A method and system for monitoring human motion

Abstract

The application provides a human motion monitoring method and system, and relates to the technical field of medical image equipment detection. The method dynamically adjusts the working state of an image acquisition device based on the real-time position of a medical image equipment probe, can make the image acquisition device accurately capture human motion signals in a differentiated device cooperative mode when the probe is at a key acquisition angle, and can quickly identify changes in the patient's body position beyond the allowed range by comparing the motion energy value with the preset threshold value and timely triggering an alarm. The entire monitoring process closely matches the actual work rhythm of the medical image equipment, effectively avoids problems such as blurred images and invalid acquisition data caused by unexpected patient motion, and ensures the quality stability of medical image acquisition.

Description

Technical Field

[0001] This application relates to the field of medical imaging equipment detection technology, and in particular to a method and system for monitoring human motion. Background Technology

[0002] In the field of medical imaging diagnostic equipment, single-photon emission computed tomography (SPECT) technology is widely used in the diagnosis of diseases such as tumors and cardiovascular diseases. The examination process requires the patient to remain still on the examination bed for 10 to 30 minutes. During this period, the SPECT dual probes rotate around the patient at a fixed step size (such as 6° / step) to collect data.

[0003] Current SPECT image acquisition has significant technical drawbacks: First, patients are prone to unexpected movements (such as head rotation or trunk shift) during prolonged periods of stillness, leading to motion artifacts in the images. In severe cases, re-examination is required, which not only prolongs the treatment time but also increases the patient's risk of radiation exposure. Second, motion monitoring relies on manual intervention, requiring doctors or technicians to observe continuously throughout the process. However, 30 minutes of monitoring can easily cause visual fatigue, making it difficult to capture subtle movements. Furthermore, the initial rotation phase of the dual probes (0°-45°, 135°-180°) can obstruct the left and right side views, creating blind spots and further reducing the reliability of monitoring.

[0004] Therefore, existing solutions cannot meet clinical needs. Designing an automatic motion monitoring solution that adapts to the rotation characteristics of SPECT dual probes, covers the entire acquisition cycle, and requires no manual intervention has become an urgent problem to be solved in this field. Summary of the Invention

[0005] To address the above problems, this application provides a method for monitoring human motion, including the following:

[0006] In a first aspect, this application provides a method for monitoring human motion, the method comprising:

[0007] To obtain the real-time position of the probe of a medical imaging device;

[0008] When the real-time position of the probe of the medical imaging device meets the preset angle, the working state of the image acquisition device is adjusted; the image acquisition device includes a first image acquisition device, a second image acquisition device and a third image acquisition device, the second image acquisition device and the third image acquisition device have the same working state, but different from the working state of the first image acquisition device;

[0009] Alternatively, the image acquisition device may include a fourth image acquisition device and a fifth image acquisition device, wherein the working state of the second image acquisition device is different from the working state of the first image acquisition device;

[0010] The motion energy value in the acquisition result of the image acquisition device is determined, and an alarm is issued when the motion energy value in the acquisition result is greater than a preset energy standard threshold.

[0011] Optionally, when the image acquisition device includes a first image acquisition device, a second image acquisition device, and a third image acquisition device, adjusting the working state of the image acquisition device when the real-time position of the medical imaging device probe meets the preset angle includes:

[0012] When the real-time position of the probe of the medical imaging device meets the first preset angle, the first image acquisition device is controlled to enter the working state, and the second and third image acquisition devices enter the non-working state; when the real-time position of the probe of the medical imaging device meets the second preset angle, the second and third image acquisition devices are controlled to enter the working state, and the first image acquisition device enters the non-working state.

[0013] Optionally, when the image acquisition device includes a fourth image acquisition device and a fifth image acquisition device, adjusting the working state of the image acquisition device when the real-time position of the medical imaging device probe meets the preset angle includes:

[0014] When the real-time position of the medical imaging device probe meets the first preset angle, the fourth image acquisition device is controlled to enter the working state, and the fifth image acquisition device enters the non-working state; when the real-time position of the medical imaging device probe meets the second preset angle, the fifth image acquisition device enters the working state, and the fourth image acquisition device enters the non-working state; the position of the fifth image acquisition device moves synchronously with the position of the medical imaging device probe.

[0015] Optionally, before acquiring the real-time position of the medical imaging device probe, the method further includes:

[0016] Before the medical imaging equipment starts working, acquire n frames of images acquired by the image acquisition device, where n is an integer greater than 1;

[0017] The acquired n frames of images are input into a preset human body segmentation model. The human body segmentation model performs semantic segmentation on the human body region in each frame of the image and outputs a human body mask that corresponds to each frame of the image.

[0018] A pixel-level intersection operation is performed on the human body masks corresponding to the n frames of images to retain the set of pixels that are all determined to be human body regions in the n frames of human body masks. Based on the set of pixels, a stable region of interest is generated that is required during the operation of the medical imaging device.

[0019] Optionally, the process for setting the energy standard threshold is as follows:

[0020] Before the medical imaging equipment starts working, each image acquisition device in the image acquisition device acquires a sets of static energy values, where a is an integer greater than 1, and the static energy values ​​are the energy values ​​of the target human body when it is kept absolutely still.

[0021] The energy standard threshold is defined as 1.5 times the maximum value among the static energy values ​​in group a.

[0022] Optionally, issuing an alarm when the motion energy value in the collected results exceeds a preset energy standard threshold includes:

[0023] An alarm is issued when the motion energy values ​​in two consecutive frames of the acquired data are both greater than the preset energy standard threshold.

[0024] Optionally, issuing an alarm when the motion energy value in the collected results exceeds a preset energy standard threshold includes:

[0025] An alarm is issued when the average value of motion energy in five consecutive frames of the acquired data exceeds the preset energy standard threshold.

[0026] Optionally, after generating the stable region of interest required for the operation of the medical imaging device based on the pixel set, the method further includes:

[0027] When the medical imaging equipment is in working condition, the image acquisition device reacquires m frames of images every preset time interval, where m is an integer greater than 1 and less than n;

[0028] The m frames of images acquired are input into the preset human body segmentation model. The human body segmentation model performs semantic segmentation on the human body region in each frame of the image and outputs a human body mask that corresponds to each frame of the image.

[0029] Perform pixel-level intersection operations on the human body masks corresponding to the m-frame images, retain the set of pixels that are all determined to be human body regions in the m-frame human body masks, and update the original region of interest based on the new pixel set.

[0030] Optionally, the first image acquisition device, the second image acquisition device, and the third image acquisition device respectively perform human motion monitoring within their respective regions of interest;

[0031] The fourth image acquisition device and the fifth image acquisition device respectively monitor human movement within their respective regions of interest.

[0032] Secondly, this application provides a human motion monitoring device, which includes:

[0033] The acquisition unit is used to acquire the real-time position of the probe of the medical imaging equipment;

[0034] The control unit is used to adjust the working state of the image acquisition device when the real-time position of the probe of the medical imaging device meets the preset angle; the image acquisition device includes a first image acquisition device, a second image acquisition device and a third image acquisition device, the second image acquisition device and the third image acquisition device have the same working state, but different from the working state of the first image acquisition device;

[0035] Alternatively, the image acquisition device may include a fourth image acquisition device and a fifth image acquisition device, wherein the working state of the second image acquisition device is different from the working state of the first image acquisition device;

[0036] The judgment unit is used to judge the motion energy value in the acquisition result of the image acquisition device, and to issue an alarm when the motion energy value in the acquisition result is greater than a preset energy standard threshold.

[0037] Optionally, when the image acquisition device includes a first image acquisition device, a second image acquisition device, and a third image acquisition device, the control unit is specifically used for:

[0038] When the real-time position of the probe of the medical imaging device meets the first preset angle, the first image acquisition device is controlled to enter the working state, and the second and third image acquisition devices enter the non-working state; when the real-time position of the probe of the medical imaging device meets the second preset angle, the second and third image acquisition devices are controlled to enter the working state, and the first image acquisition device enters the non-working state.

[0039] Optionally, when the image acquisition device includes a fourth image acquisition device and a fifth image acquisition device, the control unit is specifically used for:

[0040] When the real-time position of the medical imaging device probe meets the first preset angle, the fourth image acquisition device is controlled to enter the working state, and the fifth image acquisition device enters the non-working state; when the real-time position of the medical imaging device probe meets the second preset angle, the fifth image acquisition device enters the working state, and the fourth image acquisition device enters the non-working state; the position of the fifth image acquisition device moves synchronously with the position of the medical imaging device probe.

[0041] Optionally, the device further includes a region of interest generation unit, used to acquire n frames of images acquired by the image acquisition device before the medical imaging device starts working, where n is an integer greater than 1;

[0042] The acquired n frames of images are input into a preset human body segmentation model. The human body segmentation model performs semantic segmentation on the human body region in each frame of the image and outputs a human body mask that corresponds to each frame of the image.

[0043] A pixel-level intersection operation is performed on the human body masks corresponding to the n frames of images to retain the set of pixels that are all determined to be human body regions in the n frames of human body masks. Based on the set of pixels, a stable region of interest is generated that is required during the operation of the medical imaging device.

[0044] Optionally, the device further includes an energy standard threshold setting unit, used to collect a set of static energy values ​​through each image acquisition device in the image acquisition device before the medical imaging equipment starts working, where a is an integer greater than 1, and the static energy value is the energy value of the target human body when it is kept absolutely still;

[0045] The energy standard threshold is defined as 1.5 times the maximum value among the static energy values ​​in group a.

[0046] Optionally, the judgment unit is specifically used to issue an alarm when the motion energy values ​​in two consecutive frames of the acquisition result are both greater than the preset energy standard threshold.

[0047] Optionally, the judgment unit is specifically used to issue an alarm when the average value of motion energy values ​​in five consecutive frames of images in the acquisition results is greater than the preset energy standard threshold.

[0048] Optionally, after generating the stable region of interest required for the operation of the medical imaging device based on the pixel set, the region of interest generation unit is further configured to:

[0049] When the medical imaging equipment is in working condition, the image acquisition device reacquires m frames of images every preset time interval, where m is an integer greater than 1 and less than n;

[0050] The m frames of images acquired are input into the preset human body segmentation model. The human body segmentation model performs semantic segmentation on the human body region in each frame of the image and outputs a human body mask that corresponds to each frame of the image.

[0051] Perform pixel-level intersection operations on the human body masks corresponding to the m-frame images, retain the set of pixels that are all determined to be human body regions in the m-frame human body masks, and update the original region of interest based on the new pixel set.

[0052] Optionally, the first image acquisition device, the second image acquisition device, and the third image acquisition device respectively perform human motion monitoring within their respective regions of interest;

[0053] The fourth image acquisition device and the fifth image acquisition device respectively monitor human movement within their respective regions of interest.

[0054] Thirdly, this application provides an apparatus comprising a memory and a processor, the memory for storing instructions or code, and the processor for executing the instructions or code to cause the apparatus to perform the method described in any of the implementations of the first aspect.

[0055] Fourthly, this application provides a computer-readable storage medium storing code, wherein when the code is executed, a device executing the code implements the method described in any of the implementations of the first aspect.

[0056] This application provides a method for monitoring human motion, applied to the acquisition process of medical imaging equipment. When executing the method, the real-time position of the medical imaging equipment probe is first acquired. When the real-time position of the probe meets a preset angle, the working state of the image acquisition device is adjusted. The image acquisition device includes a first image acquisition device, a second image acquisition device, and a third image acquisition device. The second and third image acquisition devices operate in the same state, but different from the first image acquisition device. Alternatively, the image acquisition device includes a fourth and a fifth image acquisition device, where the working state of the second image acquisition device differs from that of the first image acquisition device. Then, the motion energy value in the acquisition result of the image acquisition device is determined. When the motion energy value in the acquisition result exceeds a preset energy standard threshold, an alarm is issued.

[0057] In this way, by dynamically adjusting the working state of the image acquisition device based on the real-time position of the medical imaging equipment probe, the image acquisition device can accurately capture human motion signals in a differentiated device collaboration mode when the probe is at a critical acquisition angle. This avoids the blind spots or insufficient accuracy problems of single-device acquisition and enables the rational allocation of acquisition resources through differences in the working states of different devices, reducing unnecessary computing power consumption. Furthermore, by comparing the motion energy value with a preset threshold, it can quickly identify postural changes of the patient that exceed the allowable range (such as limb swaying or postural deviation during medical image acquisition) and trigger alarms in a timely manner. This allows the entire monitoring process to closely follow the actual working rhythm of the medical imaging equipment, effectively avoiding problems such as image blurring and invalid acquisition data caused by unexpected patient movements, ensuring the quality and stability of medical image acquisition. At the same time, the differentiated image acquisition device working state design can also reduce the overall power consumption of the equipment while meeting monitoring accuracy requirements, improving system operating efficiency. This achieves a balance between medical image acquisition quality, monitoring timeliness, and equipment operating economy. This can provide more reliable motion monitoring support for medical imaging examinations, reduce the need for repeated examinations due to patient movement, shorten examination time, and improve the efficiency of medical services and the patient's medical experience. Attached Figure Description

[0058] To more clearly illustrate the technical solutions in this embodiment or the prior art, the drawings used in the description of the embodiment or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0059] Figure 1 A flowchart of a human motion monitoring method provided in an embodiment of this application;

[0060] Figure 2 This is a schematic diagram of the structure of a human motion monitoring device provided in an embodiment of this application. Detailed Implementation

[0061] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0062] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of related data must comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0063] The human motion monitoring method provided in this application is applied to the acquisition process of medical imaging equipment, specifically adaptable to Single-Photon Emission Computed Tomography (SPECT). SPECT technology falls under the category of nuclear medicine imaging technology. When using it to diagnose diseases such as tumors and cardiovascular diseases, the patient needs to remain still on the examination table for several tens of minutes. During this period, the dual probes of the SPECT equipment will rotate around the patient according to a preset trajectory, continuously acquiring internal functional metabolic data. If the patient moves during this process, motion artifacts are very likely to appear in the acquired images, and in severe cases, the acquired data may become unusable, affecting the accuracy of diagnosis.

[0064] The SPECT probe's motion parameters are as follows: it rotates around the patient in a fixed step size of 6° / step, pausing for 10 seconds after each step to complete data acquisition, with an overall rotation range covering 0°-180°. However, due to the inherent field-of-view obstruction problem of the SPECT device, specifically, during the initial rotation phase, the dual probes obstruct the left and right lateral views of the device in the 0°-45° and 135°-180° ranges, making it difficult for medical staff to observe whether the patient is moving laterally, further increasing the blind spots for manual monitoring.

[0065] To address the aforementioned issues, this application employs multiple image acquisition devices for monitoring. First, the current rotation angle of the dual SPECT probes is determined (by combining a step size of 6° / step, a dwell time of 10 seconds / step, and a rotation range of 0°-180° to accurately locate the probe's real-time position). Then, for each probe's real-time position, the corresponding image acquisition device is activated to acquire images of the target human body (patient). Subsequently, based on the acquired images, it can be determined whether the patient has moved. If movement occurs, a timely alert is issued, ensuring that the patient remains stationary on the examination table throughout the SPECT examination, thus avoiding motion artifacts affecting image quality. The following describes this application with specific embodiments:

[0066] Figure 1 This is a flowchart illustrating a human motion monitoring method provided in an embodiment of this application. (In conjunction with...) Figure 1 As shown, the human motion monitoring method provided in this application embodiment may include:

[0067] S101. Obtain the real-time position of the probe of the medical imaging equipment.

[0068] The medical imaging device probe moves in fixed steps, with a rotation range of 0°-180°. This rotation range is divided into segments: 0°-44°, 45°-135°, and 135°-180°. The 0°-44° and 135°-180° ranges are set as the first preset angle, and the 45°-135° range is set as the second preset angle. When acquiring the real-time position of the medical imaging device probe, it is further necessary to determine whether the real-time position of the medical imaging device probe is within the range of the first preset angle or the range of the second preset angle.

[0069] Since the human motion monitoring method in this application aims to determine whether the target human body, i.e., the patient to be examined, has moved and is in a non-static state, in order to make the detection results more accurate, the method further includes constructing a region of interest (ROI) before acquiring the real-time position of the medical imaging device probe. By pre-defining the ROI and using it as the monitoring area, a stable and accurate ROI can be constructed for different image acquisition devices at different working stages and viewing angles, reducing background interference and focusing on key parts of the patient. The specific process of generating the ROI is described below.

[0070] Before the medical imaging equipment starts working, n frames of images are acquired by the image acquisition device, where n is an integer greater than 1. The acquired n frames of images are input into a preset human body segmentation model. The human body segmentation model performs semantic segmentation processing on the human body regions in each frame of images and outputs a human body mask corresponding to each frame of images. Pixel-level intersection operations are performed on the human body masks corresponding to the n frames of images, and the set of pixels that are all determined to be human body regions in the n frames of human body masks is retained. Based on the pixel set, a stable region of interest required during the operation of the medical imaging equipment is generated.

[0071] The image acquisition device in this application includes multiple image acquisition devices to adapt to the viewing angle requirements of SPECT examination. Specifically, it includes two implementation methods. In one implementation method, the image acquisition device includes a first image acquisition device, a second image acquisition device, and a third image acquisition device. In another implementation method, the image acquisition device includes a fourth image acquisition device and a fifth image acquisition device.

[0072] When the image acquisition device includes a first image acquisition device, a second image acquisition device, and a third image acquisition device, the first image acquisition device is fixed directly above the diagnostic bed, with its lens perpendicular to the bed surface and a distance of 2-2.5 meters from the bed surface. Its viewing angle can completely cover the patient's entire body (from head to toe). The viewing angle is affected by the SPECT probe angle: when the probe is in the 45°-135° range, the image is partially obstructed by the probe, and only part of the patient's body can be observed; when the probe is in the 0°-45° or 135°-180° range, the image is unobstructed, and the entire patient's body can be clearly observed. The second image acquisition device is fixed outside the initial position of the SPECT left probe, installed at a height parallel to the bed surface, with its lens facing the patient's body. Its viewing angle specifically covers the patient's left half of the body (including the head, left torso, and left arm). The occlusion pattern of the third image acquisition device is the opposite of that of the first device: when the probe is in the 0°-45° or 135°-180° range, the image is partially obstructed by the probe, and only the left side of the patient's body can be observed; when the probe is in the 45°-135° range, the image is unobstructed, and the entire left side of the patient's body can be clearly observed. The third image acquisition device is fixed outside the initial position of the SPECT right-side probe, installed at a height parallel to the bed surface, with the lens facing the patient's body, and its viewing angle specifically covers the right half of the patient's body (including the head, right torso, and right arm). Its occlusion pattern is consistent with that of the second device: when the probe is in the 0°-45° or 135°-180° range, the image is partially obstructed; when it is in the 45°-135° range, the image is unobstructed, and the entire right side of the patient's body can be observed.

[0073] When the image acquisition device includes a fourth image acquisition device and a fifth image acquisition device, the fixed position, lens orientation, distance from the bed surface, and viewing angle coverage of the fourth image acquisition device are completely consistent with the first image acquisition device in the first implementation. That is, it is fixed directly above the diagnostic bed, with the lens pointing vertically downwards, 2-2.5 meters from the bed surface, and the viewing angle covers the entire patient. The image is partially obstructed in the 45°-135° range of the probe, and unobstructed in the 0°-45° and 135°-180° ranges. The fifth image acquisition device is fixed to the side of the diagnostic bed and establishes a linkage relationship with the SPECT probe through a mechanical slide rail. When the SPECT probe rotates around the patient, this device moves synchronously with the probe, thus replacing the symmetrical dual cameras in the first implementation. This setup eliminates the need to install devices on both sides; a single device can completely cover the patient's side view, saving examination room space while ensuring the monitoring range.

[0074] Therefore, when generating regions of interest, if there are three image acquisition devices, the regions of interest need to be generated separately for the first, second, and third image acquisition devices using the method described above. Similarly, if there are two image acquisition devices, the regions of interest need to be generated separately for the fourth and fifth image acquisition devices using the method described above.

[0075] It should be noted that the methods used to generate the corresponding region of interest are the same, whether it is based on the first, second, and third image acquisition devices or the fourth and fifth image acquisition devices. The specific process of generating the region of interest will be introduced below, taking the image acquisition devices located directly above the diagnostic bed and the image acquisition devices located to the side of the diagnostic bed as examples.

[0076] The ROI construction and monitoring of the image acquisition devices directly above (such as the first and fourth image acquisition devices) focuses on the preparation stage before SPECT acquisition and the unobstructed area of ​​the probe:

[0077] The SPECT scan is initiated one minute before the actual data acquisition begins, at which point the patient is already lying on the examination bed, and the SPECT probe has not yet started rotating. The control device continuously acquires 10 frames of patient images. Each of the 10 acquired frames is then input into a pre-defined human segmentation model. The model performs semantic segmentation on the human body region in each frame, outputting a "human body mask" (a pixel matrix that only marks the human body region) corresponding to each frame. A pixel-level "intersection operation" is performed on the 10 human body masks, retaining only the set of pixels that are all identified as human body regions in all 10 masks. A stable region of interest (ROI) is generated based on this set. Motion monitoring is immediately initiated after the ROI is constructed (at this time, the SPECT probe rotation angle is 0°). When the probe rotation angle is greater than 45° but less than 135°, monitoring is paused because the device's field of view is incomplete due to probe occlusion. When the probe rotation angle is greater than 135°, the field of view is restored to unobstructed, and monitoring is restarted.

[0078] The ROI construction and monitoring of side image acquisition devices (such as the second, third, and fifth image acquisition devices) are primarily adapted to the unobstructed side view range of the SPECT probe. No fixed timing is preset; ROI construction is triggered only when the SPECT probe rotation angle reaches ≥45° and <135° (within this range, the probe is unobstructed, and the device can fully capture the patient's side image). The construction steps are completely consistent with those of the image acquisition device directly above. Continuous motion monitoring is initiated immediately after ROI construction is completed, and monitoring is paused when the SPECT probe rotation angle exceeds 135° (entering the probe obstruction range, where the device's view is blocked).

[0079] To ensure the accuracy of motion detection, this application configures an independent monitoring algorithm module for each image acquisition device. The core functions are as follows: Each camera independently performs motion detection based on its own constructed ROI. It analyzes the displacement changes of pixels within the ROI using optical flow to quantify and calculate the intensity of the patient's motion, such as the pixel displacement corresponding to limb swaying or trunk shift. This design can completely avoid signal interference between different cameras. For example, the detection results of devices directly above and to the side do not affect each other. Only when a single device detects that the motion intensity exceeds a preset threshold will the corresponding alarm signal be triggered, effectively ensuring the accuracy of the alarm.

[0080] S102. When the real-time position of the probe of the medical imaging device meets the preset angle, adjust the working state of the image acquisition device.

[0081] As mentioned in step S101 above, the probe of the medical imaging device moves with a fixed step size and rotates within a range of 0°-180°. The rotation range is divided into segments of 0°-44°, 45°-135°, and 135°-180°. The ranges of 0°-44° and 135°-180° are set as the first preset angle, and the range of 45°-135° is set as the second preset angle.

[0082] In one implementation of this application, the image acquisition device employs a combination of a first image acquisition device, a second image acquisition device, and a third image acquisition device. The second and third image acquisition devices maintain the same operating state, while their operating states differ from those of the first image acquisition device. The specific adjustment logic is as follows: When the real-time position of the medical imaging device probe is at a first preset angle (0°-44° or 135°-180°), the first image acquisition device is controlled to enter the working state, while the second and third image acquisition devices enter the non-working state. When the real-time position of the probe is at a second preset angle (45°-135°), the second and third image acquisition devices are controlled to enter the working state, while the first image acquisition device enters the non-working state. Through this state adjustment, the first, second, and third image acquisition devices can independently perform human motion monitoring within their respective corresponding regions of interest, thereby achieving blind-spot-free coverage at different probe rotation angles.

[0083] In another implementation of this application, the image acquisition device adopts a combination of a fourth image acquisition device and a fifth image acquisition device, and the two have different working states, which can be dynamically adjusted according to the real-time position of the medical imaging device probe.

[0084] Specifically, based on the aforementioned preset angle divisions (the first preset angle is 0°-44° and 135°-180°, and the second preset angle is 45°-135°), when the probe's real-time position is at the first preset angle, the fourth image acquisition device is activated, while the fifth image acquisition device is deactivated. When the probe's real-time position is at the second preset angle, the fifth image acquisition device is activated, and the fourth image acquisition device is deactivated. In other words, the startup and shutdown logic of the fifth image acquisition device is: it is activated when the probe rotation angle is ≥45° and <135°, and it is deactivated when the probe rotation angle >135° (entering the obstruction zone). It should be noted that the position of the fifth image acquisition device will adjust synchronously with the movement of the medical imaging device probe to maintain effective lateral view coverage of the patient. Through this state switching, the fourth and fifth image acquisition devices can independently monitor human movement within their respective regions of interest, adapting to monitoring needs at different angles.

[0085] To accommodate the slight positional drift that patients may experience during the acquisition process of medical imaging equipment such as SPECT, such as slow sliding down when lying down or slight shoulder shift when turning to the side, and to avoid a decrease in monitoring accuracy of the fixed initial region of interest (ROI) due to changes in position, this application will update the ROI during the operation of the medical imaging equipment to ensure that the ROI is accurately matched with the patient's actual position throughout the entire acquisition cycle.

[0086] After generating the stable region of interest (ROI) required for the operation of the medical imaging equipment based on pixel sets, the top and side image acquisition devices each complete ROI reconstruction independently, depending on the different implementation methods of the image acquisition device. The specific process is as follows: When the image acquisition device includes a first image acquisition device for acquiring frontal images (top of the diagnostic bed), a second image acquisition device for acquiring side images (left side of the SPECT), and a third image acquisition device (right side of the SPECT), during the operation of the medical imaging equipment, the three devices independently perform ROI update operations at preset time intervals: First, the first, second, and third image acquisition devices each perform ROI update operations at preset time intervals (e.g., the preset time for the first image acquisition device is 1 second). (The preset time for the second and third image acquisition devices can be consistent with that of the first image acquisition device or set separately according to the side monitoring requirements.) Each device independently acquires m frames of images, where m is an integer greater than 1 and less than the number of frames n acquired during the initial ROI construction, in order to reduce the computational power consumption of the devices while ensuring update accuracy. Then, each device inputs its newly acquired m frames of images into a preset human segmentation model. The model performs semantic segmentation processing on the human regions in each frame of image and outputs a human mask corresponding to each frame of image. Subsequently, pixel-level intersection operations are performed on the human masks corresponding to the m frames of images of each device, retaining the set of pixels in the m frames of human masks that are all determined to be human regions, and updating the original region of interest corresponding to itself based on the new pixel set.

[0087] When the image acquisition device includes a fourth image acquisition device (directly above the examination bed) for acquiring frontal images and a fifth image acquisition device (linked with the SPECT probe) for acquiring side images, the update logic is consistent with the above three-device combination method: During the operation of the medical imaging equipment, the fourth and fifth image acquisition devices independently acquire m frames of images at preset time intervals. Each device inputs the acquired m frames of images into a preset human segmentation model to generate a corresponding human body mask. Then, it performs pixel-level intersection operations on its own m-frame human body masks, retaining the set of pixels that are all determined to be human body regions. Based on the new pixel set, it updates its corresponding original region of interest. Through the above update mechanism, the regions of interest from the top and sides can adapt to changes in patient position in real time, effectively improving monitoring stability.

[0088] S103. Determine the motion energy value in the acquisition result of the image acquisition device. If the motion energy value in the acquisition result is greater than the preset energy standard threshold, issue an alarm.

[0089] The calculation of motion energy value and alarm triggering logic are as follows: Each image acquisition device independently performs motion detection, calculates the pixel displacement between the current frame and the previous frame in the corresponding region of interest using optical flow, and quantifies the motion energy value. To avoid false alarms caused by single noise, this application sets dual alarm triggering conditions, which can be selected according to the actual scenario: The first case is that when the motion energy values ​​of two consecutive frames (within a 0.5-second interval, i.e., within 1 second) are both greater than the preset energy standard threshold, an independent alarm is triggered; The second case is that a "sliding window judgment" mechanism is used to calculate the average motion energy value of 5 consecutive frames (within 2.5 seconds). If the average value is greater than the preset energy standard threshold, an alarm is triggered to reduce false alarms caused by instantaneous noise such as sudden electrical sparks in the equipment, and further reduce the false alarm rate.

[0090] Meanwhile, the alarm focus of image acquisition devices at different locations is different. For example, the first and fourth image acquisition devices directly above focus on monitoring the patient's overall body movement, such as trunk shift and large head rotation, while the second, third, and fifth image acquisition devices on the side focus on monitoring the patient's side movements, such as raising the left arm and turning to the right.

[0091] The process of setting the preset energy standard threshold is as follows: Before the medical imaging equipment starts working, collect a group of static energy values ​​through each image acquisition device, where a is an integer greater than 1. The static energy value is the energy value when the target human body is kept absolutely still; multiply the maximum value in the group of static energy values ​​by 1.5, and the result is the energy standard threshold.

[0092] The implementation process of the human motion monitoring method in this application is described below with examples:

[0093] Preparation phase (1 minute): The image acquisition device directly above acquires 10 frames of images, generates ROIs covering the patient's entire body through the human body segmentation model (marking the head and torso as high-priority regions), and obtains a threshold of 12 (unit of energy value) based on static energy value calibration.

[0094] 0°-44° rotation phase (2 minutes): The image acquisition device directly above continuously monitors the patient's head. During this period, the patient's head turns slightly. The motion energy value of two consecutive frames is 18 (greater than the threshold of 12), which triggers the alarm directly above. After the doctor reminds the patient, the patient returns to stillness.

[0095] 45°-135° rotation phase (2 minutes): The monitoring from directly above stops, and the image acquisition devices on the left and right sides start monitoring and constructing the ROI (the left ROI covers the left arm and left torso). When the patient raises his left arm, the left device detects a motion energy value of 20 (greater than the left threshold of 15), triggering the left alarm. After the doctor intervenes, the patient is adjusted to a static state.

[0096] 135°-180° rotation phase (2 minutes): The top monitoring restarts, while the left and right side monitoring stops. During this period, the patient's head turns slightly again. The motion energy value of two consecutive frames is 18 (greater than the threshold of 12), triggering the top alarm. After the doctor reminds the patient, the patient returns to stillness.

[0097] Data collection ends (1 minute): All monitoring stops and monitoring logs are automatically saved.

[0098] The following section, using the SPECT full acquisition cycle scenario as an example, explains the application of the "sliding window judgment" mechanism in top-down and side-mounted monitoring:

[0099] During the preparation phase before starting the medical imaging equipment, threshold calibration is first performed: the top-view image acquisition device (ROI covering the whole body) collects 6 sets of static energy values, which are 7, 8, 9, 7, 10, and 8 when the patient is at rest. The maximum value of 10 is taken as 1.5 times, and the top-view threshold is set to 15. The side-view image acquisition device collects 5 sets of static energy values, which are 6, 5, 7, 6, and 8 when the patient is at rest on their side. The maximum value of 8 is taken as 1.5 times, and the side-view threshold is set to 12.

[0100] Full lifecycle monitoring process:

[0101] 0°-44° rotation phase (monitoring activated from directly above): The patient experiences slight trunk swaying due to discomfort while lying down. The motion energy values ​​of the device directly above for 5 consecutive frames (within 2.5 seconds) are 13, 14, 16, 14, and 13 respectively, with an average value of (13+14+16+14+13) / 5=14, which is less than the threshold of 15, so no alarm is triggered (minor adjustment movements are filtered out). Subsequently, the patient unconsciously turns over, and the energy values ​​for 5 consecutive frames are 18, 20, 19, 21, and 17, with an average value of 19>15, triggering the alarm from directly above. After being reminded by the doctor, the patient returns to stillness.

[0102] 45°-135° rotation phase (lateral monitoring activated): When the patient moves on the left side, the energy values ​​of the lateral device for 5 consecutive frames are 10, 11, 13, 11, 10, with an average of 11 < 12, and no alarm is triggered (physiological tremor is filtered); a moment later, when the patient moves on the right side, the energy values ​​for 5 consecutive frames are 14, 15, 13, 16, 14, with an average of 14.4 > 12, triggering the lateral alarm. After the doctor intervenes, the patient remains still.

[0103] 135°-180° rotation phase (overhead monitoring restarted): The patient's head moves, and the energy values ​​for 5 consecutive frames are 14, 15, 14, 16, and 15, with an average of 14.8 < 15, so no alarm is triggered; In the final phase, the patient moves again, and the energy values ​​for 5 consecutive frames are 17, 19, 18, 20, and 18, with an average of 18.4 > 15, triggering the overhead alarm. The doctor assists in adjusting the patient's position to complete the data collection.

[0104] Based on this, the "sliding window judgment" mechanism is used to calculate the average motion energy within 2.5 seconds for the whole body and half body areas of the device above and the device on the side, respectively. This accurately captures continuous motion (such as turning over or sliding down) and effectively filters out instantaneous physiological activities (such as muscle twitching or slight shaking), achieving motion monitoring with a low false alarm rate throughout the entire acquisition cycle.

[0105] To improve the adaptability and flexibility of the applicant's human motion monitoring method, the following alternative solutions can be adopted for some implementation methods in the above embodiments, as detailed below:

[0106] I. Image Acquisition Equipment Type Replacement Solution

[0107] The ordinary visible light camera (image acquisition device) used to acquire patient images in this application can be replaced with an infrared camera. The core advantage of this replacement is that it adapts to the dark environment acquisition scenarios of SPECT equipment (some SPECT equipment requires data acquisition in low-light or no-light environments to reduce ambient light interference). The infrared camera can generate clear images by capturing human body thermal radiation, and can accurately present the patient's outline even in dark environments, ensuring that the accuracy of human body region segmentation during subsequent ROI construction is not affected by lighting conditions, further expanding the applicable environment range of the method in this application.

[0108] II. Model Replacement Scheme in the ROI Construction Phase

[0109] During the ROI construction process, the Mask-RCNN human segmentation model used for semantic segmentation of human regions can be replaced with other models that have equivalent human image segmentation capabilities, such as the PP-HumanSeg-Lite model and the MobileNet-SSD model. All of these replacement models can achieve accurate segmentation of human regions in images and output corresponding human masks to support subsequent intersection operations to generate stable ROIs. Among them, the PP-HumanSeg-Lite model is lightweight, reducing device computational power consumption, while the MobileNet-SSD model has an advantage in segmentation speed. The appropriate model can be selected based on the actual hardware configuration and real-time requirements without affecting the final ROI construction result.

[0110] III. Algorithm Replacement Scheme for Calculating Motion Energy Value

[0111] The technique of calculating motion energy values ​​using optical flow in this application can be replaced by other algorithms capable of quantifying the motion intensity between image frames, including the Lucas-Kanade optical flow algorithm, the Farneback optical flow algorithm, frame difference method, color difference method, edge difference method, and structural similarity index (SSIM) analysis. All of these replacement algorithms can quantify the patient's motion intensity by analyzing pixel changes (such as position, color, and edge structure differences) between adjacent image frames. For example, the frame difference method quickly identifies the motion region by calculating the pixel differences between adjacent frames, while the structural similarity index reflects the degree of motion by comparing the structural differences between images between frames. The appropriate algorithm can be selected based on the accuracy requirements of motion detection and the device's computing power to ensure the validity of the motion energy value calculation results and support subsequent alarm logic triggering.

[0112] The above are some specific implementations of a human motion monitoring method provided in the embodiments of this application. Based on this, this application also provides a corresponding device. The device provided in the embodiments of this application will be described below from the perspective of functional modularity.

[0113] Figure 2 This is a schematic diagram of a human motion monitoring device provided in an embodiment of this application. (Combined with...) Figure 2 As shown, the human motion monitoring device 200 provided in this application embodiment includes:

[0114] The acquisition unit 210 is used to acquire the real-time position of the probe of the medical imaging equipment;

[0115] The control unit 220 is used to adjust the working state of the image acquisition device when the real-time position of the probe of the medical imaging device meets the preset angle; the image acquisition device includes a first image acquisition device, a second image acquisition device and a third image acquisition device, the second image acquisition device and the third image acquisition device have the same working state, but different from the working state of the first image acquisition device;

[0116] Alternatively, the image acquisition device may include a fourth image acquisition device and a fifth image acquisition device, wherein the working state of the second image acquisition device is different from the working state of the first image acquisition device;

[0117] The judgment unit 230 is used to judge the motion energy value in the acquisition result of the image acquisition device, and to issue an alarm when the motion energy value in the acquisition result is greater than a preset energy standard threshold.

[0118] In one implementation of this application, when the image acquisition device includes a first image acquisition device, a second image acquisition device, and a third image acquisition device, the control unit is specifically used for:

[0119] When the real-time position of the probe of the medical imaging device meets the first preset angle, the first image acquisition device is controlled to enter the working state, and the second and third image acquisition devices enter the non-working state; when the real-time position of the probe of the medical imaging device meets the second preset angle, the second and third image acquisition devices are controlled to enter the working state, and the first image acquisition device enters the non-working state.

[0120] In one implementation of this application embodiment, when the image acquisition device includes a fourth image acquisition device and a fifth image acquisition device, the control unit is specifically used for:

[0121] When the real-time position of the medical imaging device probe meets the first preset angle, the fourth image acquisition device is controlled to enter the working state, and the fifth image acquisition device enters the non-working state; when the real-time position of the medical imaging device probe meets the second preset angle, the fifth image acquisition device enters the working state, and the fourth image acquisition device enters the non-working state; the position of the fifth image acquisition device moves synchronously with the position of the medical imaging device probe.

[0122] In one implementation of this application, the apparatus further includes a region of interest generation unit, used to acquire n frames of images acquired by the image acquisition device before the medical imaging device starts working, where n is an integer greater than 1;

[0123] The acquired n frames of images are input into a preset human body segmentation model. The human body segmentation model performs semantic segmentation on the human body region in each frame of the image and outputs a human body mask that corresponds to each frame of the image.

[0124] A pixel-level intersection operation is performed on the human body masks corresponding to the n frames of images to retain the set of pixels that are all determined to be human body regions in the n frames of human body masks. Based on the set of pixels, a stable region of interest is generated that is required during the operation of the medical imaging device.

[0125] In one implementation of this application, the device further includes an energy standard threshold setting unit, used to collect a set of static energy values ​​by each image acquisition device in the image acquisition device before the medical imaging device starts working, wherein a is an integer greater than 1, and the static energy value is the energy value when the target human body is kept absolutely still;

[0126] The energy standard threshold is defined as 1.5 times the maximum value among the static energy values ​​in group a.

[0127] In one implementation of this application, the judgment unit is specifically used to issue an alarm when the motion energy values ​​in two consecutive frames of the acquisition result are both greater than the preset energy standard threshold.

[0128] In one implementation of this application, the judgment unit is specifically used to issue an alarm when the average value of motion energy values ​​in five consecutive frames of images in the acquisition result is greater than the preset energy standard threshold.

[0129] In one implementation of this application, after generating the stable region of interest required for the operation of the medical imaging device based on the pixel set, the region of interest generation unit is further configured to:

[0130] When the medical imaging equipment is in working condition, the image acquisition device reacquires m frames of images every preset time interval, where m is an integer greater than 1 and less than n;

[0131] The m frames of images acquired are input into the preset human body segmentation model. The human body segmentation model performs semantic segmentation on the human body region in each frame of the image and outputs a human body mask that corresponds to each frame of the image.

[0132] Perform pixel-level intersection operations on the human body masks corresponding to the m-frame images, retain the set of pixels that are all determined to be human body regions in the m-frame human body masks, and update the original region of interest based on the new pixel set.

[0133] In one implementation of this application embodiment, the first image acquisition device, the second image acquisition device, and the third image acquisition device respectively perform human motion monitoring within their respective regions of interest;

[0134] The fourth image acquisition device and the fifth image acquisition device respectively monitor human movement within their respective regions of interest.

[0135] This application also provides corresponding devices and computer storage media for implementing the solutions provided in this application.

[0136] The device includes a memory and a processor. The memory stores instructions or code, and the processor executes the instructions or code to cause the device to perform the method described in any embodiment of this application.

[0137] The computer storage medium stores code, and when the code is run, the device running the code implements the method described in any embodiment of this application.

[0138] As can be seen from the above description of the embodiments, those skilled in the art can clearly understand that all or part of the steps in the methods of the above embodiments can be implemented by means of software plus a general-purpose hardware platform. Based on this understanding, the technical solution of this application can be embodied in the form of a software product. This computer software product can be stored in a storage medium, such as a read-only memory (ROM) / RAM, magnetic disk, optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, a server, or a network communication device such as a router) to execute the methods described in various embodiments or some parts of the embodiments of this application.

[0139] It is understood that in the specific embodiments of this application, the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved need to obtain user permission or consent when the above embodiments of this application are applied to specific products or technologies, and the collection, use and processing of related data need to comply with the relevant laws, regulations and standards of relevant countries and regions.

[0140] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0141] It should also be noted that the various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, for the device and apparatus embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the description of the method embodiments. The device and apparatus embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components indicated as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of the solution in this embodiment according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0142] The above description is merely one specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for monitoring human motion, characterized in that, The method, applied to the acquisition process of medical imaging equipment, includes: To obtain the real-time position of the probe of a medical imaging device; When the real-time position of the probe of the medical imaging device meets the preset angle, the working state of the image acquisition device is adjusted; The image acquisition device includes a first image acquisition device, a second image acquisition device, and a third image acquisition device. The second and third image acquisition devices operate in the same state, but different from the first image acquisition device. When the image acquisition device includes the first, second, and third image acquisition devices, adjusting the operating state of the image acquisition device when the real-time position of the medical imaging device probe meets a preset angle includes: when the real-time position of the medical imaging device probe meets a first preset angle, controlling the first image acquisition device to enter the working state, and the second and third image acquisition devices to enter the non-working state; when the real-time position of the medical imaging device probe meets a second preset angle, controlling the second and third image acquisition devices to enter the working state, and the first image acquisition device to enter the non-working state; the first image acquisition device is fixed directly above the diagnostic bed; the second image acquisition device is fixed outside the initial position of the SPECT left probe; the third image acquisition device is fixed outside the initial position of the SPECT right probe. Alternatively, the image acquisition device may include a fourth image acquisition device and a fifth image acquisition device, wherein the operating state of the fourth image acquisition device differs from that of the fifth image acquisition device; when the image acquisition device includes a fourth image acquisition device and a fifth image acquisition device, adjusting the operating state of the image acquisition device when the real-time position of the medical imaging device probe meets a preset angle includes: When the real-time position of the medical imaging device probe meets the first preset angle, the fourth image acquisition device is controlled to enter the working state, and the fifth image acquisition device enters the non-working state; when the real-time position of the medical imaging device probe meets the second preset angle, the fifth image acquisition device enters the working state, and the fourth image acquisition device enters the non-working state; the position of the fifth image acquisition device moves synchronously with the position of the medical imaging device probe; the fourth image acquisition device is fixed directly above the diagnostic bed; The motion energy value in the acquisition result of the image acquisition device is determined, and an alarm is issued when the motion energy value in the acquisition result is greater than a preset energy standard threshold.

2. The method according to claim 1, characterized in that, Before acquiring the real-time position of the medical imaging device probe, the method further includes: Before the medical imaging equipment starts working, acquire n frames of images acquired by the image acquisition device, where n is an integer greater than 1; The acquired n frames of images are input into a preset human body segmentation model. The human body segmentation model performs semantic segmentation on the human body region in each frame of the image and outputs a human body mask that corresponds to each frame of the image. A pixel-level intersection operation is performed on the human body masks corresponding to the n frames of images to retain the set of pixels that are all determined to be human body regions in the n frames of human body masks. Based on the set of pixels, a stable region of interest is generated that is required during the operation of the medical imaging device.

3. The method according to claim 1, characterized in that, The process for setting the energy standard threshold is as follows: Before the medical imaging equipment starts working, each image acquisition device in the image acquisition device acquires a sets of static energy values, where a is an integer greater than 1, and the static energy values ​​are the energy values ​​of the target human body when it is kept absolutely still. The energy standard threshold is defined as 1.5 times the maximum value among the static energy values ​​in group a.

4. The method according to claim 1, characterized in that, The alarm is triggered when the motion energy value in the collected results exceeds a preset energy standard threshold, including: An alarm is issued when the motion energy values ​​in two consecutive frames of the acquired data are both greater than the preset energy standard threshold.

5. The method according to claim 1, characterized in that, The alarm is triggered when the motion energy value in the collected results exceeds a preset energy standard threshold, including: An alarm is issued when the average value of motion energy in five consecutive frames of the acquired data exceeds the preset energy standard threshold.

6. The method according to claim 2, characterized in that, After generating the stable region of interest required for the operation of the medical imaging device based on the pixel set, the method further includes: When the medical imaging equipment is in working condition, the image acquisition device reacquires m frames of images every preset time interval, where m is an integer greater than 1 and less than n; The m frames of images acquired are input into the preset human body segmentation model. The human body segmentation model performs semantic segmentation on the human body region in each frame of the image and outputs a human body mask that corresponds to each frame of the image. Perform pixel-level intersection operations on the human body masks corresponding to the m-frame images, retain the set of pixels that are all determined to be human body regions in the m-frame human body masks, and update the original region of interest based on the new pixel set.

7. The method according to any one of claims 1-6, characterized in that, The first image acquisition device, the second image acquisition device, and the third image acquisition device respectively perform human motion monitoring in their respective regions of interest. The fourth and fifth image acquisition devices respectively monitor human movement within their respective regions of interest.

8. A human motion monitoring system, characterized in that, The system includes: The acquisition unit is used to acquire the real-time position of the probe of the medical imaging equipment; The control unit is used to adjust the working state of the image acquisition device when the real-time position of the probe of the medical imaging equipment meets the preset angle. The image acquisition device includes a first image acquisition device, a second image acquisition device, and a third image acquisition device. The second and third image acquisition devices operate in the same state, but different from the first image acquisition device. When the image acquisition device includes the first, second, and third image acquisition devices, adjusting the operating state of the image acquisition device when the real-time position of the medical imaging device probe meets a preset angle includes: when the real-time position of the medical imaging device probe meets a first preset angle, controlling the first image acquisition device to enter the working state, and the second and third image acquisition devices to enter the non-working state; when the real-time position of the medical imaging device probe meets a second preset angle, controlling the second and third image acquisition devices to enter the working state, and the first image acquisition device to enter the non-working state; the first image acquisition device is fixed directly above the diagnostic bed; the second image acquisition device is fixed outside the initial position of the SPECT left probe; the third image acquisition device is fixed outside the initial position of the SPECT right probe. Alternatively, the image acquisition device may include a fourth image acquisition device and a fifth image acquisition device, wherein the operating state of the fourth image acquisition device differs from that of the fifth image acquisition device; when the image acquisition device includes a fourth image acquisition device and a fifth image acquisition device, adjusting the operating state of the image acquisition device when the real-time position of the medical imaging device probe meets a preset angle includes: When the real-time position of the medical imaging device probe meets the first preset angle, the fourth image acquisition device is controlled to enter the working state, and the fifth image acquisition device enters the non-working state; when the real-time position of the medical imaging device probe meets the second preset angle, the fifth image acquisition device enters the working state, and the fourth image acquisition device enters the non-working state; the position of the fifth image acquisition device moves synchronously with the position of the medical imaging device probe; the fourth image acquisition device is fixed directly above the diagnostic bed; The judgment unit is used to judge the motion energy value in the acquisition result of the image acquisition device, and to issue an alarm when the motion energy value in the acquisition result is greater than a preset energy standard threshold.

Images