Radiation-based medical system and monitoring device

CN115624339BActive Publication Date: 2026-09-25SIEMENS HEALTHINEERS AG
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
CN202210817606.9
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Priority Date
2021-07-14
Filing Date
2022-07-12
Publication Date
2026-09-25
Estimated Expiration
2042-07-12

AI Technical Summary

Technical Problem

通常每天必须为校准周期或检查安排时间,在此期间不能使用医疗系统,并且必须至少有人在场以开始校准

Benefits of technology

[0024]用于检查和/或校准基于辐射的医疗系统的系统、监测装置和方法的所有优点和特征,可以类似地转用至用于训练处理单元的方法,反之亦然。

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN115624339B_ABST
    Figure CN115624339B_ABST
Patent Text Reader

Abstract

The invention relates to a radiation-based medical system (1) and a monitoring device, the medical system comprising a monitoring device for checking the presence of a person, a control unit (4) and a radiation generating device (6), the monitoring device comprising at least one sensor device (8) and a processing unit (10), the at least one sensor device (8) being configured to detect sensor data from a treatment room in which the system is arranged and to transmit them to the processing unit (10), the processing unit (10) being configured to determine from the sensor data whether a person (12) is present in the treatment room and to transmit a start signal to the control unit (4) in the event of at least one predetermined condition occurring and on the proviso that no person (12) has been determined, the control unit (4) being configured to execute, in response to the transmitted start signal, a check cycle of the radiation generating device (6) and to generate radiation, in particular X-ray radiation.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to a radiation-based medical system, a monitoring device, a method for inspecting and / or calibrating a radiation-based medical system, and a method for training a processing unit. Background Technology

[0002] Many medical systems require regular calibration cycles or inspections, during which various automated calibrations are performed. For radiation-based systems, such as CT scanners, radiation, particularly X-rays, is typically triggered. Therefore, the calibration cycle for these systems is usually subject to safety regulations. In particular, no one must be present in the room during calibration; therefore, calibration typically cannot be initiated automatically but must be started by a person trained in radiation protection. This takes time, depending on the duration of the calibration. Typically, time must be scheduled daily for calibration cycles or inspections during which the medical system cannot be used and at least one person must be present to begin the calibration. Summary of the Invention

[0003] Therefore, one object of the present invention is to provide a medical system that can perform regular calibration cycles in the most time-saving manner and with minimal interference to routine operating procedures.

[0004] According to one aspect of the invention, a radiation-based medical system, particularly a computed tomography (CT) system, is provided, comprising a monitoring device for examining the presence of a person, a control unit, and a radiation generating device. The monitoring device includes at least one sensor device and a processing unit. The at least one sensor device is configured to detect sensor data from a treatment room in which the system is arranged and transmit it to the processing unit. The processing unit is configured to determine, based on the sensor data, whether a person is present in the treatment room, and, upon the occurrence of at least one predetermined condition and provided that no one has been determined to be present, transmit a start signal to the control unit. The control unit is configured to, in response to the transmitted start signal, execute and / or trigger an examination cycle of the radiation generating device and generate radiation, particularly X-ray radiation. In this context, "radiation-based" specifically refers to the system generating radiation, particularly X-ray radiation, at least occasionally during its operation and / or when an examination cycle needs to be executed. Advantageously, the system can be used to automatically identify the presence or absence of a person, so that the examination cycle can be started, particularly without human interaction. In particular, the presence of a trained person (e.g., in a control room) is not required to start the examination cycle. The examination cycle can be adjusted to execute automatically at night and / or when the medical system is not to be used in any other way. For example, the medical system can also be used during normal working hours or treatment times, especially during the day, to save time. For instance, 15 to 30 minutes can be available each day, during which time the system can also be used for treatment, examination, and / or diagnosis.

[0005] The monitoring device includes at least one sensor device. According to a preferred embodiment, the monitoring device may include multiple, preferably 3 to 5, and particularly preferably 4 sensor devices. Multiple sensor devices can achieve coverage of the entire room, especially preventing blind spots that might not be detectable by a single sensor device. 3 to 5 sensor devices are generally well-suited for complete room coverage. In particular, it has been found that 4 sensor devices represent a good solution for typical treatment room configurations, enabling coverage of all areas and ensuring that the sensor devices are sufficiently close to all relevant locations in the treatment room for reliable human detection. At least one sensor device can be a camera, particularly an infrared camera. At least one sensor device can be arranged on the ceiling and / or in the corners of the treatment room. In particular, multiple sensor devices can be arranged so that people throughout the treatment room can be detected using these sensor devices. At least one sensor device can be integrated into an existing camera system, which can be used, for example, to monitor patients and / or check patient positioning. For example, an existing camera can be used as a sensor device. Alternatively or additionally, additional sensors, particularly infrared sensors, can be designed as upgrades to existing cameras and / or camera modules. Using existing cameras or camera systems can advantageously reduce installation workload. In particular, existing systems can be upgraded to the proposed medical system relatively easily or at a relatively low cost in this manner. Sensor data detected by at least one sensor device can be, in particular, image data, preferably infrared or thermal image data. Sensor data can include time series of images, especially image data showing processes of change over time, including infrared video data.

[0006] The processing unit can communicate with at least one sensor device, particularly wirelessly or via a wired connection. The processing unit may include a processor unit and / or a memory unit, specifically for processing sensor data from at least one sensor device. The processing unit may be a microcontroller or part of a microcontroller. Alternatively or additionally, the processing unit may be integrated into a computer unit such as a personal computer or tablet computer. According to one embodiment, the processing unit can be designed autonomously, particularly separately from the main computer of the medical system. The processing unit may preferably be located in a treatment room or a room adjacent to the treatment room. In particular, the autonomous design allows for easy upgrading of existing systems. On the other hand, this can sometimes prevent the already considerable complexity of the system's main computer, and consequently, its inherent error variability, from being further increased. The processing unit is particularly configured to determine the presence and / or absence of people using sensor data. The absence of a person specifically means that no one is present in the treatment room, while the presence specifically means that at least one person is present in the treatment room. For example, the processing unit may include, and / or have access to, a database for classifying objects (e.g., equipment, tables, and / or walls) and people located in the room. This classification may include identification criteria for identifying objects and / or people. Identification criteria may be, for example, geometric patterns, movement sequences (e.g., moving a movable treatment table), thermal characteristics (e.g., heating equipment during operation), and / or positioning or arrangement within the treatment room. The processing unit can be configured to distinguish between objects and people using the identification criteria, and more specifically, to determine the presence or absence of a person in the treatment room. The processing unit can be configured to determine the presence of a person given an assigned identification criterion. The processing unit can be configured to infer the presence of a person based on the identification criterion exceeding a predetermined threshold. The processing unit and / or monitoring device can be configured to perform presence checks and / or detect sensor data only when predetermined conditions occur, particularly at predetermined time points. The predetermined conditions can be stored on the processing unit, particularly in the processing unit's storage unit. Predetermined conditions may be, for example, time points, particularly at night. For example, it can be specified that a check cycle is performed periodically at 3:00 AM, and correspondingly, a presence check is performed at or shortly before this time (e.g., 1-10 seconds prior). The processing unit can be configured to automatically determine the appropriate time points for the check cycle, particularly by means of past detected person presence times and / or system uptime. Alternatively or additionally, the specified time points of the inspection cycle can be set by the user.

[0007] The processing unit is preferably communicatively connected to the control unit, particularly via a wireless or wired connection. The processing unit can be connected directly to the control unit or via an intermediate connection medium such as a router, hub, or additional computer unit. This connection may exist, for example, through an interface of the control unit, which is also used to receive emergency stop commands for the control unit's emergency stop function. The control unit can be configured to interrupt the inspection cycle via the emergency stop function. Advantageously, an existing interface for the emergency stop function can be used to communicate with the processing unit, thus allowing for relatively easy upgrades to existing systems with ongoing inspections. The control unit can be the main computer and / or control computer of the medical system. The control unit can be configured to monitor or control the inspection and / or treatment work of the medical system. In particular, the control unit is configured to execute and / or initiate inspection cycles for the medical system and / or its radiation-generating devices. Radiation-generating devices may include, for example, X-ray tubes. In particular, the inspection cycle may also include calibration of the radiation-generating devices and / or the medical system. The inspection cycle can be used to ensure the quality of medical system functions, such as imaging. Radiation, particularly X-ray radiation, may occur during the examination cycle. Due to safety regulations, this usually means that no one should be in the treatment room during the examination cycle. This can be advantageously ensured, especially through monitoring devices.

[0008] According to one embodiment, the control unit may include a storage unit for storing predetermined time points and / or include an input interface including an option to input the predetermined time points, particularly by user input. Specifically, the control unit may be configured to send a request signal to the processing unit at the predetermined time point, the request signal initiating a check for the presence of a person in the treatment room. The request signal may be a predetermined condition of the processing unit. The control unit may be advantageously configured to allow the user to input a schedule of check cycles into the control unit, wherein the processing unit is specifically configured to perform an automatic presence check at the start of the check cycle. The processing unit may be configured to interrupt the check cycle once a person is detected in the treatment room. The interruption may be transmitted to the control unit in the form of an emergency stop function. For example, the processing unit may be configured to prevent the check cycle by persistently activating the emergency stop function transmitted to the control unit when a person is present. In this case, the start signal may be a deactivation of the transmitted emergency stop function.

[0009] According to one implementation, the at least one predetermined condition may be: a predetermined time point when the system is not normally in use; a specified minimum time interval exceeding the interval since the last executed examination cycle; a pause in the transmission of treatment and / or examination plans to the processing unit; a malfunction of the radiation generating device; and / or at least one operating parameter of the radiation generating device exceeding or falling below a predetermined limit. The predetermined time point may, for example, be a fixed time, particularly at night. The specified minimum time interval can be used to ensure that examination cycles are not executed too frequently and / or too infrequently. It is conceivable that several conditions are summarized in the predetermined conditions, for example, a pause detected in the treatment and / or examination plan combined with a detected and / or transmitted malfunction. For example, the time point may be a daily recurring time point, particularly a specific time, and / or a variable time point depending on the day of the week.

[0010] According to one embodiment, the inspection cycle may include heating the X-ray tube, restoring or improving the vacuum in the radiation generating device, checking the function of the X-ray tube, checking and / or adjusting the operating parameters of the radiation generating device, and / or testing system function. Heating the X-ray tube can be used to improve or restore the vacuum in the X-ray tube. For example, a getter material such as titanium can be heated to capture or bind gas particles in the vacuum. Additionally or alternatively, the inspection cycle may include checking and / or adjusting the operating parameters of the medical system, particularly the radiation generating device and / or detectors. Therefore, it is advantageous to automatically check or improve the operation or function of key components.

[0011] According to one embodiment, the sensor device and / or monitoring device may include one or more, particularly three to five infrared sensors and / or infrared cameras. Advantageously, infrared sensors are relatively inexpensive to obtain. Infrared sensors or cameras can advantageously enable the identification of people in a room in a particularly simple manner by providing infrared images, since people can be identified as heat sources. In particular, the processing unit may include a database for classifying heat sources, particularly those caused by people and objects, and / or have access to such a database. The processing unit may be configured to distinguish typical thermal images generated by people from those generated by objects. Therefore, infrared sensors and / or cameras can be a relatively simple, inexpensive, and simultaneously reliable option for checking the presence of people.

[0012] According to one embodiment, one or more sensor devices can be arranged in the treatment room such that the distance between each location in the treatment room and the nearest sensor device is at most 3-8 meters, preferably at most 4-6 meters. A distance of 3-8 meters, especially 4-6 meters, allows for particularly reliable identification of people and / or differentiation of people from objects using the detected data, especially image data. A distance of 4 to 6 meters also allows for the use of low-resolution sensor devices, especially infrared cameras, while still ensuring reliable identification of people.

[0013] According to one embodiment, the sensor device can be configured to capture low-resolution images. Specifically, the sensor device can be configured to capture images, preferably infrared images, at a resolution ranging from 4×4 pixels to 64×64 pixels, preferably from 8×8 pixels to 32×40 pixels. Due to the low resolution, the sensor device can be manufactured inexpensively. A resolution of 4×4 pixels to 64×64 pixels allows for the generation of image material requiring minimal storage space and processed or analyzed by a processing unit with relatively low computational load. In particular, the processing unit can be relatively simple, especially equipped with low computational power. A resolution of 8×8 pixels to 32×40 pixels also enables reliable identification of the presence of people, particularly moving and stationary people, while excluding individual identification. The inability to identify individuals may be advantageous for data protection and the maintenance of individual rights, especially by not generating related personal data.

[0014] According to one embodiment, the monitoring device can be configured to record the temporal progression of detected thermal radiation, wherein the processing unit is specifically configured to distinguish between the sudden appearance of heat, particularly due to a person entering the treatment room, and the slow heating, particularly of equipment in operation. In the environment of medical systems such as CT systems, there are typically numerous heat sources, making it difficult to distinguish objects from people based on thermal radiation alone. Observing the temporal progression of heat can provide a good solution to this problem. When a person enters the treatment room, an immediate heat source, particularly one at body temperature, can be detected. In contrast, equipment is expected not to reach its operating temperature immediately upon being put into operation, but rather to gradually heat up over a certain period, such as 1 to 10 minutes. For example, an automatically moving workbench with a motor can only slowly reach its operating temperature when put into operation. Thermal radiation emitted from walls typically also changes very slowly throughout the day. Therefore, the temporal progression of detected thermal radiation, particularly with regard to time-related thermal progression, can be a relatively easy means to distinguish between people and objects entering the room. The processing unit can be configured to classify objects and people using the temporal progression of heat.

[0015] According to one implementation, the processing unit can be trained via machine learning to recognize the presence of a person and distinguish them from inanimate components or objects. For example, the system may have been running for some time, with inspection cycles manually initiated. The processing unit can be configured to use the time period of the executed inspection cycle as a reference for the absence of people during the learning phase. Alternatively or additionally, the processing unit can be configured to use the time period in which the inspection was performed as a reference for the presence of at least one person during the learning phase. Alternatively or additionally, the processing unit can also be trained by having the user manually input the number of people present in the room. The system can therefore adapt particularly well to new environments, especially treatment rooms.

[0016] According to one implementation, the processing unit can be trained using machine learning to determine predetermined time points when the system is not typically in use. The processing unit can be configured to determine appropriate time points for executing check cycles, using manually executed check cycles and / or records of typical pause times. Training via machine learning allows for particularly simple system setup.

[0017] According to one embodiment, the system may include at least one emergency stop function, which is activated, in particular, via an emergency stop switch and / or a door contact at the entrance door of the treatment room. The processing unit and / or control unit are configured to trigger the emergency stop function when they detect the presence of at least one person. The door contact may be a contact at the entrance door of the treatment room, triggered, for example, by a door handle or by opening the door, and is designed to detect a person entering the treatment room. The processing unit may be configured to continue checking for the presence of a person during a check cycle, and, if it determines that at least one person is present, to interrupt the check cycle, in particular, by sending a signal to the control unit. The control unit may detect the presence of a person indirectly via a signal from the processing unit. The control unit may include an interface for receiving the emergency stop function, wherein, in particular, an existing interface may be used. Advantageously, the interface of the control unit configured to receive signals from the emergency stop function may also be used for signals from the processing unit. In particular, it may be specified that the control unit does not receive any information regarding the cause of the emergency stop function, i.e., a cause, in particular, caused by the emergency stop switch, the door contact, and / or the processing unit. Therefore, it is advantageous to upgrade an existing system using such an interface at a relatively low cost.

[0018] According to one implementation, the processing unit can be configured to trigger the detection only randomly before and / or upon the occurrence of at least one predetermined condition. Advantageously, energy can thus be saved by specifying the detection and processing of sensor data only when absolutely necessary.

[0019] According to another aspect of the invention, a monitoring device, particularly as described herein, can be provided for detecting a person in the environment and / or treatment room using a radiation-based medical system. This monitoring device includes at least one sensor device, particularly an infrared sensor and / or an infrared camera, and includes a processing unit configured to use the at least one sensor device to detect whether a person is present in the environment and / or treatment room, and to transmit information about the person's presence to the system's control unit. All the advantages and features of the system can be similarly transferred to the monitoring device, and vice versa. The processing unit can be configured to communicate with an interface of the control unit as described herein. In particular, this interface can be an interface for activating an emergency shutdown command. Advantageously, the monitoring device can therefore be connected to existing interfaces of the medical system and is particularly useful for easily implemented system upgrades. The at least one sensor device can be configured to detect sensor data from the environment or treatment room and transmit it to the processing unit. The processing unit can be configured to determine whether a person is present in the environment or treatment room based on the sensor data, and to transmit a start signal to the system's control unit when at least one predetermined condition is met and when it has been determined that no one is present. Pre-defined conditions can be, in particular, pre-defined conditions as described in the context of a healthcare system.

[0020] According to another aspect of the invention, a method is provided for inspecting a radiation-based medical system, particularly the system described herein, in a treatment room. The inspection may include calibrating and / or optimizing the system. The method includes the following steps: a) Determine the possible presence of a person in the treatment room using a monitoring device that includes at least one sensor device, particularly using an infrared sensor and / or a camera; b) The monitoring device detects the occurrence of at least one predetermined condition, and upon the occurrence of the predetermined condition and provided that no one is present, transmits a start signal from the monitoring device to the system's control unit; and c) In response to the transmission start signal, the control unit initiates the inspection cycle of the system's radiation generating device and generates radiation.

[0021] All the advantages and features of the system and monitoring device can be similarly transferred to methods for inspecting and / or calibrating radiation-based medical systems, and vice versa.

[0022] The method may further include the following steps: determining the possible presence of a person in the treatment room during the examination cycle, and interrupting the examination cycle upon determining the presence of a person. Alternatively or additionally, the method may optionally include the following steps: optimizing the repetition time of the examination cycle by storing the presence time of the person and determining the periodically repeating time period when no one is present in the treatment room, and / or by storing the pause times during medical system operation and determining the periodically repeating pause times. Advantageously, the processing unit can thus learn over time when the examination cycle can be executed without interruption.

[0023] According to another aspect of the present invention, a method is provided for training a processing unit as described herein using machine learning. The method includes the following steps: a) Provide monitoring devices in the treatment room in conjunction with radiation-based medical systems; b) Run the system according to routine procedures for several consecutive days, including having the user manually start the check cycle at appropriate times; c) Based on the assumption that there is no person in the treatment room during the manually started examination cycle and / or at least one person in the treatment room during the examination, adjust the detection of people by the monitoring device through machine learning; d) Based on a manually selected start time and / or based on the system’s operating time and associated pause time, the appropriate start time for the inspection cycle may optionally be determined automatically through machine learning.

[0024] All the advantages and features of systems, monitoring devices, and methods used for inspecting and / or calibrating radiation-based medical systems can be similarly transferred to methods used for training processing units, and vice versa.

[0025] Routine operation can involve using the system for examinations and / or treatments during daily use. Specifically, during routine operation, examination cycles are manually initiated by at least one user. The timing of examination cycles and the duration of system usage can be advantageously used by the processing unit as a learning reference. Training takes place over several days, particularly at least one week. At least one week of training allows for the determination of weekly repetition cycles. It is also conceivable that training may take place over several weeks, such as two or three weeks, to detect deviations from the typical weekly cycles. Measurements of the treatment room dimensions and / or equipment located within the treatment room can be specified and input into the processing unit as a basis and / or aid to training. During training, the processing unit can, in particular, categorize equipment, walls, and / or people, which can serve as a basis for determining the presence of a person. During training, the processing unit can learn when it is best to ensure uninterrupted execution of examination cycles.

[0026] Advantageously, the monitoring device can be pre-trained in a training environment before being placed in the treatment room. For example, the monitoring device can be trained in a training room of similar size to the treatment room. For example, the dimensions of the treatment room can be measured in advance and / or existing equipment can be determined in order to design the training room accordingly. For example, the processing unit can be notified by manual user input whether there is someone in the training room, so as to correct or optimize the detection of people. Advantageously, the on-site training phase can be shortened through pre-training. Alternatively or additionally, the prescribed inspection time or usage time in the treatment room for pre-training can also be input into the processing unit.

[0027] The above features and implementation methods can be combined with each other, and the advantages associated with each feature also apply to the combination of these features. Attached Figure Description

[0028] Other advantages and features of the invention will become apparent from the following description of preferred embodiments of the subject matter according to the invention, with reference to the accompanying drawings. The following description is for illustrative purposes only and should not be construed as limiting the appended claims to any of these embodiments.

[0029] Figure 1 A schematic diagram of a radiation-based medical system according to an embodiment of the present invention is shown;

[0030] Figure 2 A schematic diagram of a method for inspecting and / or calibrating a radiation-based medical system according to an embodiment of the present invention is shown;

[0031] Figure 3 A schematic diagram of a method for training a processing unit according to an embodiment of the present invention is shown. Detailed Implementation

[0032] Figure 1A radiation-based medical system 1 in a treatment room is shown. Two walls 16 and a ceiling 14 of the treatment room are shown here. The radiation-based system includes a radiation-generating device 6, such as a CT scanner, with a control unit 4. The radiation-generating device 6 must be maintained regularly, particularly in the form of inspection cycles. Inspection cycles may include, for example, heating the X-ray tube, restoring or improving the vacuum, checking the function of the X-ray tube, checking and / or adjusting operating parameters and / or testing system functionality. Radiation, particularly radioactive radiation, may occur during these inspection cycles, therefore no one 12 is allowed in the treatment room during these cycles. According to the prior art, therefore, the user (e.g., a technician) must check that no one is present, and the inspection cycle must be started manually by the user. To avoid this necessity, the medical system 1 has a monitoring device, in this case, comprising multiple sensor devices 8 in the form of infrared cameras and a processing unit 10. These sensor devices 8 are arranged on the ceiling 14 or in the corner between the ceiling 14 and the walls 16. The sensor devices 8 acquire sensor data, particularly as infrared images, and transmit them to the processing unit 10. The sensor device 8 is preferably designed to acquire sensor data at a low resolution, such as 8×8 pixels or 32×40 pixels. As a result, relatively inexpensive sensors can be used, and the storage and processing workload is correspondingly lower. Furthermore, with low resolution, the personal rights of person 12 in the treatment room can be prevented from being violated by making it impossible to distinguish each person 12 individually. The processing unit 10 is configured to use the sensor data to determine whether person 12 is in the room. This can be accomplished, in particular, by analyzing the time course of thermal radiation detected by the sensor device 8. When predetermined conditions are met, particularly when a specific time point is reached, such as 3:00 AM, if the processing unit 10 has determined that no person 12 is in the treatment room, the processing unit 10 sends a start signal to the control unit 4. It can be specified here that the presence of person 12 is checked only when predetermined conditions are met, in order to save energy, for example. The control unit 4 then initiates a check cycle based on the start signal. If the monitoring device detects that at least one person 12 is in the treatment room, the check cycle is interrupted when the door contact of the door 20 leading to the treatment room indicates that person 12 is expected to enter the treatment room, or when the emergency stop switch 22 is triggered. The emergency stop switch 22 or an additional emergency stop switch 22 may also be located in another room, such as an adjacent room.

[0033] Figure 2A schematic diagram of a method for inspecting a radiation-based medical system 1 according to an embodiment of the present invention is shown. In a first step 101, a monitoring device determines whether someone may be present in the treatment room. In a second step 102, it is checked whether a predetermined condition has been met, such as reaching a certain time point. If the condition has been met, and no person is detected in the treatment room during the inspection, the monitoring device sends a start signal to the control unit 4. In the next step 103, the control unit 4 starts the inspection cycle of the radiation generating device 6 in response to the start signal.

[0034] Figure 3 A schematic diagram of a method for training a processing unit 10 according to an embodiment of the present invention is shown. In a first step 202, a monitoring device and a radiation-based medical system 1 are provided in a treatment room. In a second step 203, the system operates periodically for several days, wherein a control cycle is manually initiated by the user. Specifically, during the control cycle, the user checks whether there is no one 12 in the treatment room. In the next step 204, the monitoring device or the processing unit 10 of the monitoring device automatically adapts to the detection of a person 12 in the treatment room through machine learning, in that it assumes that there is no one 12 in the treatment room during the manually initiated check cycle. Optionally, in a further step 205, a suitable start time for the check cycle can also be automatically determined through machine learning. In particular, the start time and / or pause time selected by the user in the operation of the medical system 1 can serve as a basis. Furthermore, optionally, the monitoring device can be pre-trained in the preceding step 201 before being placed or provided in the treatment room, for example, by being trained in a test environment. In this way, the training time in the treatment room can be shortened in particular.

Claims

1. A radiation-based medical system (1) comprising a monitoring device for checking the presence of a person, a control unit (4) and a radiation generating device (6). in, The monitoring device includes at least one sensor device (8) and a processing unit (10). The at least one sensor device (8) is configured to detect sensor data from the treatment room where the medical system is arranged and transmit it to the processing unit (10). The processing unit (10) is configured to determine whether there is a person (12) in the treatment room based on the sensor data, and to transmit a start signal to the control unit (4) when at least one predetermined condition is met and on the premise that it has been determined that there is no one (12). The control unit (4) is configured to, in response to the transmitted start signal, execute the inspection cycle of the radiation generating device (6) and generate radiation. The processing unit (10) is trained in the following manner: a) Provide the monitoring device in the treatment room together with the medical system; b) Run the medical system for several consecutive days as usual, including having the user manually start the examination cycle at appropriate times; c) Based on the assumption that no one (12) is in the treatment room during the manually started inspection cycle, the monitoring device is adjusted to detect the person (12) by machine learning; d) Based on a manually selected start time and / or based on the operating time and associated pause time of the medical system, the appropriate start time of the examination cycle is automatically determined by machine learning.

2. The medical system according to claim 1, in, The at least one predetermined condition is: the occurrence of a predetermined time point when the medical system is not normally used; the interval between the last performed examination cycle and a predetermined minimum time interval is exceeded; a treatment and / or examination plan transmitted to the processing unit (10) is suspended; the radiation generating device (6) malfunctions; or at least one operating parameter of the radiation generating device (6) exceeds or falls below a predetermined limit.

3. The medical system according to claim 1 or 2, in, The inspection cycle includes: heating an X-ray tube; restoring or improving the vacuum in the radiation generating device (6); checking the function of an X-ray tube; checking and / or adjusting the operating parameters of the radiation generating device (6); and / or testing the system function.

4. The medical system according to claim 1 or 2, in, The sensor device (8) includes one or more infrared sensors and / or infrared cameras.

5. The medical system according to claim 1 or 2, in, One or more sensor devices (8) are arranged in the treatment room such that the distance between each location in the treatment room and the nearest sensor device (8) is at most 4-6 meters.

6. The medical system according to claim 1 or 2, in, The sensor device (8) is configured to capture images at a resolution ranging from 4×4 pixels to 64×64 pixels.

7. The medical system according to claim 1 or 2, in, The monitoring device is configured to record the temporal progression of the detected thermal radiation. The processing unit (10) is configured to distinguish between the sudden heat caused by a person (12) entering the treatment room and the slow heat generation of the equipment in operation.

8. The medical system according to claim 1 or 2, in, The processing unit (10) is trained by machine learning to recognize the presence of a person (12) and distinguish them from inanimate components.

9. The medical system according to claim 1 or 2, in, The processing unit (10) is trained by machine learning to determine predetermined time points when the medical system is not typically used.

10. The medical system according to claim 1 or 2, in, The medical system includes at least one emergency stop function that is activated by an emergency stop switch (22) and / or a door contact at the entrance door (20) of the treatment room, wherein the processing unit (10) and / or the control unit (4) is configured to trigger the emergency stop function when it detects the presence of at least one person (12).

11. The medical system according to claim 1 or 2, in, The processing unit (10) is configured to immediately trigger the detection of the person (12) only before the occurrence of the at least one predetermined condition.

12. The medical system according to claim 1, wherein the medical system is a computed tomography (CT) scanning system.

13. The medical system of claim 1, wherein the radiation is X-ray radiation.

14. The medical system according to claim 1 or 2, wherein the sensor device (8) comprises 3 to 5 infrared sensors and / or infrared cameras.

15. The medical system according to claim 1 or 2, wherein the sensor device (8) is configured to capture images at a resolution ranging from 8×8 pixels to 32×40 pixels.

16. A monitoring device for use in a medical system according to any one of claims 1 to 15 to detect a person (12) in the environment and / or treatment room, comprising: At least one sensor device (8); and A processing unit (10) is configured to use the at least one sensor device (8) to detect whether there is a person (12) in the environment and / or in the treatment room, and to transmit information about the presence of the person (12) to a control unit (4) of the medical system.

17. The monitoring device according to claim 16, wherein the sensor device (8) is an infrared sensor and / or an infrared camera.

18. A method for inspecting a medical system in a treatment room according to any one of claims 1 to 15, comprising the following steps: a) Determine the possible presence of a person (12) in the treatment room by means of a monitoring device including at least one sensor device (8). b) The monitoring device detects the occurrence of at least one predetermined condition, and upon the occurrence of the predetermined condition and provided that no one (12) is present, transmits a start signal from the monitoring device to a control unit (4) of the medical system; and c) In response to the transmitted start signal, the control unit (4) initiates an inspection cycle of a radiation generating device (6) of the medical system and generates radiation.

19. A method for training a processing unit (10) of a monitoring device for a radiation-based medical system (1) according to any one of claims 1 to 15 using machine learning, comprising the following steps: a) Provide the monitoring device in the treatment room together with the medical system; b) Run the medical system for several consecutive days as usual, including having the user manually start the examination cycle at appropriate times; c) Based on the assumption that no one (12) is in the treatment room during the manually started inspection cycle, the monitoring device is adjusted to detect the person (12) by machine learning; d) Based on a manually selected start time and / or based on the operating time and associated pause time of the medical system, the appropriate start time of the examination cycle is automatically determined by machine learning.

20. The method according to claim 19, in, The monitoring device was pre-trained in a training environment before being placed in the treatment room.

Citation Information

Patent Citations

  • Industrial plant having an area relevant to safety

    CN101490283A

  • Radiation treatment of moving targets

    CN102512762A

  • Radiation tomography imaging apparatus and program

    JP2019010426A

  • Air conditioner detecting human body and control method of the same

    KR1020100092573A