Technology for path planning of a medical device

Abstract

A technique for planning a path of an automatically moving medical device comprises a computer-implemented method having a step of receiving sensor data regarding a possible range of motion of an automatically moving medical device (300). The presence of obstacles (308) in the possible range of motion is detected based on the received sensor data. Actions of the detected existing obstacles (308) are predicted based on the received sensor data. The prediction takes place taking into account medical processes in the range of motion of the automatically moving medical device (300). A path (302') of the automatically moving medical device (300) is planned based on the predicted actions of the obstacles (308).

Description

[0001] The invention relates to a method for planning a path of an automatically moving medical device, a computing device, an automatically moving medical device comprising the computing device, a system comprising the automatically moving medical device, a computer program (and / or computer program product) and a computer-readable storage medium.

[0002] In hospitals, there are typically unstable environmental conditions for autonomous robots or mobile devices, sometimes due to medical emergencies. For example, trash cans are moved, materials are delivered and stored, equipment is moved, or people walk through the corridors.

[0003] Autonomous mobile robots are robots that can move and act independently in their environment. Currently, there are various levels of autonomy, i.e., the robot's independence. For example, a mobile robot can explore its environment independently and complete assigned tasks, but must always return to a charging station. The degree of autonomy of a mobile robot can be defined in a specification and depends on the mobile robot's task.

[0004] For spatial navigation, simultaneous positioning and mapping (SLAM) is conventionally used. Mapping is typically performed using laser scans (e.g., LiDAR). For obstacle detection, RGB-D cameras and lasers are conventionally used, as described, for example, by Alessandro Di Fava in Wiki: Robots / TIAGo++ / Tutorials / Navigation / Localization (last edited 2023-03-30), the content of which is incorporated herein by reference.

[0005] A camera also enables vital parameter recording, as described by Tobias Haist et al. in "Camera-based Recording of Vital Parameters" tm - Technisches Messen, vol. 86, no. 7-8, 2019, pp. 354-361, the content of which is hereby incorporated by reference. The contactless recording of vital parameters with as little restriction as possible for the user and / or patient is of great practical interest for a wide range of potential applications. Pulse and respiration can be detected with high precision and robustness using a camera. The use of suitable wavelength channels is essential for pulse measurement, and the use of the holographic multipoint method for respiration detection is essential.

[0006] For path planning, public algorithms such as the Robot Operating System (ROS) or ROS2, available at http: / / wiki.ros.org / or http: / / docs.ros.org / , or proprietary algorithms such as SIMOVE ANS+ from Siemens AG can be used. For path planning, SIMOVE ANS+, for example, uses multiple paths, and if an obstacle is detected, the autonomous mobile robot uses a new path.

[0007] The disadvantage is that conventional path planning is not suitable for the complex and frequently changing ranges of motion of a medical device in a medical facility.

[0008] It is therefore an object of the present invention to provide a solution for optimized planning of a path of an automatically moving medical device. Alternatively or additionally, the object is to optimize a path of an automatically moving medical device with regard to freedom from collisions with obstacles, wear on the moving components, safety for persons (especially medical personnel), and a smooth (especially without time delays) treatment process for patients.

[0009] This object is achieved by a method for planning a path of an automatically moving medical device, by a computing device, by an automatically moving medical device comprising the computing device, by a system comprising the automatically moving medical device, by a computer program (and / or computer program product), and by a computer-readable storage medium according to the appended independent claims. Advantageous aspects, features, and embodiments are described in the dependent claims and in the following description, along with advantages.

[0010] In the following, the inventive solution is described with reference to the claimed method for planning a path of an automatically driving medical device and with reference to the claimed computing device. Features, advantages, or alternative embodiments herein can be assigned to the other claimed subject matter (e.g., the automatically driving medical device, the system, the computer program, or a computer program product), and vice versa. In other words, the claims for the devices and system can be improved by features described or claimed in connection with the method. In this case, the functional features of the method are embodied by structural units of the devices and / or the system, and vice versa.

[0011] According to one method aspect, a computer-implemented method for planning a path of an automatically moving medical device is provided. The method comprises a step of receiving sensor data regarding a possible range of motion of the automatically moving medical device. The method further comprises a step of detecting the presence of one or more obstacles in the possible range of motion based on the received sensor data. The method further comprises a step of predicting actions of the detected one or more obstacles based on the received sensor data. The prediction is performed taking into account medical processes in the range of motion of the automatically moving medical device.The method further comprises a step of planning a path of the automatically moving medical device based on the predicted actions of the one or more obstacles.

[0012] The technology according to the invention enables efficient, collision-free, and / or accident-free planning of a path (also known as path planning, path planning, and / or trajectory planning) of an automatically moving medical device (device for short). This can, for example, optimize the transport of medical products within a medical facility, particularly while simultaneously reducing the burden on the medical staff at the facility.

[0013] The path (also: trajectory) is (e.g., at least the spatial part) of a trajectory along which the medical device moves or is intended to move. The path can comprise a sequence of positions, orientations, velocities, and / or accelerations. An initial position of the path can be predetermined, in particular corresponding to the current position of the automatically moving medical device. Alternatively or additionally, an intermediate position and / or an end position of the automatically moving medical device can be predetermined, for example, due to a scheduled use of the medical device.

[0014] The automatically driving medical device can also be called an autonomous robot.

[0015] The automatically moving medical device can comprise a mobile device, in particular a transport robot (e.g. for medications, laboratory samples, laboratory equipment, treatment materials, instruments, medical handheld devices and / or surgical instruments), a laboratory device and / or a device in the medical environment (e.g. laptop). Alternatively or additionally, the automatically moving medical device can comprise a (particularly mobile) imaging device (also: medical scanner, in short: scanner). The imaging device (and / or the scanner) can comprise, for example, an ultrasound device and / or an X-ray device. Furthermore, alternatively or additionally, the automatically moving medical device can comprise a (particularly mobile) C-arm.

[0016] The automatic driving of the medical device can include at least partial automation.

[0017] Detecting the presence of obstacles may include object detection.

[0018] The obstacle(s) may be transportable, movable, and / or non-static. Alternatively or additionally, the obstacle(s) may include movable (medical and / or non-medical) equipment present in a medical facility.

[0019] An obstacle may, for example, include a medical imaging device, such as a C-arm device. Alternatively or additionally, the obstacle may include a patient bed parked (e.g., in a corridor), a delivery of goods, a cleaning cart, and / or a trash can.

[0020] An obstacle may also include a person, in particular medical personnel and / or a patient.

[0021] The actions (e.g., comprising at least one action) may imply a possible future path of the obstacle. Alternatively or additionally, the actions may include executing functions of the obstacle, such as taking an image from a CT, PET, MRI, US, and / or X-ray device. The actions may imply or cause a change in the position and / or orientation of the obstacle (e.g., pivoting a component of the device, adjusting its height, etc.).

[0022] The medical-technical processes can be stored in a database as a file (e.g., as an XML file). The medical-technical processes can be based on rules that transfer one state to a subsequent state. The medical-technical processes can refer to the medical device and / or to detected obstacles. The medical-technical processes can be specific to an area of ​​the medical facility in which the movement area of ​​the automatically moving medical device is located. For example, dental-technical processes can be known in a dentist's office or dental clinic, and / or operation-specific processes in an operating room.

[0023] The medical-technical processes can be known based on historically recorded data (historical data for short), for example, as statistical distributions. Alternatively or additionally, the medical-technical processes can be learned by artificial intelligence (AI), in particular based on the received sensor data and / or the historically recorded data. For example, the (particularly trained) AI can predict an individual medical-technical process based on sensor data, inputs, and / or planning by medical personnel (e.g., a surgeon or dentist).

[0024] The path planning can be done in the area of ​​the medical facility depending on the medical-technical processes.

[0025] The method may further comprise a step of receiving a map of the movement area.

[0026] Detecting obstacles, predicting obstacle actions, and / or planning the path can be done relative to a position and / or orientation of the map.

[0027] Detecting the obstacles and / or predicting the actions of the obstacles (especially relative to the map) can also be referred to as creating and / or evaluating a world model (technically: world model).

[0028] The map may include immobile objects, e.g., walls and fixed furnishings (e.g., a counter, table, and / or seating) and / or fixed medical devices. In one embodiment, the map may include height restrictions and / or clearance heights (e.g., a mobile robot may be able to move under a tabletop and / or barrier at a predetermined height).

[0029] The map must be generated on the basis of a (e.g. architectural) floor plan and / or on the basis of (e.g. LiDAR) sensor data (and / or laser scanner data).

[0030] The sensor data can be received by at least one (particularly external) sensor of a sensor system via a wireless data connection.

[0031] The sensor data can be collected by a sensor of a sensor system located inside (internal) and / or outside (external) the automated medical device. Reception can occur wirelessly, for example, via a local Wi-Fi, Bluetooth, ZigBee, and / or a cellular network.

[0032] In one embodiment, the sensor data can be preprocessed. For example, object detection may have already been performed on the automatically driving medical device before the sensor data was received. The sensor data can include the result of the object detection.

[0033] The at least one sensor can be included in a sensor system. The sensor data can thus be received by a sensor system. The sensor system can be optical, acoustic, resistive, mechanical, inductive, magnetic, thermal, piezoelectric, photoelectric, and / or thermoelectric. The optical sensor system can, in particular, comprise a camera and / or a LiDAR sensor. Optionally, at least one sensor of the sensor system, in particular a camera and / or a LiDAR sensor, can be arranged outside a currently optically visible area of ​​the automatically driving medical device.

[0034] The sensor system may comprise one or more sensors and / or combinations of different sensor types. For example, the sensor system may comprise one or more cameras and / or LiDAR sensors that are permanently arranged in the medical facility (e.g., in rooms and / or corridors). Alternatively or additionally, the sensor system may comprise sensors arranged on other mobile devices. The sensor data may comprise image data, in particular from a video camera, and / or LiDAR data.

[0035] The sensor data may include control instructions for and / or inputs relating to the medical device and / or another medical device, in particular a mobile C-arm.

[0036] The sensor data can relate to a planned technical process in the form of control instructions. The additional medical device can, for example, comprise a device (e.g., an imaging device) mounted on a C-arm (or other support structure). The device can, for example, comprise an X-ray device, an ultrasound device, a computed tomography (CT) and / or magnetic resonance imaging (MRI) device, a laboratory device, and / or a measuring device.

[0037] The control instructions for the (further) medical device can, for example, include data relating to a medical function. For example, the further medical device (e.g., an X-ray system) can be used to control an image acquisition of one or more body parts (e.g., the legs) of a patient.

[0038] When capturing control instructions indicating a planned image acquisition, for example, a path can be selected to minimize radiation exposure to the automatically moving medical device. Alternatively or additionally, it may be known from the medical process that doors are closed at the time of image acquisition and / or that medical personnel are entering or leaving a room immediately before and / or after the acquisition.

[0039] Alternatively or additionally, the control data can include a movement of the further medical device (e.g., moving into or out of a parking position). For example, the control data can encode a speed and / or a (e.g., piecewise) path of the movement. Alternatively or additionally, a path of the movement and / or another action (e.g., taking an image) can be extrapolated based on the received control data by taking the medical-technical processes into account.

[0040] The sensor data can include vital parameters of a patient in the range of motion.

[0041] Based on vital signs, and taking medical-technical processes into account, the status of a patient's treatment (e.g., a surgical procedure, also known as an operation, or OP for short) can be detected and future actions can be predicted accordingly. For example, at the end of treatment, medical personnel can move from a position assumed during treatment.

[0042] The vital parameters may include a pulse (e.g. determined based on wavelength channels) and / or respiration (e.g. determined based on a holographic multi-point methodology).

[0043] The patient's vital parameters can be obtained indirectly by means of a camera (particularly an external one) aimed at displaying vital parameters and / or directly by means of a patient monitoring device.

[0044] For example, a threshold value of a vital parameter may be associated with an unstable condition of the patient, which triggers (e.g., modified) path planning.

[0045] The method may include a step of aggregating the received sensor data.

[0046] The sensor data can be received from a plurality of sensors and / or a sensor system. Detecting the presence of obstacles and / or predicting the actions of the obstacles can be done based on the aggregated (also: combined) sensor data. For example, camera data can be combined with control data from another medical device to optimize path planning while taking medical processes into account.

[0047] The detection step may further comprise automatic, in particular algorithm-based, person recognition (in particular, the recognition that an unidentified person, for example, is located within the possible movement range of the automatically moving medical device). Optionally, the person recognition may comprise person identification. The person recognition may be configured to distinguish medical personnel from non-medical personnel, in particular a patient.

[0048] Automatic person recognition (and / or person identification) as medical personnel can include identification based on job-specific clothing (e.g., a surgical gown) of the medical personnel. Alternatively or additionally, person recognition (and / or person identification) as medical personnel can be based on sensor data (e.g., received and / or transmitted signals, in particular from a radio receiver, also known as a pager) that can be spatially assigned to a person. Alternatively or additionally, person recognition (and / or person identification) as medical personnel can be based on recognition of a transmitting device and / or a receiving device (e.g., based on object recognition in image data). Furthermore, alternatively or additionally, person recognition (and / or person identification) as medical personnel can include facial recognition.

[0049] Person recognition (and / or person identification) as a patient can comprise identification based on a position on a patient couch and / or based on treatment-specific clothing (e.g., a patient gown). Alternatively or additionally, person recognition (and / or person identification) as a patient can comprise recognition of one or more sensors arranged on the patient. For example, the sensor can comprise an SpO2 sensor that measures the oxygen saturation of the blood. The recognition of the sensor(s) can be based on object recognition in image data and / or on transmitted and / or received signals. Furthermore, alternatively or additionally, person recognition (and / or person identification) as a patient can comprise recognition of a medical device on the body, for example an access (e.g., for intravenous therapy) and / or one or more electrodes (e.g.,for an electrocardiogram, ECG, and / or an electroencephalogram, EEG).

[0050] If a person is identified as medical personnel, a first (e.g., shorter) predetermined distance can be used for path planning. If a person is identified as a patient, a second (e.g., longer) predetermined distance can be used for path planning.

[0051] Alternatively or additionally, if the person is identified as medical personnel, the automatically moving medical device can be enabled to receive and implement control commands from the medical personnel. If the person is identified as a patient, the receipt of control commands from the patient can be blocked.

[0052] The operational reliability of the automatically moving medical device can be increased through person-specific planning of distances and the possible reception of control commands.

[0053] The consideration of medical-technical processes can be based on historical data.

[0054] The historical data can include (e.g., aggregated) historical sensor data (and / or process data). For example, process data from a treatment (especially a surgical procedure) for the stages of preparation, execution, and follow-up, as well as optionally from image acquisition, can be aggregated and compiled into a medical-technical process. Before, during, and / or after image acquisition, it may be necessary, for example, to keep areas around doors clear and / or to consider the entry and / or exit of medical personnel into the operating room.

[0055] The historical data can be recorded statistically. Alternatively or additionally, the historical data can optionally be processed with the sensor data by artificial intelligence (AI).

[0056] The medical-technical processes can be recorded and / or stored as statistical probabilities of work steps, movements and / or behavior of the medical staff (e.g. in an operating room) and incorporated into the path planning.

[0057] Alternatively or additionally, work steps, movements, and / or behavior of medical personnel can be analyzed using AI and transformed into a model of a medical-technical process. The model can be (e.g., continuously) expandable.

[0058] Path planning can rely on defaults stored in memory, especially in the absence of detected obstacles.

[0059] A preferred default may include driving to the right of the space in the absence of detected obstacles.

[0060] Alternatively or additionally, the standard specifications may include driving near (particularly automatic) doors and / or elevator doors.

[0061] In one embodiment, the received sensor data may include indications of a planned door opening. When the doors are planned to open, the default setting may be a greater distance to the door than when driving past a closed door.

[0062] According to a first aspect of the device, a computing device for planning a path of an automatically moving medical device is provided. The computing device comprises a sensor data receiving interface configured to receive sensor data regarding a possible range of motion of the automatically moving medical device. The computing device further comprises a detection module configured to detect the presence of one or more obstacles in the possible range of motion based on the received sensor data. The computing device further comprises a prediction module configured to predict actions of the detected one or more obstacles based on the received sensor data. The prediction is performed taking into account medical processes in the range of motion of the automatically moving medical device.Furthermore, the computing device comprises a planning module which is designed to plan a path of the automatically moving medical device based on the predicted actions of the one or more obstacles.

[0063] The computing device may be configured to perform the method according to the method aspect. Alternatively or additionally, the computing device may comprise one or each of the features disclosed in the method aspect.

[0064] According to a second aspect of the device, an automatically moving medical device is provided. The automatically moving medical device comprises a computing device according to the first aspect of the device.

[0065] According to one system aspect, a system for planning a path of an automatically driving medical device is provided. The system comprises a sensor system. The sensor system comprises at least one sensor. The system for planning the path further comprises at least one automatically driving medical device according to the second device aspect (and / or comprising a computing device according to the first device aspect). The receiving interface of the at least one automatically driving medical device is configured to receive sensor data from the sensor system.

[0066] According to a further aspect, a computer program product is provided with program elements that cause a computing device (e.g., according to the first device aspect) to execute the steps of the method for planning a path of an automatically moving medical device according to the method aspect when the program elements are loaded into a memory of the computing device.

[0067] According to yet another aspect, a computer-readable medium is provided on which program elements are stored that can be read and executed by a computing device (eg according to the first device aspect) to perform steps of the method for planning a path of an automatically driving medical device according to the method aspect when the program elements are executed by the computing device.

[0068] The above-described properties, features, and advantages of the present invention, as well as the manner in which they are achieved, will become clearer and more understandable in light of the following description and the exemplary embodiments explained in more detail in conjunction with the drawings. This following description does not limit the invention to the embodiments included. Like components or parts may be provided with the same reference numerals in different figures. In general, the figures are not to scale.

[0069] It is understood that a preferred embodiment of the present invention may also be any combination of the dependent claims or the above embodiments with the respective independent claim.

[0070] These and other aspects of the invention will become apparent from and will be explained by reference to the embodiments described below. BRIEF DESCRIPTION OF THE DRAWINGS Fig. 1 is a flowchart of a method for planning a path of an automatically driving medical device according to a preferred embodiment of the present invention. Fig. 2 is an overview of the structure and layout of a computing device for planning a path of an automatically driving medical device according to a preferred embodiment of the present invention. Fig. 3A schematically shows an embodiment of path planning in an aisle with an obstacle based on the field of view of an automatically moving medical device. Fig. 3B schematically shows an embodiment of path planning in a corridor with obstacles based on external sensors in addition to the field of view of the automatically driving medical device. Fig. 4A shows a first embodiment of path planning of an automatically moving medical device in an operating room with a moving obstacle in a supply corridor. Fig. 4B shows a second embodiment of a path planning of an automatically moving medical device in an operating area with a moving obstacle in a supply corridor after determining that the shortest path from Fig. 4A is unsuitable. Fig. 4C shows the second embodiment of path planning from Fig. 4B after a few minutes, during which other moving obstacles, especially medical personnel, took the shortest route out of Fig. 4A, while the longer route from Fig. 4B remains free.

[0071] Any reference signs in the claims are not to be understood as limiting the scope of application.

[0072] Fig. 1 schematically shows a flow diagram of a computer-implemented method 100 for planning a path of an automatically moving medical device.

[0073] The method 100 comprises a step S102 of receiving sensor data regarding a possible range of movement of an automatically moving medical device.

[0074] The method 100 further comprises a step S104 of detecting a presence of one or more obstacles in the possible movement area based on the received S102 sensor data.

[0075] The method 100 further comprises a step S106 of predicting actions of the one or more obstacles detected S104 based on the received S102 sensor data. Prediction S106 takes into account medical-technical processes within the range of motion of the automatically moving medical-technical device.

[0076] Furthermore, the method 100 comprises a step S108 of planning a path of the automatically moving medical device based on the predicted S106 actions of the one or more obstacles.

[0077] Optionally, the method 100 comprises a step S101 of receiving a map of the movement area.

[0078] The method 100 may further include a step S103 of aggregating the received S102 sensor data.

[0079] Fig. 2 schematically shows a computing device 200 for planning a path of an automatically moving medical device.

[0080] The computing device 200 comprises a sensor data receiving interface 202, which is designed to receive sensor data detection module 204 regarding a possible range of movement of an automatically moving medical device.

[0081] The computing device 200 further comprises a detection module 204, which is designed to detect the presence of one or more obstacles in the possible movement range of the automatically driving medical device based on the received sensor data.

[0082] Computing device 200 further includes a prediction module 206 configured to predict actions of the detected one or more obstacles based on the received sensor data. The prediction is performed taking into account medical-technical processes within the possible range of motion of the automatically moving medical device.

[0083] Furthermore, the computing device 200 comprises a planning module 208 which is designed to plan a path of the automatically moving medical device based on the predicted actions of the one or more obstacles.

[0084] Optionally, the computing device 200 comprises a map receiving interface 201, which is designed to receive a map of the possible movement area of ​​the automatically driving medical device.

[0085] The computing device 200 may further include an aggregation module 203 configured to aggregate the received sensor data.

[0086] The sensor data receiving interface 202 and / or the optional card receiving interface 201 can be implemented by a transmit and receive interface 210.

[0087] The detection module 204, the prediction module 206, and / or the planning module 208 may be implemented by a processor 212 (e.g., a central processing unit, CPU, or a graphics processing unit, GPU). Optionally, the aggregation module 203 may also be implemented on the processor 212 (not shown). Fig. 2 shown).

[0088] The computing device 200 may further include a computer-readable memory 214.

[0089] The computing device 200 may be configured to execute the method 100.

[0090] The computing device 200 may be arranged in the automatically moving medical device for which the path is planned.

[0091] The technology according to the invention (in particular comprising the method 100, the computing device 200, an automatically driving medical device 300, and / or a system comprising a sensor system and the medical device 300) can also be referred to as predictive path planning in the clinical environment and / or as predictive navigation and path planning of autonomous mobile robots (and / or autonomous medical devices).

[0092] Unpredictable events occur in hospitals. These are detected by the inventive technology using internal and / or external sensors and integrated into path planning and navigation.

[0093] According to a first embodiment, unpredictable events are detected by sensors (e.g., cameras) and integrated into the path planning. Technically, this is done, for example, by cameras and / or LiDAR sensors outside the field of view of the automatically driving medical device. In particular, the cameras and / or LiDAR sensors can be located in another room. The detected unpredictable events are incorporated into the remote path planning. Conventional path planning is expanded.

[0094] Fig. 3A shows an example of an automatically moving medical device (short: device) 300 that enters a corridor (e.g., a hospital, a doctor's office, and / or a medical facility). The corridor includes walls 304 and openings 306 in the walls. Each opening 306 can be closed by a door (not shown). In the example of Fig. 3A, an obstacle 308 is located on the right side of the aisle in the direction of travel of the device 300. The planning of the path 302 in Fig. 3A includes the requirement to always drive on the right side of the aisle and to avoid obstacles 308 if they come into view. Fig. Path 302 shown in Figure 3A has abrupt changes of direction to meet this requirement.

[0095] Fig. Figure 3B shows the same corridor with walls 304 and openings 306 as well as the obstacle 308 as in Fig. 3A. In addition, in the embodiment of the Fig. 3B, a plurality of sensors 310 (in particular cameras and / or LiDAR sensors) are arranged along the aisle, so that the obstacle 308 can be detected even before the device 300 enters the aisle. For the planning of the path 302', a requirement can again be to always drive to the right in the aisle, if possible, and to avoid it if an obstacle 308 comes into view. In addition, an obstacle 302 in the distance can already be integrated into the planning of the path 302'. The path 302' in Fig. 3B includes, for example, fewer changes of direction before the obstacle 308 than the path 302 in Fig. 3A.

[0096] According to a number of further embodiments, the planning of the path of the automatically moving medical device comprises an integration into an operating (OR) system in order to include the sequence of an operation and / or in an OR area (comprising, for example, one or more supply corridors, operating rooms and / or operating rooms, preparation rooms, changing rooms, rooms for materials and / or rooms for mobile devices that can be used variably) in the path planning.

[0097] When planning the path in an operating room, the automatically moving medical device (also known as a robot) 300 takes into account the (e.g., low) probability that people will leave the sterile area in the operating room system (also known as the OR). Alternatively or additionally, there is a (e.g., high) probability that a mobile C-arm (e.g., as another medical device) will be moving, and / or that the route from the patient to the changing room is (e.g., more frequently) used.

[0098] Certain events, such as triggering an image of the C-arm (e.g., as another medical device) and / or the command to return the C-arm (e.g., to a parking position), can also indicate that the operation is over and there is now a very high probability that people (e.g., the operating room staff, in particular the surgeon and / or assistants) will leave the sterile area. Alternatively or additionally, triggering the image and / or the command to return the C-arm can indicate that other people (e.g., nurses) who are collecting the patient are about to enter the operating room (and / or room). The automatically moving medical device (also: robot) 300 can take this into account, in particular each of the events and the resulting changes in the movement area (in particular, expected people as moving obstacles), and plan a path (and / or its routes) accordingly.

[0099] In one embodiment, the automatically driving medical device (also: robot) 300 can be capable of distinguishing between staff (e.g., doctors and / or nurses) and patients and adapting its behavior accordingly. The automatically driving medical device (also: robot) 300 can, for example, maintain a small (and / or shorter) distance from staff and / or respond to instructions from the staff (e.g., via robot-human interaction). Alternatively or additionally, the automatically driving medical device (also: robot) 300 can, for example, maintain a larger (also: longer) distance from a patient and / or not accept any instructions from the patient.

[0100] Fig. 4A shows an example of planning a path in an operating room (also: operating theatre), in which an intended, in particular short, path is taken by the automatically moving medical device (also: robot) 300.

[0101] The operating room in Fig. 4A includes several supply corridors 404, changing rooms 412, preparation rooms 408 and operating rooms 410. The exemplary operating room area in Fig. 4A further includes a room 414 for materials (e.g., including medications, bandages, syringes, surgical supplies such as sutures, and / or surgical instruments) and a room 416 provided for mobile devices that are used in a variety of ways. The mobile devices can, for example, include the automatically moving medical device (also: robot) 300.

[0102] In Fig. 4A shows an obstacle (e.g., a patient bed) in one of the supply aisles 404. The automatically moving medical device (also: robot) 300 takes the shortest route 402 to the planned location, one of the operating rooms 410.

[0103] In Fig. 4A also schematically depicts walls 304 and doors 406 of varying widths. Doors 406 to the changing rooms 412, for example, can be narrow and designed only for people. Doors 406 to preparation rooms 408, operating rooms 406, the material room 414, and / or the equipment room 416 can, for example, be wide enough to allow a path for the automatically moving medical device (also: robot) 300.

[0104] Each of the Fig. The doors 406 shown in Figure 4A may be openable by means of a swinging door and / or a sliding door. Path 402 may be designed for open doors.

[0105] Fig. 4B shows the surgical area of ​​the Fig. 4A with an alternative planned path 402' of the automatically moving medical device (also: robot) 300 to the location in one of the operating rooms 410. The alternative path 402' can be planned if a hectic situation is detected in the operating room area (also: operating room wing), in particular based on the received sensor data regarding the possible range of movement of the automatically moving medical device (also: robot) 300.

[0106] A hectic situation can arise if, for example, the patient exhibits critical vital signs and / or becomes unstable. Vital signs are usually monitored during surgery and can be evaluated, as most patients are connected to a monitoring center via monitoring devices.

[0107] Alternatively or additionally, vital signs can be recorded using a camera system, for example, in the ceiling of the operating room (especially operating room 410). In this case, the automatically moving medical device (also known as a robot) 300 avoids the busy area, as well as aisles 404 near the busy area, in order not to block potentially approaching personnel.

[0108] Normally, the operating room supply corridor 404 should be kept clear for large equipment. However, since the inventive technology in the exemplary embodiment takes into account, for example, the patient's vital parameters, the automatically moving medical device (also known as a robot) 300 proactively avoids the hectic area.

[0109] Fig. 4C shows the hectic situation Fig. a few (e.g. two) minutes later. The emergency (e.g. due to critical vital parameters of the patient) creates hectic activity and the automatically moving medical device (also: robot) 300 would now be in the way. For example, in Fig. 4C at reference number 418 persons (in particular additional medical personnel) in the operating room supply corridor 404.

[0110] By way of example, the method 100 (and / or the planning algorithm) can comprise the following steps. Based on a map recorded in advance (e.g., using a laser scanner), the best path is planned (e.g., with regard to static obstacles). The path can be refined and / or controlled using live data (in particular from external sensors) available along the relevant possible movement areas (e.g., as steps S102 to S108). Alternatively or additionally, the automatically driving medical device (also: robot) 300 can be driven using real-time data from the sensors (also: real-time sensor data) on the device 300. Optionally, data from external sensors that are not attached to the device 300 are also included.

[0111] The sensors can be permanently installed in the rooms and / or arranged on other mobile devices. Data transmission from the sensors to the computing device 200 and / or the automatically moving medical device (also: robot) 300 can be wireless (e.g., via a cellular network, Wi-Fi, Bluetooth, and / or ZigBee).

[0112] For predictive path planning (also: navigation), the information from a large number of different sensors (particularly internal and / or external) can be combined into a coherent environment model (and / or "world model"). The automatically driving medical device (also: autonomous mobile robot) 300 can access the environment model (and / or world model) (e.g., via a cloud connection) and optimize the path planning (and / or navigation) accordingly. Alternatively or additionally, for the automatically driving medical device (also: autonomous mobile robot and / or autonomous medical device) 300, the environment, immediate obstacles (particularly detected by sensors on the device and / or robot 300, e.g., in step S104) in the immediate vicinity, and mobile and / or portable obstacles along the entire path (particularly detected by sensors such as a camera in the long-range, e.g.,in the neighboring room, within the entire map, also: map, in which the device and / or the robot 300 can move, e.g. in step S104).

[0113] The external sensors can be installed in the rooms and / or on other mobile devices. For example, two mobile C-arms can each comprise one or more sensors and can be tracked. Alternatively or additionally, the speed and / or planned path of a mobile C-arm can be extrapolated. The extrapolation of the speed and / or planned path of the mobile C-arm can be fed into the environment model (and / or world model) so that other mobile devices (in particular the automatically driving medical device 300) consider the extrapolated speed and / or the extrapolated planned path as parameters in the planning.

[0114] To anticipate foreseeable situations, accumulated process knowledge for standard procedures in the operating room (especially in the operating room) can be incorporated into planning. For example, a standard procedure might include preparing for a surgery, performing the surgery, post-operative care, and / or imaging during a surgery. During imaging, staff may need to leave the operating room for a short time, and doors may need to remain clear.

[0115] Each situation and / or standard procedure may require the automatically moving medical device (also: mobile robot) 300 to adjust its current position and / or path planning so that the device 300 does not disrupt the course of the operation (and / or the situation and / or the standard procedure).

[0116] Alternatively or additionally, a natural human movement pattern can be incorporated into path planning. For example, the automatically moving medical device (also known as a mobile robot) 300 can preferentially (and / or always, especially if there are no obstacles in the way) move to the right of the room.

[0117] Furthermore, alternatively or additionally, situations can be included in the planning of the path, such as the anticipation of hectic activity due to (e.g. unhealthy and / or critical values ​​of) the patient's vital parameters.

[0118] The technology according to the invention can comprise the automatically driving medical device (also: autonomous mobile robot and / or autonomous medical device) 300. The device 300 can be designed to carry out the method 100 (and / or comprise a computing device 200). For path planning, a possible movement area (also: environment), the (e.g. immediate) obstacles (in particular detected by sensors on the device and / or robot 300, in particular in step S104) in the immediate vicinity and (e.g. portable and / or mobile) obstacles in the entire path (in particular detected by sensors such as a camera in the long-range area, e.g. in the adjacent room, within the entire map and / or map in which the device and / or robot 300 can move, in particular in step S104) can be included. Based on detected obstacles, the path (and / or behavior) of the device and / or robot 300 can change, e.g. byit drives slower.

[0119] Sensors can be used to record scenarios in the operating area (and / or operating room), and the movement sequences (and / or behavioral sequences) can be evaluated statistically and / or using artificial intelligence (AI). During the evaluation, the movement sequence (and / or behavior) can be linked to specific roles in the operating room. For example, people standing in the sterile area can fulfill a different specific role than people standing outside the sterile area. The data (e.g., the evaluation of the movement sequence and / or behavior) can be played back and used for path planning of the device and / or robot 300.

[0120] Alternatively or additionally, clinical processes can be integrated, such as an elevator ride to the floor where the automatically moving medical device (also: autonomous mobile robot and / or autonomous medical device) 300 is located, and the disembarkation of passengers and / or the unloading of a patient bed, cleaning materials, and / or medical devices. Planning the path may include the automatically moving medical device (also: autonomous mobile robot and / or autonomous medical device) 300 not traveling directly in front of the elevator door, but rather avoiding it.

[0121] If hectic activity is detected in the operating room (e.g., in the operating theater), a slowed-down movement can be planned. Alternatively or additionally, if it is detected that an X-ray will soon be taken in the operating room, the device and / or the robot 300 can clear a path (e.g., for medical personnel leaving the room, in particular the operating theater, during the X-ray) and / or open a door. Furthermore, alternatively or additionally, if it is detected that a C-arm and / or a patient is being pushed out of the operating room (and / or operating theater, generally: room), it can be planned that the device and / or the robot 300 will not move into the room.

[0122] The inventive technology enables natural, energy-saving, and / or (particularly hardware-friendly) movements of the automatically driving medical device (also known as an autonomous system) and / or minimizes wear. Alternatively or additionally, predictive driving of the automatically driving medical device enables a path to be cleared and improves safety, particularly for medical personnel.

[0123] The inventive technology can be observed during path planning. For example, signals from internal and / or external sensors as well as the operation of software (in particular detection, prediction, and / or planning) can be characteristic of reproducible situations, e.g., driving the automatically driving medical device around a corner. A comparison of a path (and / or route) with and without an obstacle that cannot be detected by sensors on the chassis can be characteristic of the inventive technology.

[0124] Unless already explicitly described, individual embodiments or their individual aspects and features described with reference to the drawings may be combined or exchanged with one another without limiting or expanding the scope of the described invention, provided such a combination or exchange is reasonable and in the spirit of the present invention. Advantages described with reference to a specific embodiment of the present invention or with reference to a specific figure are, wherever applicable, also advantages of other embodiments of the present invention. Cited documents [1] Alessandro DiFava. “Localization and path planning”. Wiki: Robots / TIAGo++ / Tutorials / Navigation / Localization (last edited 2023-03-30) [2] http: / / wiki.ros.org / ; http: / / docs.ros.org / [3] Haist, Tobias, Reichert, Carsten, Würtenberger, Felicia, Lachenmaier, Lena and Faulhaber, Andreas. “Camera-based recording of vital parameters” tm - Technisches Messen, vol. 86, no. 7-8, 2019, pp. 354-361. List of reference symbols 100 procedures S101 Step of receiving a card S102 Step of receiving sensor data S103 Step of aggregating the sensor data S104 Obstacle detection step S106 Step of predicting actions of obstacles S108 Step of planning a path 200 Calculating device 201 Card receiving interface 202 Sensor data reception interface 203 Aggregation module 204 Detection module 206 Predicating module 208 Planning module 210 Send and receive interface 212 processor 214 memory 300 Automatically moving medical device 302; Path of the automatically moving medical device 302' 304 Wall 306 Opening 308 Obstacle 310 Sensor, e.g. camera 402; Path of the automatically moving medical device 402' 404 Supply corridor for operating area 406 Door 408 Preparation room 410 Operating Room 412 Changing room 414 Room for materials 416 space for variable (e.g. mobile) devices 418 Other variable obstacle, in particular person QUOTES CONTAINED IN THE DESCRIPTION

[0000] This list of documents submitted by the applicant was generated automatically and is included solely for the convenience of the reader. This list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions. Cited non-patent literature

[0000] Tobias Haist et al. in "Camera-based recording of vital parameters" tm - Technisches Messen, vol. 86, no. 7-8, 2019, pp. 354-361

[0005] http: / / wiki.ros.org /

[0006] http: / / docs.ros.org /

[0006] http: / / wiki.ros.org / ; http: / / docs.ros.org /

[0124] Haist, Tobias, Reichert, Carsten, Würtenberger, Felicia, Lachenmaier, Lena, and Faulhaber, Andreas. "Camera-based recording of vital parameters." tm - Technisches Messen, vol. 86, no. 7-8, 2019, pp. 354-361

[0124]

Claims

[1] Computer-implemented method (100) for planning a path of an automatically moving medical device, comprising the steps: - receiving (S102) sensor data regarding a possible range of movement of the automatically moving medical device; - detecting (S104) the presence of one or more obstacles in the possible movement area based on the received (S102) sensor data; - predicting (S106) actions of the detected (S104) one or more existing obstacles on the basis of the received (S102) sensor data, wherein the prediction (S106) takes into account medical-technical processes in the movement range of the automatically moving medical-technical device; and - Planning (S108) a path of the automatically moving medical device based on the predicted (S106) actions of the one or more obstacles. [2] The method (100) of claim 1, further comprising the step: - Receiving (S101) a map of the movement area. [3] Method (100) according to one of the preceding claims, wherein the sensor data is received by means of a wireless data connection from at least one sensor. [4] Method (100) according to one of the preceding claims, wherein the sensor data are received (S102) by a sensor system, wherein the sensor system is optical, acoustic, resistive, mechanical, inductive, magnetic, thermal, piezoelectric, photoelectric and / or thermoelectric and / or wherein sensors of the sensor system are arranged outside a currently optically visible area of ​​the automatically moving medical device. [5] Method (100) according to one of the preceding claims, wherein the sensor data comprise control instructions of another medical device, in particular a mobile C-arm. [6] Method (100) according to one of the preceding claims, wherein the sensor data comprise vital parameters of a patient in the range of motion. [7] Method (100) according to one of the preceding claims, further comprising the step: - Aggregating (S103) the received (S102) sensor data. [8] Method (100) according to one of the preceding claims, wherein the step of detecting (S104) further comprises a person recognition, optionally comprising a person identification, wherein the person recognition is designed to distinguish medical personnel from non-medical persons, in particular a patient. [9] Method (100) according to one of the preceding claims, wherein the consideration of the medical-technical processes is based on historical data. [10] Method (100) according to the directly preceding claim, wherein the received sensor data and / or the historical data are statistically recorded and / or wherein the received sensor data and / or the historical data are processed by an artificial intelligence, AI, in particular to determine the medical-technical processes. [11] Method (100) according to one of the preceding claims, wherein the planning (S108) of the path relies on standard specifications, in particular in the absence of detected (S104) obstacles. [12] Computing device (200) for planning a path of an automatically moving medical device, comprising: - A sensor data receiving interface (202) designed to receive sensor data regarding a possible range of movement of an automatically moving medical device; - A detection module (204) configured to detect the presence of one or more obstacles in the possible movement area based on the received sensor data; - A prediction module (206) designed to predict actions of the detected one or more obstacles on the basis of the received sensor data, wherein the prediction takes place taking into account medical-technical processes in the movement range of the automatically moving medical-technical device; and - A planning module (208) configured to plan a path of the automatically moving medical device based on the predicted actions of the one or more obstacles. [13] Computing device (200) according to the immediately preceding claim, further comprising one or each of the features according to method claims 2 to 11, and / or wherein the computing device (200) is configured to perform one or each of the steps according to method claims 2 to 11. [14] Automatically moving medical device (300) comprising a computing device (200) according to claim 12 or 13. [15] System for planning a path of an automatically moving medical device, comprising: - A sensor system comprising at least one sensor; and - At least one automatically driving medical device (300) according to the directly preceding claim, wherein the receiving interface (202) of the at least one automatically driving medical device (300) is designed to receive sensor data from the external sensor system.

Images