least partially, consist of an electronic hardware circuit or the like. The evaluation unit is designed to evaluate at least the operator property data, the object property data, and the intervention-related operation data, and, depending on the evaluation, to determine and issue a recommendation for action. For this purpose, the evaluation unit may process the data accordingly so that the desired recommendation for action can be determined. The output of the recommendation for action can, for example, be in the form of a signal that can be audibly transmitted to the operator via an output unit.The output unit can be, for example, acoustic, visual, and / or haptic. However, the output unit is not limited to these configurations. Furthermore, a combination of the aforementioned options is also possible. The delivery of the recommended action can at least also include saving the information.

[0020] The evaluation unit communicates with the other units of the support device to perform the corresponding evaluation. The recommended action could, for example, be an instruction for the surgeon to perform a specific task at a particular time. It could also be an instruction for the surgeon not to perform a specific action or activity at that time. For example, the recommended action could be an instruction for the surgeon to operate or use one or more surgical instruments in a prescribed manner.

[0021] The invention thus makes it possible, particularly when considering the operator's characteristics, to optimize the recommended course of action not only with regard to the object and the procedure to be performed, but also additionally by taking into account the operator's characteristics. This demonstrates that the reliability of performing the procedure with a specific, identified operator and / or the time required to perform the procedure can be optimized or improved.

[0022] According to a training program, it is proposed that the support device include an initial data acquisition unit for recording the current situation during the surgical procedure and providing situational data. The evaluation unit is designed to further analyze this situational data and determine a recommended course of action based on this analysis. This training program enables the device to update the recommended course of action or issue a new or updated one if deemed necessary based on the current situational data. This allows the operator to receive appropriate recommendations for subsequent steps in the surgical procedure, enabling the operator to optimize the procedure by considering these recommendations.This advanced training proves particularly advantageous when one or more unexpected events occur during surgery that could influence or even hinder the procedure. Specifically, it can be arranged that the current situation is recorded repeatedly at predetermined times and / or upon the occurrence of specific events, and that a supplementary or new course of action is determined and provided. This allows for monitoring the surgical procedure and, for example, providing the surgeon with ongoing, time-sensitive, or continuous updates with specific course of action. The data acquisition unit could be, for example, an MRI scanner, a CT scanner, an ultrasound device, a combination thereof, or similar equipment. The current situation can reflect the ongoing progress of the surgical procedure on the patient, such as the positioning and / or use of an instrument in connection with the procedure.

[0023] The treatment instrument or instrument can be, for example, a scalpel, a cannula, a stent, a thread, a needle, combinations thereof and / or the like.

[0024] The treatment instrument serves to support the surgical procedure on the object.

[0025] The support device, in particular the evaluation unit, may comprise an electronic circuit and / or a computing unit. A computing unit may be understood to be, in particular, a data processing device containing a processing circuit. The computing unit may therefore, in particular, process data to perform arithmetic operations. This may also include operations to perform indexed accesses to a data structure, for example, a lookup table (LUT). The computing unit may, in particular, contain one or more computers, one or more microcontrollers, and / or one or more integrated circuits, for example, one or more application-specific integrated circuits (ASICs), one or more field-programmable gate arrays (FPGAs), and / or one or more systems on a chip (SoCs).The computing unit can also include one or more processors, for example, one or more microprocessors, one or more central processing units (CPUs), one or more graphics processing units (GPUs), and / or one or more signal processors, in particular one or more digital signal processors (DSPs). The computing unit can also include a physical or virtual array of computers or other units of the aforementioned type. In various embodiments, the computing unit has one or more hardware and / or software interfaces and / or one or more memory units.

[0026] A storage device within the meaning of this disclosure may be a volatile data storage device, for example, a dynamic random access memory (DRAM) or a static random access memory (SRAM), or a non-volatile data storage device, for example, a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), a flash memory or flash EEPROM, a ferroelectric random access memory (FRAM), or a magnetoresistive random access memory.It can be designed as MRAM (magnetoresistive random access memory) or as phase-change random access memory, PCRAM (phase-change random access memory).

[0027] According to a further training proposal, the evaluation unit comprises a processing unit that utilizes a machine learning method. This processing unit is trained using at least a two-stage learning process, with the first stage involving training the processing unit using at least the procedure-related operational data. This first-stage training identifies the processing unit for the intended surgical procedure. For this purpose, the procedure-related operational data acquired and stored by the preparation unit can be used.For example, if a surgical procedure on an organ, such as the patient's liver, is planned, available data regarding this procedure from other surgeries and their outcomes can be provided to train the processing unit accordingly. Based on this training, the processing unit is then prepared or configured for use in this specific surgical procedure. Additionally, operator and object property data can be used for training. This makes it possible to train the processing unit specifically for the object and the operator.

[0028] For this purpose, the evaluation unit, in particular the processing unit, can incorporate a neural network. Here and in the following, an artificial neural network can be understood as program code stored on a computer-readable storage medium that represents one or more interconnected artificial neurons or can replicate their function. The program code can also contain multiple program code components, which, for example, may have different functions. In particular, an artificial neural network can implement a nonlinear model or a nonlinear algorithm that maps an input to an output, where the input is given by an input feature vector or an input sequence, and the output can, for example, include a category output for a classification task, one or more predicated values, or a predicated sequence.

[0029] Machine learning algorithms can be considered computer algorithms for the automated execution of a processing task. A processing task can be understood, for example, as the extraction of information from available data. In some cases, the processing task could, in principle, be performed by a human capable of perceiving information corresponding to the data. However, in this context, processing tasks are also performed automatically without human intervention.

[0030] A computer algorithm may, for example, include a processing algorithm or a data analysis algorithm that is or has been trained through machine learning and may be based on an artificial neural network, particularly a convolutional neural network. The computer algorithm may, for example, include an object detection algorithm, an obstacle detection algorithm, an object tracking algorithm, a classification algorithm, a semantic segmentation algorithm, and / or a depth estimation algorithm.

[0031] Corresponding algorithms can also be implemented based on data other than information perceptible to humans. For example, point clouds or images from infrared cameras, lidar systems, etc., can also be evaluated using appropriately adapted computer algorithms. Strictly speaking, these algorithms are not processing algorithms because the corresponding sensors can operate in ranges that are imperceptible to humans, such as the infrared range. Therefore, within the scope of the present invention, such algorithms are referred to as automatic perception algorithms. Automatic perception algorithms thus include, among other things, automatic visual perception algorithms, but are not limited to human perception.Consequently, according to this understanding, an automatic perception algorithm can include a computational algorithm for the automatic execution of a perception task, which is or has been trained, for example, through machine learning and may, in particular, be based on an artificial neural network. Such generalized automatic perception algorithms can also include object detection algorithms, object tracking algorithms, classification algorithms, and / or segmentation algorithms, such as semantic segmentation algorithms.

[0032] If an artificial neural network is used to implement an automatic processing algorithm, one architecture employed can be that of a convolutional neural network (CNN). Specifically, a 2D CNN can be applied to corresponding 2D camera images. CNNs can also be used for other automatic processing algorithms. For example, 3D CNNs, 2D CNNs, or 1D CNNs can be applied to point clouds, depending on the spatial dimensions of the point cloud and the processing requirements.

[0033] The result or output of an automatic processing algorithm can depend on the specific underlying processing task. For example, the output of an object recognition algorithm might include one or more bounding boxes that define a spatial position and, optionally, an orientation of one or more corresponding objects in the environment, and / or corresponding object classes for the one or more objects. The output of a semantic segmentation algorithm applied to a camera image might include a pixel-level class for each pixel in the image. Similarly, the output of a semantic segmentation algorithm applied to a point cloud might include a corresponding point-level class for each point. The pixel-level and point-level classes, respectively, might define an object type to which the respective pixel or point belongs.Of course, it is also possible to consider additional data relevant to the surgical procedure for training purposes. For example, supplementary data available regarding the object can be taken into account, such as data relating to one or more previous surgical procedures on the object, even if these procedures were not performed in the same area. This allows for the consideration of object-specific characteristics during the determination of the recommended course of action, which may become relevant depending on the course of the surgical procedure. This, in turn, allows for further improvement of the support device, the procedure, and the computer program.

[0034] Furthermore, it is proposed that the processing unit be trained to perform a second stage of adaptive learning, at least using situational data. For this purpose, the situational data collected during the surgical procedure can be used to supplement the training of the processing unit during the procedure itself, or to conduct adaptive learning based on this data. This allows the evaluation unit, particularly the processing unit, to be further trained and its adaptation improved during its use, preferably during the surgical procedure.

[0035] It is further proposed that the support device include an object detection unit for capturing object properties. The object detection unit can provide corresponding object data that can be used supplementarily by the evaluation unit, in particular the processing unit. The object properties can include, for example, temperature, humidity, pressure, as well as mechanical dimensions and / or the like. In the case of a dynamic object, especially a patient, vital parameters can also be captured using suitable sensors. Vital parameters in a patient can include, for example, blood pressure, pulse rate, respiration, brain waves, and / or the like. Among other possibilities, the object's body temperature can be measured using a suitable thermometer, in particular an infrared thermometer.

[0036] Furthermore, it is proposed that the support device include an operator detection unit for recording operator characteristics. The operator detection unit can have one or more sensors by which the appropriate or desired parameter can be recorded, preferably in real time. For example, these physical parameters can include height, arm length, leg length, weight, but also vital parameters such as heart rate, blood pressure, pulse rate, skin conductance, and / or the like. This allows for the recording of corresponding operator characteristics and the provision of appropriately assigned data within the support device for its function. This data can be additionally considered by the evaluation unit, particularly the processing unit, with particular advantage.The object's properties can be recorded, for example, using a biomechanical jacket worn by the surgeon, or similar equipment. Of course, other or supplementary sensors or sensor units can also be used to record the desired parameters of the surgeon. For example, the surgeon's body temperature can be measured using an infrared thermometer.

[0037] The operator detection unit is designed to record at least one of the operator's positions or orientations relative to the surgical site during the procedure. This allows the recommended course of action to be supplemented by the operator's relative position to the object. The recommended course of action includes a change in the operator's position relative to the object, or similar adjustments. This makes it possible to guide the operator into the most advantageous position for performing the surgical procedure.

[0038] It is particularly advantageous if the support device is designed to record operator-related data during the procedure. This can include, for example, monitoring the operator's physical exertion to reduce their workload. Furthermore, it can be designed to assess the operator's overall capacity and, especially during lengthy or complex procedures, suggest measures to maintain or restore their performance. These measures might include taking breaks, consuming food and / or drink, and / or similar actions.Furthermore, the instruction may also include a change of operator if the support device has determined that the operator is unable to provide the necessary knowledge and / or physical performance to continue the surgical procedure.

[0039] Preferably, the evaluation unit is designed to perform the second stage of training additionally, depending on the collected operator-related data. This allows operator-related data acquired during the progress of the surgical procedure to be used to further train the evaluation unit, particularly the processing unit. For example, this allows the determination and output of the action recommendation to be adapted so that the surgeon receives instructions tailored to their individual ergonomic capabilities.

[0040] Furthermore, it is proposed that the support device be designed to provide at least one recommendation for action prior to the procedure, depending on the specific intervention to be performed on the object. This allows the surgeon to be provided with relevant information to prepare for the surgical procedure, enabling better planning. Essentially, one or more recommendations for action can be used to at least partially, but preferably completely, prepare for the procedure, particularly by simulating it. For example, the recommendation could also include a predicted course of the surgical procedure, allowing the surgeon to prepare accordingly.

[0041] Furthermore, it is proposed that the support device include a device detection unit for detecting devices used to support and / or perform the procedure within a surgical environment. This allows the devices to be considered when determining the recommended course of action. For example, the recommended course of action might include the use of a specific device during the surgical procedure at a particular time or in a specific situation. The surgical environment refers to the environment in which the surgical procedure is performed on the object. The surgical environment could, for example, be a spatial area encompassing the object and the surgeon, and preferably also the support device. The surgical environment could, for example, be a room in which the procedure is performed. The device could, for example, be a treatment instrument or a sensor unit for detecting a specific parameter.The device detection unit can, for example, include a camera that can detect and identify devices in the intervention area. The camera can provide relevant camera data for further use by the support device. Image analysis can then be used to identify the devices present. Furthermore, it is also possible to mark one or more devices with a wirelessly identifiable tag, such as an RFID tag, and to equip the device detection unit with a corresponding sensor for reading such identification tags. This allows for a simple determination of which devices are available in the intervention area and / or how many devices are present.

[0042] According to a training course, it is proposed that the evaluation unit be trained to determine the recommended course of action based on the evaluation of situational data and the recorded devices. This makes it possible to adapt the recommended course of action even more specifically to the available constraints, thereby further improving the execution of the surgical procedure. This data can be used particularly advantageously by the processing unit. Furthermore, this data can also be used to provide supplementary adaptive training for the processing unit.

[0043] The advantages and effects specified for the support device according to the invention naturally apply equally to the method according to the invention and to the computer program product according to the invention, and vice versa. Accordingly, device features can also be formulated as method features and vice versa.

[0044] The features and combinations of features mentioned above in the description, as well as the features and combinations of features mentioned below in the figure description and / or shown in the figures alone, can be used not only in the combination specified in each case, but also in other combinations without leaving the scope of the invention.

[0045] For use cases or application situations that may arise during the procedure and are not explicitly described here, it may be provided that, according to the procedure, an error message and / or a request for user feedback is issued and / or a default setting and / or a predetermined initial state is set.

[0046] The embodiments described below are preferred embodiments of the invention. The features and combinations of features specified above in the description, as well as those mentioned in the following description of embodiments and / or shown in the figures, are not only usable in the combinations specified, but also in other combinations. Thus, embodiments not explicitly shown and explained in the figures are also included in the invention and are considered disclosed, but can be derived and generated from the described embodiments by separate combinations of features.The features, functions, and / or effects illustrated by the exemplary embodiments can each, considered independently, represent individual features, functions, and / or effects of the invention, each of which further develops the invention independently. Therefore, the exemplary embodiments are intended to include combinations other than those described in the embodiments. Furthermore, the described embodiments can also be supplemented by additional features, functions, and / or effects of the invention already described.

[0047] In the figures, the same reference symbols denote the same features or functions.

[0048] Regardless of the grammatical gender of a particular term, persons with male, female or other gender identities are included.

[0049] They show: Fig. 1 a magnetic resonance imaging (MRI) apparatus comprising a magnetic resonance imaging (MRI) machine and an MR local coil connected to the MRI machine, positioned on a patient for performing an examination on the patient; Fig. 2 a perspective view of the magnetic resonance device according to Fig. 1 with an object stage arranged at the edge of an opening of a through-hole of the magnetic resonance device; and Fig. 3. A schematic block diagram illustrates the functional sequence of the support device.

[0050] Fig. Figure 1 shows a partially cutaway view of a magnetic resonance imaging (MRI) setup 10, which includes a magnetic resonance imaging (MRI) scanner 14 as the imaging device and a local coil 16. In this case, the MRI setup 10 is used to perform an imaging procedure for the medical examination of a patient 12 as the subject of the examination. The examination is performed by a surgeon 64 during a surgical procedure on the patient 12 to monitor the progress of the operation. However, the surgical procedure and the MRI setup 10 are not limited to this application and can be used in a variety of other surgical procedures on virtually any subject instead of the patient 12.

[0051] The magnetic resonance imaging (MRI) device 14 has a main magnet 18, which serves to generate a strong, and in particular time-constant, main magnetic field in one direction 20. The MRI device 14 also has a through-hole 22 as a detection area, which is provided here as a tunnel and serves to receive the patient 12. The through-hole 22 is essentially cylindrical and is surrounded in one circumferential direction by the main magnet 18. In principle, however, the through-hole 22 can also be configured differently.

[0052] The magnetic resonance imaging (MRI) device 14 further comprises a patient positioning device 24 or an object positioning device, which makes it possible to move the patient 12 or the object into the through-opening 22. For this purpose, the patient positioning device 24 has a patient table 26, arranged to be movable at least partially within the through-opening 22, which serves as the object table. However, the invention is not limited to this design of the detection area of ​​the MRI device 14. It is provided that an examination area 30, in which the surgical procedure on the patient 12 is to be performed, is arranged within the detection area of ​​the MRI device 14 in order to monitor the course or progress of the surgical procedure.

[0053] The magnetic resonance imaging (MRI) device 14 further comprises a gradient coil unit 28, by means of which magnetic field gradients can be generated that can be used for spatial coding during an imaging examination. The gradient coil unit 28 is controlled here by a gradient control unit 32 of the MRI device 14. The main magnet 18 and the gradient coil unit 28 form a magnetic field source.

[0054] The magnetic resonance imaging (MRI) device 14 further comprises a high-frequency antenna unit 34, which in the present embodiment is permanently integrated into the MRI device 14 as a body coil. The high-frequency antenna unit 34 is controlled by a high-frequency antenna control unit 36 ​​of the MRI device 14. This makes it possible to generate high-frequency magnetic resonance sequences in the examination area 30. During normal operation of the MRI device 14, excitation of atomic nuclei in the examination area 30 of the patient 12 can be achieved. Relaxation of the excited atomic nuclei generates MR signals. The MR signals can be received by means of the high-frequency antenna unit 34.

[0055] To control the components of the magnetic resonance imaging (MRI) device 14, the MRI device 14 has a device control unit 38. The device control unit 38 centrally controls the MRI examination setup 10, for example, with regard to performing a predetermined examination on the patient 12, in particular performing a predetermined imaging gradient echo sequence. Furthermore, the device control unit 38 has an MR evaluation unit (not shown) by means of which received MR signals acquired during the examination can be evaluated.

[0056] The magnetic resonance imaging (MRI) device 14 also has a user interface 40, which is communicatively connected to the control unit 38. The user interface 40 has a display unit 42 and an input unit 44, each of which is communicatively connected to the device control unit 38. Data and information, evaluation results, such as imaging results, parameters relating to the MRI evaluation, and / or the like, can be visually displayed as needed via the display unit 42. The display unit 42 can, for example, include a monitor or the like. Using the user interface 40, a user of the MRI examination setup 10, or the operator 64, can supply data to the device control unit 38 as needed, such as parameters for conducting the examination, control data, information, and / or the like.

[0057] Out of Fig. Figure 1 shows that a local coil 16 is arranged on the patient 12 in the examination area 30. The local coil 16 is located on a surface of the patient 12 in the examination area 30. The local coil 16 has a connecting cable 46, which is connected in a connection area 52 to a connecting cable 54 of the magnetic resonance imaging (MRI) device 14 for communication and / or signal purposes. A coil connection element 48 is provided in the connection area 52 for this purpose and is permanently connected to the connecting cable 46. The coil connection element 48 is designed as a plug.

[0058] In connection area 52, a device connection element 50 is also provided, which is permanently connected to the connecting cable 54. The coil connection element 48 and the device connection element 50 allow for a detachable connection between the local coil 16 and the magnetic resonance device 14 in connection area 52. This makes it possible to couple each local coil 16 to the respective magnetic resonance device 14. The device connection element 50 can be configured as a socket. Connection area 52 is a standardized connection area, enabling the connection of any local coil 16 to any magnetic resonance device 14 based on this standard. Both the coil connection element 48 and the device connection element 50 are shielded.

[0059] In Fig. In this configuration, only a single connection unit 52 is provided for connecting a single local coil 16. Alternative configurations may, of course, provide for multiple connection units 52, allowing the connection of more than one local coil. Furthermore, the device connection element 50 may be located at any suitable position on the patient table 26. It is also possible to provide for more than one device connection element 50, preferably with the multiple device connection elements 50 being located at different positions, particularly on the patient table 26. The high-frequency antenna unit 34 can be adapted as required. This makes it possible to adapt the magnetic resonance imaging setup 10 to the specific surgical procedure being performed.

[0060] The magnetic resonance imaging apparatus 10 also includes a power supply unit 56. The power supply unit 56 is connected to a three-phase power supply network 58.

[0061] Fig. Figure 2 shows a perspective view of the magnetic resonance imaging (MRI) device 14 with a specimen table or patient table 26 arranged at one edge of an opening 60 of the through-hole 22 of the MRI device 14. The specimen table or patient table 26 is designed to be movable and has, for this purpose, unspecified casters. The specimen table 26 can be positioned on the MRI device 14 as needed.

[0062] The object table or patient table 26 has an object support or patient support 62 on which the patient 12 can be positioned. The patient support 62 is designed to be movable in the longitudinal direction of the patient table 26, so that a patient 12 positioned on the patient support 62 can be moved as needed into the passage opening 22 by means of the patient support 62.

[0063] As from Fig. As can be seen in Figure 1, a surgical procedure is being performed on patient 12, who is the subject of this study. For this purpose, the surgeon 64 uses a surgical instrument 98, which in this case is a scalpel. To perform the surgical procedure, the surgeon 64 manually moves the scalpel to the area under investigation where the procedure is to take place. The surgical procedure is performed in the area of ​​the local coil 16, so that the magnetic resonance imaging (MRI) device 14 is able to record the progress of the surgical procedure on patient 12.

[0064] Due to the operating situation and the dimensions of the magnetic resonance imaging (MRI) machine 14, the area where the surgical procedure is performed on patient 12 is poorly visible to the surgeon 64 and is also only accessible in an inconvenient manner. Therefore, a support device 70 is provided to assist the surgeon 64 during the surgical procedure on patient 12.

[0065] The support device 70 serves to provide a recommendation for action to the operator 64, for which purpose the support device 70 has an acoustic output unit in the form of a loudspeaker 82, which makes it possible to transmit the recommendations for action 110 ( Fig. 3) to be output acoustically to the surgeon 64. The surgeon 64 can therefore, using the action recommendation 110, optimize his work during the surgical procedure on patient 12 and thus improve the treatment outcome.

[0066] In order to provide the action recommendations 110, the support device 70 includes an object identification unit 66 for identifying the object, in this case, patient 12. The object identification unit 66 can identify patient 12, for example, using camera images and capturing identifying features of patient 12's body and / or the like. Furthermore, identification is also possible using an identification element, such as a chip card belonging to patient 12 that has an individual patient identification number, an RFID tag attached to a part of patient 12's body, and / or the like. The object identification unit 66 is designed to detect the chip card or RFID tag and read and process the individual identification stored on it.

[0067] Furthermore, the support device 70 includes an operator identification unit 68 for identifying the operator 64. Similar options for identifying the operator 64 as those used for identifying the patient 12 can be provided here. In addition, it can also be provided that the operator 64 enters an individual personal identification number into an input unit of the operator identification unit 68 in order to identify themselves to the support device 70.

[0068] The support device 70 further comprises a first storage unit 72, which serves to store descriptive patient property data 12 for the identified patient as object property data. This data can be obtained, for example, by the object identification unit 66 or by another unit of the support device 70. For this purpose, the support device 70 can communicate with other databases that can provide corresponding patient property data, such as body dimensions, the patient's weight, and other medical-physiological data such as blood pressure, body temperature, disabilities, and / or the like.Furthermore, the patient attribute data may also include data collected in relation to previous surgical procedures performed on patient 12, such as previous operations, data relating to anesthesia during a previous operation and / or the like.

[0069] The support device 70 also includes a second storage unit 74, which serves to store operator characteristic data describing the identified operator 64. This operator characteristic data can also include, for example, body dimensions, level of training, physical constitution, and / or the like relating to the operator 64. Thus, the support device 70 contains a detailed database relating to the operator 64, enabling the recommendation for action 110 to be determined and issued in a way that is specifically tailored to the characteristics of the operator 64. The operator identification unit 68, or another unit of the support device 70, can also access further databases to obtain specific operator characteristic data. This data can, for example, also contain information from previously performed surgical procedures.

[0070] The support device 70 is therefore preferably connected to a data cloud 100 via communication technology, which can, for example, include the Internet and through which it is possible to access a large number of different databases in order to obtain the desired data.

[0071] The support device 70 further comprises a control unit 76, which serves to specify the procedure to be performed on the patient 12. For example, the procedure may be to be performed on an organ of the patient 12, such as a liver, a kidney, an intestine, or the like. The control unit 76 can implement the specification of the surgical procedure based on or taking into account patient characteristic data. For this purpose, the control unit 76 may be provided with an input unit that enables the surgeon 64 or other personnel to specify the surgical procedure accordingly. The control unit 76 may also be communicatively connected to a remotely located input unit, so that the specification of the surgical procedure can be made from the remote location.For example, it could be stipulated that the surgical procedure for patient 12 is prescribed by a specialist who, based on an examination of patient 12, deems the surgical intervention necessary. The surgical procedure is then performed by surgeon 64, for example, in a hospital. This makes it possible to network different institutions, such as a hospital and a specialist's practice, and thus improve overall efficiency.

[0072] Furthermore, the support device 70 includes a preparation unit 78, which communicates with the pre-setup unit 76. The preparation unit 78 serves to prepare the surgical procedure on patient 12 and, for this purpose, to obtain and store procedure-related surgical data. The procedure-related surgical data may also include patient attribute data. In particular, the procedure-related surgical data may relate to data pertaining to patient 12, specifically to an area of ​​their body on which the surgical procedure is to be performed. For example, this data may include detailed information relating to an organ of patient 12, such as vascular anatomy and / or the like.

[0073] The preparation unit 78 can use data from the input unit 76 to determine an area on the patient's body 12 where the surgical procedure is to be performed. This can also be used, among other things, by the magnetic resonance imaging setup 10 to position the patient 12 appropriately in the opening 22 so that the area where the surgical procedure is to be performed on the patient 12 can be clearly visualized by the magnetic resonance device 14.

[0074] The support device 70 further comprises an evaluation unit 80, which is designed to evaluate at least the operator's attribute data, the patient's attribute data, and the procedure-related operational data, and, depending on the evaluation, to determine the action recommendation 110. The action recommendation 110 determined in this way can then be output acoustically via the acoustic output unit 82, so that the operator 64 can perceive it acoustically.

[0075] The evaluation unit 80 has an electronic hardware circuit and a computer unit for this purpose, which realizes the desired functionality using one or more suitable computer program products.

[0076] Fig. Figure 3 shows a schematic flowchart illustrating the functional sequence of the support device 70. It can be seen that the evaluation unit 80 issues a determined action recommendation 110 to the surgeon 64 – as previously explained. The evaluation unit 80 has a processing unit 86 that uses a machine learning method. The processing unit 86 is trained using a two-stage learning process. In the first stage, the processing unit 86 is trained 102, at least using the procedure-related surgical data 104. In this case, the training 102 includes at least instructions based on the present case, an optimized surgical technique, a reduced risk of infection, a shorter hospital stay for the patient 12, and / or the like. The training also includes improving the resection during the surgical procedure and improving the patient's quality of life 12.The training also includes procedure-specific recommendations and summaries, such as disease-specific recommendations, the best surgical approach during the procedure with the fewest possible complications, and suggestions for execution and prognosis. For this purpose, the processing unit 86 can include a corresponding computing unit that, among other things, provides a neural network which can be trained using the given data, in particular using the data presented or provided in the following steps 104 to 108.

[0077] In the first step, training 102 is conducted using the data presented or provided in step 104. This data consists of intervention-related operational data. This data can include registration data, such as relevant data from a collaborative quality initiative, KOL (Key Opinion Leader) data, board data, and / or the like. Furthermore, stored data can be used for managing, storing, reporting, supporting registrations, providing recommendations for workflows based on registration data, and the like. Additionally, carbon intelligence data can be considered, which may relate to training, education, and / or peer data. Finally, databases relating to medical publications, Google Scholar, conference data, and others can be considered.Furthermore, it is possible to include data on support staff such as caregivers, nurses, anesthesiologists, radiologists, cardiologists, and / or the like. Finally, alternatively or additionally, in step 104, medical device data or workflow data, such as the angle to be applied to a treatment needle, anatomical data (e.g., based on MRI data), protocols, screen data, and / or the like, can also be included. This data can be obtained, at least in part, from other databases via the data cloud 100 or the internet and stored in the support device 70.

[0078] Taking into account the aforementioned data, appropriate training procedures can be carried out in step 102 that can achieve the aforementioned goals for step 102.

[0079] Furthermore, it is possible to implement basic training of processing unit 86, taking into account step 106, in which patient-related data and / or therapy data for patient 12 are considered. In this context, for example, an EMR, a diagnosis, a patient history, and / or the like can be taken into account.

[0080] Out of Fig. As can be seen in Figure 1, the surgeon 64 wears a biometric jacket 88, which can record the surgeon's position, orientation, and / or posture. The biometric jacket 88 communicates with a surgeon acquisition unit 90 of the support device 70. This allows a body model of the surgeon 64 to be created using the evaluation unit 80, in particular the processing unit 86, on the basis of which the action recommendations 110 can be determined. The body model is preferably dynamic, meaning that it can be continuously adapted like a digital twin. This also allows, among other things, certain surgical steps to be simulated during the procedure before the surgeon 64 actually performs them. This can preferably be implemented during the surgical procedure on the patient 12.

[0081] The support device 70 further comprises a first acquisition unit 84 for recording the current situation during the surgical procedure on patient 12. The acquisition unit 84 can provide corresponding situational data. The situational data can be made available to the evaluation unit 80, in particular the processing unit 86, in order to determine the recommended course of action by evaluating and additionally considering the situational data. The situational data can be, among other things, in Fig. 3 in a step 108 for training the processing unit 86 in a second stage according to the nature of adaptive learning. The situational data can, for example, show a relative position of the operator 64 in relation to the patient 12 and / or the like.

[0082] Using the evaluation unit 80, it is particularly possible to determine one or more corresponding recommendations for action 110 for carrying out the surgical procedure and to provide them to the surgeon 64 even before the surgical procedure is performed. This allows the surgeon 64 to better prepare for the surgical procedure.

[0083] The support device 70 further comprises a device detection unit 92, which is configured to detect devices for supporting and / or performing the procedure in a surgical environment. Such devices can be, for example, a surgical needle 94 and an aspiration line 96. However, the invention is not limited to this. The devices can also include screens, headphones, tables, the magnetic resonance imaging (MRI) scanner 14, an ultrasound device, a C-arm, a CT scanner, navigation systems and / or the like for registering a patient, robotic devices and communication devices, MR data, various protocols, surgical procedures, as well as hazards and / or the like.

[0084] In the present case, the evaluation unit 80 is further equipped to determine the action recommendation 110 additionally depending on the evaluation of the situation data and the recorded devices 94, 96.

[0085] The description of the figures serves solely to explain the invention and is not intended to limit it.

[0086] In particular, the invention is of course not limited to medical applications in the treatment of a patient. The invention can be used on virtually any object, as is also explained in the general description section.

Claims

[1] Support device (70) for issuing a recommendation for action (110) to an operator (64) who is performing a surgical procedure on an object (12), with - an object identification unit (66) for identifying the object, - an operator identification unit (68) for identifying the operator (64), - a first storage unit (72) for storing object property data describing the identified object (12), - a second storage unit (74) for storing operator property data describing the identified operator (64), - a specification unit (76) for specifying the intervention to be carried out on the object (12), - a preparation unit (78) in communication with the specification unit (76), which is trained to obtain and store intervention-related operational data (104), and - is trained in an evaluation unit (80) to evaluate at least the operator property data, the object property data and the intervention-related operation data (104) and, depending on the evaluation, to determine and issue the recommendation for action (110), characterized by , that the support device includes an operator detection unit (90) for detecting operator characteristics, wherein the operator detection unit (90) is designed to detect at least one position or orientation of the operator (64) in relation to an intervention site on the object (12) during the intervention. [2] Support device according to claim 1, characterized bya first recording unit (84) for recording a current situation during the operative intervention on the object (12) and for providing situation data, wherein the evaluation unit (80) is designed to additionally evaluate the situation data and to additionally determine the recommendation for action (110) depending on the evaluation of the situation data. [3] Support device according to any of the preceding claims, characterized by , that the evaluation unit (80) has a processing unit (86) which uses a machine learning method, wherein the processing unit (86) is trained by means of a learning method with at least two stages, wherein in a first stage the processing unit (86) is trained at least using the intervention-related operation data (104). [4] Support device according to claim 3, characterized by, that the processing unit (86) is trained, in a second stage to carry out training of the processing unit (86) in the manner of adaptive learning, at least using the situational data. [5] Support device according to any of the preceding claims, characterized by an object acquisition unit (88) for acquiring object properties. [6] Support device according to any of the preceding claims, characterized by , that the support device (70) is designed to record operator-related data during the performance of the procedure. [7] Support device according to claim 6, characterized by , that the evaluation unit (80) is trained to carry out the second stage of training additionally depending on the operator-related data recorded. [8] Support device according to claim 6 or 7, characterized by, that the evaluation unit (80) is trained to additionally take into account the recorded operator-related data during evaluation. [9] Support device according to any of the preceding claims, characterized by , that the support device (70) is designed to provide at least one recommendation for action (110) before the intervention is carried out, depending on the specified intervention to be carried out on the object (12). [10] Support device according to any of the preceding claims, characterized by a device acquisition unit (92) for acquiring devices (94, 96) to support and / or carry out the intervention in an intervention environment. [11] Support device according to claim 9, characterized by , that the evaluation unit (80) is trained to determine the recommendation for action (110) additionally depending on the evaluation of the situation data and the devices recorded (94, 96). [12] Method for issuing a recommendation for action (110) by means of a support device (70) according to one of claims 1 to 11, to an operator (64) who is performing a surgical procedure on an object (12), wherein - the object (12) is identified by means of an object identification unit (66), - the operator (64) is identified by means of an operator identification unit (68), - the identified object (12) describing object property data are stored in a first storage unit (72), - operator property data describing the identified operator (64) are stored in a second storage unit (74), - the intervention to be carried out on the object (12) is specified by means of a specification unit (76), - intervention-related operational data (104) are obtained and stored by means of a preparation unit (78) communicating with the specification unit (76), and - by means of an evaluation unit (80) at least the operator property data, the object property data and the intervention-related operation data (104) are evaluated and, depending on the evaluation, the recommendation for action (110) is determined and issued, characterized by , that the operator detection unit (90) detects at least one position or orientation of the operator (64) in relation to a site of intervention on the object (12) during the intervention. [13] Computer program product comprising a program for a computer unit of a support device (70) according to one of claims 1 to 11, wherein the program includes program code sections for executing the following steps of a method for issuing a recommendation for action (110) to an operator (64) who is performing an operative procedure on an object (12), in which - an object identification unit (66) of the support device identifies the object (12), - an operator identification unit (68) of the support device identifies the operator (64), - the identified object (12) describing object property data are stored in a first storage unit (72) of the support device (70), - operator property data describing the identified operator (64) are stored in a second storage unit (74) of the support device (70), - a predefined unit (76) of the support device (70) specifies the intervention to be carried out on the object (12), - a preparation unit (78) of the support device (70) which communicates with the specification unit (76) procures and stores intervention-related operational data (104), - an evaluation unit (80) of the support device (70) evaluates at least the operator property data, the object property data and the intervention-related operation data (104) and, depending on the evaluation, determines and issues the action recommendation (110), characterized by , that the operator detection unit (90) detects at least one position or orientation of the operator (64) in relation to a site of intervention on the object (12) during the intervention.

Images