Method of determining a placement area for a body surface marker and related products

Abstract

The embodiment of the present application provides a method for determining a body surface marker placement area and related products, which comprises: obtaining sample data of a plurality of patients, extracting two body surface contours of each patient according to two medical image data of each patient in the sample data, calculating a displacement change amount of each pixel point of the body surface of each patient under two different breathing states according to the two body surface contours of each patient and performing clustering processing to obtain L-class clustering results of the body surface movement of each patient; dividing the body surface of the patient into N marker placement areas according to the distribution of the L-class clustering results of the body surface movement of the plurality of patients; mapping the L-class clustering results of the body surface movement of each patient to the N marker placement areas; and obtaining M marker placement areas with larger displacement amounts according to the displacement amounts corresponding to the N marker placement areas. The accuracy of a respiratory motion correlation model established based on the body surface marker movement can be improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of respiration measurement, in particular to a method for determining a body surface marker placement area and related products. BACKGROUND

[0002] At present, in order to obtain the position of a lesion of an internal organ, a correlation model between a body surface marker and the internal organ can be established, and the position of the lesion of the internal organ can be predicted through the position of the body surface marker. How to place the body surface marker directly relates to important indicators such as the accuracy, robustness, compatibility and uniformity of the correlation model. Therefore, how to place the body surface marker on the body surface of a patient becomes a problem to be solved. SUMMARY

[0003] Embodiments of the present application provide a method for determining a body surface marker placement area and related products, which can improve the accuracy of a respiration motion correlation model established based on body surface marker motion by reasonably placing the body surface marker.

[0004] A first aspect of embodiments of the present application provides a method for determining a body surface marker placement area, the method being applied to an electronic device, and the method comprising:

[0005] obtaining sample data of a plurality of patients, the sample data comprising two medical image data at two different respiration phases;

[0006] extracting two body surface contours of each patient from the two medical image data of the patient, and calculating a displacement change amount of each pixel point of the body surface of the patient at the two different respiration states according to the two body surface contours of the patient;

[0007] performing clustering processing on the displacement change amount of each pixel point of the body surface of the patient at the two different respiration states to obtain an L-class clustering result of the body surface motion of the patient, L being a positive integer greater than or equal to 2;

[0008] dividing the body surface of the patient into N marker placement areas according to the distribution of the L-class clustering result of the body surface motion of the plurality of patients on the body surface contour, N being a positive integer greater than 2;

[0009] mapping the L-class clustering result of the body surface motion of each patient to the N marker placement areas, each marker placement area of each patient belonging to one clustering result or an empty clustering result; obtaining M marker placement areas with larger displacement amounts from the N marker placement areas corresponding to the displacement amounts obtained according to the L-class clustering result of the body surface motion of the plurality of patients, M being a positive integer less than or equal to N.

[0010] Wherein, the patient's posture and body position are unchanged when obtaining the two medical image data, and L can be selected as a positive integer between 3 and 6.

[0011] Optionally, after the M marker placement regions with larger displacement amounts are obtained, the method further comprises:

[0012] According to the physiological structure of the human body, the position of the physiological feature on the surface of the human body contained in each of the M marker placement regions is selected as the position description of each of the M marker placement regions on the surface of the human body.

[0013] Optionally, the displacement amount includes any one of an absolute displacement amount, an x-axis displacement amount, a y-axis displacement amount, and a z-axis displacement amount.

[0014] The corresponding displacement amounts of the N marker placement regions obtained according to the L-class clustering results of the body surface movements of the plurality of patients, to obtain M marker placement regions with larger displacement amounts, include:

[0015] According to the displacement amount corresponding to the clustering result to which each marker placement region of each patient belongs, the displacement amount corresponding to the clustering result to which each marker placement region of each patient belongs is sorted; the displacement amount sorting includes: the displacement amount corresponding to the clustering result to which each marker placement region of each patient belongs, in all L clustering results corresponding displacement amounts, sorted from small to large;

[0016] According to the displacement amount corresponding to the clustering result to which each marker placement region of each patient belongs, the displacement amount corresponding to the clustering result to which each marker placement region of each patient belongs is sorted; the displacement amount sorting includes: the displacement amount corresponding to the clustering result to which each marker placement region of each patient belongs, in all L clustering results corresponding displacement amounts, sorted from small to large;

[0017] The clustering result corresponding displacement amount sorting mean value of the N marker placement regions is sorted from large to small, to obtain the top M M marker placement regions.

[0018] Optionally, the displacement amount includes any one of an absolute displacement amount, an x-axis displacement amount, a y-axis displacement amount, and a z-axis displacement amount.

[0019] The corresponding displacement amounts of the N marker placement regions obtained according to the L-class clustering results of the body surface movements of the plurality of patients, to obtain M marker placement regions with larger displacement amounts, include:

[0020] According to the first displacement corresponding to the cluster result to which each marker placement region of each patient belongs, a displacement sorting of the first displacement corresponding to the cluster result to which each marker placement region of each patient belongs is obtained; the first displacement is any one of the absolute displacement, the x-axis displacement, the y-axis displacement and the z-axis displacement; the displacement sorting of the first displacement includes: the first displacement corresponding to the cluster result to which each marker placement region of each patient belongs, and the first displacement corresponding to all L cluster results is sorted from small to large;

[0021] According to the displacement sorting of the first displacement corresponding to the cluster result of each marker placement region of multiple patients, a first displacement sorting mean value corresponding to the cluster result of each marker placement region is obtained;

[0022] The first displacement sorting mean value corresponding to the cluster result of each marker placement region is sorted from large to small to obtain a sorting of the first displacement sorting mean value corresponding to the cluster result of each marker placement region;

[0023] According to the sorting of the absolute displacement sorting mean value, the x-axis displacement sorting mean value, the y-axis displacement sorting mean value and the z-axis displacement sorting mean value corresponding to the cluster result of each marker placement region, a weighted sum value of the sorting of the displacement sorting mean value corresponding to the cluster result of each marker placement region is obtained;

[0024] The weighted sum value of the sorting of the displacement sorting mean value corresponding to the cluster result of the N marker placement regions is sorted from small to large to obtain the top M marker placement regions.

[0025] Optionally, the first time body surface contour of each patient includes a first body surface pixel set, and the second time body surface contour of each patient includes a second body surface pixel set;

[0026] The displacement change of each pixel point of the body surface of each patient in the two different breathing states is calculated according to the two body surface contours of each patient, including:

[0027] According to the coordinates of the corresponding pixel points on the two body surface contours of each patient in the two different breathing states, the displacement change of the corresponding pixel points of each patient in the two different breathing states is calculated.

[0028] Optionally, the L-class cluster results of the body surface movement of each patient are obtained by clustering the displacement changes of each pixel point of the body surface of each patient in the two different breathing states, including:

[0029] The L-class clustering results of the body surface motion of each patient are mapped to the N marker placement regions, to obtain initial clustering results of the N marker placement regions of each patient.

[0030] Optionally, the step of mapping the L-class clustering results of the body surface motion of each patient to the N marker placement regions comprises:

[0031] The L-class clustering results of the body surface motion of each patient are mapped to the N marker placement regions, to obtain initial clustering results of the N marker placement regions of each patient.

[0032] If the initial clustering result of the first marker placement region completely belongs to one of the L-class clustering results, the one of the L-class clustering results is taken as the clustering result to which the first marker placement region belongs; the first marker placement region is any one of the N marker placement regions.

[0033] If the initial clustering result of the first marker placement region contains at least two of the L-class clustering results, one of the at least two of the L-class clustering results is determined as the clustering result to which the first marker placement region belongs, or the clustering result to which the first marker placement region belongs is attributed to an empty cluster.

[0034] Optionally, before the step of mapping the L-class clustering results of the body surface motion of each patient to the N marker placement regions, the method further comprises:

[0035] The body surface contour is divided according to human physiological structures, to obtain K physiological structure regions, K being a positive integer.

[0036] The step of dividing the body surface of a patient into N marker placement regions according to the distribution of the L-class clustering results of the body surface motion of multiple patients on the body surface contour comprises:

[0037] The body surface of a patient is divided into N marker placement regions according to the distribution of the L-class clustering results of the body surface motion of multiple patients on the body surface contour and the K physiological structure regions.

[0038] A second aspect of the embodiment of the application provides a device for determining a body surface marker placement region, the device being applied to an electronic device, and the device comprises:

[0039] An acquisition unit is configured to acquire sample data of multiple patients, the sample data comprising two pieces of medical image data at two different respiratory phases.

[0040] a calculation unit configured to extract, according to the two pieces of medical image data of each patient, two body surface contours of each patient, and calculate, according to the two body surface contours of each patient, a displacement change amount of each pixel point of the body surface of each patient in the two different respiratory states;

[0041] a clustering unit configured to perform clustering processing according to the displacement change amount of each pixel point of the body surface of each patient in the two different respiratory states, to obtain L-class clustering results of the body surface motion of each patient, L being a positive integer greater than or equal to 2;

[0042] a mapping unit configured to divide the body surface of a patient into N marker placement regions according to the L-class clustering results of the body surface motion of the plurality of patients in the distribution of the body surface contour, N being a positive integer greater than or equal to 2; and map the L-class clustering results of the body surface motion of each patient to the N marker placement regions, each marker placement region of each patient belonging to one class of clustering results or an empty clustering result;

[0043] a region determination unit configured to obtain, according to the displacement amount corresponding to the N marker placement region clustering results obtained from the L-class clustering results of the body surface motion of the plurality of patients, M marker placement regions with larger displacement amounts, M being a positive integer less than or equal to N.

[0044] Optionally, the device can further include a region position description unit configured to select, according to the physiological structure of the human body, a body surface physiological feature position contained in each of the M marker placement regions, and take the body surface physiological feature position contained in each of the M marker placement regions as a position description of the M marker placement regions on the body surface. In this case, the patient's body posture and body position are unchanged when the two pieces of medical image data are acquired, and L can be selected as a positive integer between 3 and 6.

[0045] A third aspect of an embodiment of the present application provides an electronic device including a processor and a memory, the memory being configured to store a computer program including program instructions, and the processor being configured to invoke the program instructions to execute the steps as described in the first aspect of the embodiment of the present application.

[0046] A fourth aspect of an embodiment of the present application provides a computer readable storage medium, wherein the computer readable storage medium stores a computer program including program instructions, and the program instructions, when executed by a processor, cause the processor to execute some or all of the steps as described in the first aspect of the embodiment of the present application.

[0047] The fifth aspect of the embodiments of the present application provides a computer program product, wherein the computer program product comprises a computer program, the computer program comprises program instructions, and the program instructions, when executed by a processor, cause the processor to perform part or all of the steps described in the first aspect of the embodiments of the present application. The computer program product can be a software installation package.

[0048] In the embodiments of the present application, sample data of multiple patients is acquired, the sample data comprising two medical image data at two different respiratory phases; two body surface contours of each patient are extracted according to the two medical image data of each patient, and a displacement change amount of each pixel point of the body surface of each patient at the two different respiratory states is calculated according to the two body surface contours of each patient; clustering processing is performed according to the displacement change amount of each pixel point of the body surface of each patient at the two different respiratory states, to obtain a clustering result of L classes of body surface motion of each patient; N marker placement regions of the body surface of the patient are divided according to the distribution of the L class clustering results of the body surface motion of the multiple patients on the body surface contour, N being a positive integer greater than 2, and the L class motion clustering result of the body surface of each patient is mapped to the N marker placement regions, each marker placement region of each patient belonging to one class of clustering result or an empty clustering result; M marker placement regions with a larger displacement amount are obtained according to the displacement amount corresponding to the N marker placement regions obtained from the L class clustering results of the body surface motion of the multiple patients, M being a positive integer less than or equal to N. In the embodiments of the present application, the L class clustering results of the body surface motion of the multiple patients are analyzed, and M marker placement regions with a larger displacement amount are analyzed from the N marker placement regions of the body surface of the multiple patients in the sample data at the two different respiratory phases, so that the M marker placement regions with a larger displacement amount in the process of respiratory motion of the body surface are screened out. By placing the body surface markers in the M marker placement regions, a respiratory motion correlation model with higher precision and stronger stability can be obtained in the process of establishing the respiratory motion correlation model based on the displacement amount of the body surface markers placed in the M marker placement regions, and thus the quality of the respiratory motion correlation model established based on the motion of the body surface markers is improved. BRIEF DESCRIPTION OF DRAWINGS

[0049] In order to more clearly illustrate the technical solutions of the embodiments of the present application or the prior art, the drawings needed in the embodiments or the prior art description will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0050] Figure 1 is a flowchart of a method for determining a body surface marker placement region provided by the embodiments of the present application;

[0051] Figure 2 is a flowchart of another method for determining a body surface marker placement region provided by an embodiment of the present application;

[0052] Figure 3 is a diagram illustrating a region of a lung body surface divided according to a human physiological structure provided by an embodiment of the present application;

[0053] Figure 4 is a diagram illustrating clustering results of motions of each point of a body surface of three samples in a three-dimensional space obtained by a clustering algorithm provided by an embodiment of the present application;

[0054] Figure 5 is a diagram illustrating mapping of clustering results of motions of body surfaces of three samples to body surfaces provided by an embodiment of the present application;

[0055] Figure 6 is a diagram illustrating setting of candidate placement regions of body surface markers and coding of each region provided by an embodiment of the present application;

[0056] Figure 7 is a diagram illustrating obtaining of displacement ranking values of four displacement indicators of each clustering region of sample 5 provided by an embodiment of the present application;

[0057] Figure 8 is a diagram illustrating calculating of mean values of displacement rankings of clustering regions and ranking according to the mean values of displacement rankings provided by an embodiment of the present application;

[0058] Figure 9 is a diagram illustrating calculating of a priority order of candidate placement regions of lung body surface markers according to respiratory motion displacements provided by an embodiment of the present application;

[0059] Figure 10 is a diagram illustrating a priority selection order of candidate placement regions of lung body surface markers provided by an embodiment of the present application;

[0060] Figure 11 is a diagram illustrating a marker placement scheme and a description of each marker placement position provided by an embodiment of the present application;

[0061] Figure 12 is a structural diagram of a device for determining a body surface marker placement region provided by an embodiment of the present application;

[0062] Figure 13 is a structural diagram of an electronic device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0063] With reference to the drawings, the technical solutions in the embodiments of the present application will be clearly and completely described below. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts are within the scope of the present application.

[0064] The terms "first", "second", and the like in the description and claims of the present application and the above drawings are used to distinguish different objects, rather than to describe a particular order. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units is not limited to the listed steps or units, but can optionally include steps or units not listed, or can optionally include other steps or units inherent to the process, method, product, or device.

[0065] In the present application, the phrase "embodiment" means that the specific features, structures, or characteristics described in connection with the embodiment can be included in at least one embodiment of the present application. The phrase appears at various places in the specification does not necessarily refer to the same embodiment, nor is it mutually exclusive or alternative to other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described in the present application can be combined with other embodiments.

[0066] At present, a large category of respiratory motion models is established by correlation between a surrogate and a treated organ (such as a thoraco-abdominal treated organ) in the body. The surrogate should have two characteristics: one is that its change characteristics are clearly correlated with the respiratory motion change of the treated organ in real time, and the other is that the change signal of the surrogate is easy to track. A more intuitive surrogate is the human body surface that changes with the respiratory motion. The respiratory motion change of the body surface can be monitored by using a body surface marker tracking instrument, a pressure detection sensor, etc., and the former is more simple and universal. One of the greatest difficulties in establishing a respiratory motion model by using a body surface marker as a surrogate is how to place the marker on the patient's body surface. The placement strategy of the body surface marker is directly related to important indicators such as the accuracy, robustness, compatibility, and uniformity of the respiratory motion model, and no special research or recognized results have been found at present.

[0067] The embodiment of the present application provides a method for determining a body surface marker placement area and related products, a clustering result of a body surface respiratory motion change amount is analyzed, N marker placement areas with larger displacement amounts in two different respiratory phases are analyzed from sample data of a plurality of patients, thereby screening M marker placement areas with larger displacement amounts in the process of respiratory motion, and placing the body surface markers in the M marker placement areas, in the process of establishing a respiratory motion correlation model based on the displacement amounts of the body surface markers placed in the M marker placement areas, a respiratory motion correlation model with higher precision and stronger stability can be obtained, and the quality of the respiratory motion correlation model established based on the body surface marker motion is improved.

[0068] Please refer to Figure 1 , Figure 1 is a flowchart of a method for determining a body surface marker placement area provided by the embodiment of the present application. As shown in Figure 1 , the method can include the following steps.

[0069] 101, the electronic device obtains sample data of a plurality of patients, and the sample data includes two medical image data in two different respiratory phases.

[0070] The respiratory motion correlation model of the embodiment of the present application can be established based on a statistical method of probability distribution, or can be any one of a linear regression (LR) model, a neural network (NN) model, and a support vector machine (SVR) model.

[0071] In order to obtain a body surface marker placement scheme, sample data of a plurality of patients is required. The number of the plurality of patients can be Q, and generally, the more the number of Q, the more accurate the established body surface marker placement scheme, and the stronger the universality. For example, taking the number of patients Q=30 as an example, for the sample data of 30 patients, the sample data of each patient can include two medical image data in two different respiratory phases.

[0072] The medical image data can include any one of computed tomography (CT) data, magnetic resonance imaging (MRI) data, and X-ray data.

[0073] For example, the two medical image data can include CT data in two different respiratory phases.

[0074] Respiratory phase, which can also be referred to as respiratory state, refers to a state of a person in the process of breathing. Taking CT data as an example of medical image data, when a patient is taking a CT, the patient needs to be in a breath-holding state, which is to maintain a respiratory state. For example, the patient can take a CT image in a first respiratory state (such as a state of taking a deep breath) and take another CT image in a second respiratory state (such as a state of taking a deep breath), thereby obtaining two CT data of the patient. For all patients, two CT data can be obtained in the above manner.

[0075] In an embodiment, the electronic device can establish a data transmission channel (such as through serial communication or a network cable) with a medical image device (such as a CT device or an MRI device or an X-ray imaging device), and the electronic device can directly obtain, from the medical image device, two medical image data of a plurality of patients in two different respiratory phases.

[0076] In an embodiment, after the medical image device obtains the medical image data, the medical image device can upload the medical image data to an image data management system of a hospital, such as a picture archiving and communication system (PACS). The electronic device can obtain, from the PACS, two medical image data of a plurality of patients in two different respiratory phases.

[0077] In an embodiment, the two medical image data of a plurality of patients in two different respiratory phases can be stored in a specific external storage, and the electronic device can establish a data connection channel with the external storage to obtain, from the external storage, the two medical image data of a plurality of patients in two different respiratory phases.

[0078] Data collection: The sample data of each (patient) can include two medical image data in two different respiratory phases. The medical image data covers the body surface of the patient's torso and the complete treated organ, and the scanning layer thickness can be according to the highest treatment accuracy requirement of the model (such as the scanning layer thickness is not more than 1.5 mm). For example, when taking a CT image, the patient is in a body position in treatment. The body surface marker placement scheme applied in clinical application, in order to ensure the accuracy and universality of the respiratory motion correlation model, the number of samples can be greater than or equal to 30 cases.

[0079] It should be noted that the body state and body position of each patient are unchanged when the two medical image data are obtained.

[0080] 102, the electronic device extracts two body surface contours of each patient according to the two medical image data of each patient, and calculates the displacement change amount of each pixel point of the body surface of each patient in two different respiratory states according to the two body surface contours of each patient.

[0081] In the embodiments of the present application, the patient body surface contour can be extracted from the medical image data, and the displacement change amount of each pixel point of the patient body surface in two different breathing states of the same patient can be calculated. The registration algorithm can be used to register each pixel point of the patient body surface in the two medical image data, so as to calculate the displacement of the same pixel point of the patient body surface in the two medical image data. The registration algorithm is a non-rigid transformation registration algorithm. The displacement index (displacement change amount) used in the embodiments of the present application can be calculated based on different coordinate systems, such as Cartesian coordinate system, polar coordinate system, etc.

[0082] For example, taking CT images as an example, the patient body surface contour can be extracted from the CT images. The patient body surface contour in the CT images can be automatically recognized and extracted by software, or the patient body surface contour can be extracted from the CT images by manual labeling.

[0083] The displacement change amount can include at least one of the absolute displacement change amount of the Cartesian coordinate system, the x-axis displacement change amount, the y-axis displacement change amount, and the z-axis displacement change amount.

[0084] Optionally, the first body surface contour of each patient includes a first body surface pixel set, and the second body surface contour of each patient includes a second body surface pixel set.

[0085] The electronic device calculates the displacement change amount of each pixel point of the patient body surface in the two different breathing states of each patient according to the two body surface contours of each patient, including:

[0086] The electronic device calculates the displacement change amount of the corresponding pixel point of each patient in the two different breathing states according to the coordinates of the corresponding pixel points on the two body surface contours of each patient in the two different breathing states.

[0087] In the embodiments of the present application, the displacement change amount of the corresponding pixel point of each patient in the two different breathing states can include the Euclidean distance (absolute displacement change amount) of the spatial displacement of each pixel point in the Cartesian coordinate system (x, y, z), and the x-axis displacement change amount (x2-x1), the y-axis displacement change amount (y2-y1), and the z-axis displacement change amount (z2-z1) of the corresponding pixel point of each patient in the two different breathing states can be calculated according to the two groups of coordinates (x1, y1, z1), (x2, y2, z2) of the corresponding pixel points on the two body surface contours of each patient in the two different breathing states.

[0088] The two medical image data of each patient cover the patient trunk body surface and the complete treated organ. The corresponding pixels of the two medical images of each patient can be found by registration software to calculate the displacement field.

[0089] 103, the electronic device performs clustering processing on the displacement change of each pixel point on the body surface of each patient in two different breathing states, and obtains an L-class clustering result of the body surface movement of each patient.

[0090] L is a positive integer greater than or equal to 2. For example, L can be selected as a positive integer between 3 and 6.

[0091] In the embodiments of the present application, the clustering algorithm can be used to perform clustering processing on the displacement change of each pixel point on the body surface, and obtain an L-class clustering result of the body surface movement of each patient. The displacement change can include three-dimensional space displacement change in each direction. The L-class clustering result of the body surface movement of each patient can include L classes. The respiratory movement change characteristics of the body surface pixel points in the same class are similar. For example, according to the accuracy requirement and operability, the body surface of each patient is clustered into 3-6 classes.

[0092] It should be noted that each patient is clustered according to the same clustering algorithm, and the number of classes of the body surface movement clustering result of each patient is the same. The clustering algorithm can include any one of a fuzzy clustering algorithm, a partition-based clustering algorithm, a hierarchical clustering algorithm, a density-based clustering algorithm, a network-based clustering algorithm, and a model-based clustering algorithm. For example, the body surface movement of each patient is clustered into 5 classes using a fuzzy clustering algorithm (Fuzzy C means).

[0093] Optionally, step 103 can include the following steps:

[0094] The electronic device performs fuzzy clustering processing on the displacement change of each pixel point on the body surface of each patient in the two different breathing states, and obtains an L-class clustering result of the body surface movement of each patient.

[0095] In the embodiments of the present application, the fuzzy clustering processing refers to processing using a fuzzy clustering method. Since the displacement change of each pixel point on the body surface of each patient in two different breathing states is close to a normal distribution, using a fuzzy clustering algorithm for clustering processing can improve the accuracy of the body surface movement clustering result of each patient.

[0096] 104, the electronic device divides the body surface of the patient into N marker placement regions according to the distribution of the body surface contour of the L-class clustering result of the body surface movement of the plurality of patients, maps the L-class clustering result of the body surface movement of each patient to the N marker placement regions, and each marker placement region of each patient belongs to a class clustering result or an empty clustering result.

[0097] N is a positive integer greater than 2.

[0098] In the embodiments of the present application, the body surface of the patient is divided into N marker placement regions according to the distribution of the L-class clustering results of the body surface movement of a plurality of patients on the body surface contour, wherein the body surface of each patient is divided into N marker placement regions in the same position, so that each marker placement region of each patient belongs to one clustering result as much as possible and belongs to the empty clustering result as little as possible.

[0099] The L-class clustering results of the body surface movement of each patient can be mapped to the N marker placement regions. For one patient, one marker placement region thereof belongs to one clustering result or the empty clustering result. One marker placement region of a certain patient belongs to the empty clustering result, which indicates that the marker placement region of the patient does not belong to a single cluster, and therefore the marker placement region of the patient does not participate in the average value calculation.

[0100] Optionally, in step 104, the electronic device mapping the L-class clustering results of the body surface movement of each patient to the N marker placement regions can include the following steps:

[0101] (11) The electronic device maps the L-class clustering results of the body surface movement of each patient to the N marker placement regions to obtain the initial clustering results of the N marker placement regions of each patient.

[0102] (12) If the initial clustering result of the first marker placement region completely belongs to one clustering result in the L classes, the electronic device takes the one clustering result as the clustering result to which the first marker placement region belongs; the first marker placement region is any one of the N marker placement regions.

[0103] (13) If the initial clustering result of the first marker placement region contains at least two clustering results in the L classes, the electronic device determines one clustering result from the at least two clustering results as the clustering result to which the first marker placement region belongs, or attributes the clustering result of the first marker placement region to the empty clustering result.

[0104] In the embodiments of the present application, after the L-class clustering results of the body surface movement of each patient are mapped to the N marker placement regions, the initial clustering result of each marker placement region can include only one clustering result or at least two clustering results. If it includes only one clustering result, the one clustering result is directly taken as the clustering result of the marker placement region; if it includes at least two clustering results, one clustering result is determined from the at least two clustering results as the clustering result of the marker placement region, or the clustering result of the marker placement region is attributed to the empty clustering result.

[0105] The determining the one type of clustering result from the at least two types of clustering results as the clustering result to which the first marker placement region belongs, or attributing the clustering result of the first marker placement region to the empty clustering result, can include the following steps:

[0106] determining the number of pixel points in the first marker placement region that are attributed to each of the at least two types of clustering results, and selecting the one type of clustering result with the largest number of pixel points from the at least two types of clustering results;

[0107] If the proportion of the number of pixel points in the one type of clustering result with the largest number of pixel points to all pixel points in the first marker placement region is greater than a first threshold, the one type of clustering result with the largest number of pixel points is taken as the clustering result of the first marker placement region.

[0108] If the proportion of the number of pixel points in the one type of clustering result with the largest number of pixel points to all pixel points in the first marker placement region is less than the first threshold, the clustering result of the first marker placement region is attributed to the empty clustering result.

[0109] In the embodiments of the present application, the first threshold can be set in advance, and the first threshold can be set to 50% or more, for example, the first threshold can be set to 60%.

[0110] The embodiments of the present application provide a method for determining the clustering result of a first marker placement region. In the case that the initial clustering result of the first marker placement region contains at least two types of clustering results, whether to attribute the clustering result of the first marker placement region to the empty clustering result can be determined according to the proportion of the number of pixel points in the one type of clustering result with the largest number of pixel points. The accuracy of the clustering result to which the first marker placement region belongs can be improved when the initial clustering result of the first marker placement region is relatively concentrated (the proportion of the number of pixel points in the one type of clustering result with the largest number of pixel points to all pixel points in the first region is greater than the first threshold), and the first marker placement region can be avoided from being attributed to an inaccurate clustering result when the initial clustering result of the first marker placement region is relatively dispersed (the proportion of the number of pixel points in the one type of clustering result with the largest number of pixel points to all pixel points in the first region is less than the first threshold).

[0111] 105. The electronic device obtains the displacement amount corresponding to the N marker placement regions according to the L type of clustering results of the body surface movement of the plurality of patients, and obtains M marker placement regions with larger displacement amount, M being a positive integer less than or equal to N.

[0112] In the embodiments of the present application, the displacement amount corresponding to each marker placement area can be obtained by performing mathematical operation on the displacement amount corresponding to the cluster center of the marker placement area of the plurality of patients. The mathematical operation can include any one or several of the following: average value operation, weighted average value operation, average value operation of sorted displacement amounts, and weighted average value operation of sorted displacement amounts.

[0113] M can be selected according to clinical requirements. For example, assuming that the clinical requirement is that the number of patient body surface markers should not exceed 10 when treatment is required, M can be 10.

[0114] Optionally, after step 105 is performed, the following steps can also be performed:

[0115] According to the physiological structure of the human body, the positions of the physiological features on the body surface included in each of the M marker placement areas are selected, and the positions of the physiological features on the body surface included in each of the M marker placement areas are taken as the position description of the M marker placement areas on the body surface.

[0116] In the embodiments of the present application, after the M marker placement areas are selected, in order to facilitate the staff to paste the M markers in the M marker placement areas, each marker placement area needs to have a specific position description. This description can be the most easily implemented scheme, and can be named according to the position of each placement area according to the features of the body surface. For example, as shown in FIG. 6, the positions of the M marker placement areas are described as follows: the left side of the neck, the right side of the neck, the left side of the chest, the right side of the chest, the left side of the abdomen, the right side of the abdomen, the left side of the hip, the right side of the hip, the left side of the knee, and the right side of the knee. Figure 11 As shown in FIG. 6, since the K physiological structure regions of the body surface contour are referred to for division when the N candidate marker placement areas are determined, the position description of the M marker placement areas is also associated with the division of the K physiological structure regions. According to the physiological structure feature description of the M marker placement areas on the body surface, the uniformity of marker placement during sample collection and model implementation can be ensured.

[0117] Optionally, the displacement amount includes any one of an absolute displacement amount, an x-axis displacement amount, a y-axis displacement amount, and a z-axis displacement amount. In step 105, the electronic device obtains the displacement amount corresponding to the N marker placement areas according to the L cluster results of the body surface movement of the plurality of patients, and obtains the M marker placement areas with larger displacement amounts, which can include the following steps:

[0118] (21) The electronic device obtains the displacement amount corresponding to the cluster result to which each marker placement area of each patient belongs, according to the cluster center corresponding to the cluster result of each marker placement area of each patient. The displacement amount corresponding to the cluster result to which each marker placement area of each patient belongs is sorted in all L cluster results according to the order from small to large.

[0119] (22) The electronic device sorts the displacement amounts corresponding to the clustering results of each marker placement area of the plurality of patients according to the displacement amounts, and calculates the average of the displacement amounts corresponding to the clustering results of each marker placement area;

[0120] (23) The electronic device sorts the average of the displacement amounts corresponding to the clustering results of the N marker placement areas from large to small, and obtains the top M marker placement areas.

[0121] Each clustering result has a cluster center, and each cluster can be a three-dimensional space range composed of pixel points in a three-dimensional space (see Figure 4 ). The cluster center can be a central pixel point that can represent the clustering result among all pixel points belonging to the clustering result. The displacement amount corresponding to the cluster center of the clustering result can be the displacement amount of the central pixel point of the clustering result.

[0122] Optionally, after step (23) is performed, step (24) can also be performed, wherein:

[0123] (24) According to the human physiological structure, the body surface physiological feature positions contained in each of the M marker placement areas are selected, and the body surface physiological feature positions contained in each of the M marker placement areas are taken as the position description of the M marker placement areas on the body surface.

[0124] In the embodiments of the present application, N = 19 (the region codes of the 19 marker placement areas are B1-B6, G1-G5, Y1-Y4, and L1-L4, respectively), as shown in Figure 6 , L = 5, the plurality of patients are Q patients, and Q = 9 is taken as an example, and Figure 4 and Figure 5 Sample 4 (patient A), sample 5 (patient B), and sample 7 (patient C) are taken as examples. The L-class clustering results include 5-class clustering results. Each clustering result corresponds to an absolute displacement amount, an x-axis displacement amount, a y-axis displacement amount, and a z-axis displacement amount according to the cluster center position. For example (see Figure 5 , Figure 6The absolute displacement amount corresponding to the clustering result to which the marker placement region B4, G1, G2, G3, G5, Y1 of patient A belongs is 1.83, the x-axis displacement amount is 0.03, the y-axis displacement amount is 1.82, and the z-axis displacement amount is -0.19. The absolute displacement amount corresponding to the clustering result to which the marker placement region B1, B3 of patient A belongs is 0.48, the x-axis displacement amount is -0.42, the y-axis displacement amount is 0.21, and the z-axis displacement amount is -0.12. The absolute displacement amount corresponding to the clustering result to which the marker placement region L2, L3, L4 of patient A belongs is 4.51, the x-axis displacement amount is 0.37, the y-axis displacement amount is 4.23, and the z-axis displacement amount is -1.53. The absolute displacement amount corresponding to the clustering result to which the marker placement region B2, B5, Y2 of patient A belongs is 2.53, the x-axis displacement amount is -0.46, the y-axis displacement amount is 1.79, and the z-axis displacement amount is -1.73. The absolute displacement amount corresponding to the clustering result to which the marker placement region B6, G4, Y3 of patient A belongs is 2.78, the x-axis displacement amount is 0.27, the y-axis displacement amount is 2.38, and the z-axis displacement amount is -1.42. The marker placement region L1, Y4 of patient A belongs to the empty clustering result, and the corresponding empty value is obtained.

[0125] The absolute displacement amount corresponding to the clustering result to which the marker placement region B2, B5, G1 of patient B belongs is 0.93, the x-axis displacement amount is 0.35, the y-axis displacement amount is -0.64, and the z-axis displacement amount is -0.58. The absolute displacement amount corresponding to the clustering result to which the marker placement region B4, G2, Y1, Y2, L2 of patient B belongs is 0.66, the x-axis displacement amount is -0.23, the y-axis displacement amount is -0.31, and the z-axis displacement amount is 0.54. The absolute displacement amount corresponding to the clustering result to which the marker placement region B1, G5, Y4, L3, L4 of patient B belongs is 0.85, the x-axis displacement amount is -0.25, the y-axis displacement amount is 0.79, and the z-axis displacement amount is -0.21. The absolute displacement amount corresponding to the clustering result to which the marker placement region B3, B6, G4, Y3 of patient B belongs is 1.27, the x-axis displacement amount is 1.19, the y-axis displacement amount is 0.11, and the z-axis displacement amount is -0.43. The absolute displacement amount corresponding to the clustering result to which the marker placement region G3, L1 of patient B belongs is 1.11, the x-axis displacement amount is 0.83, the y-axis displacement amount is -0.13, and the z-axis displacement amount is 0.72.

[0126] The absolute displacement amount corresponding to the clustering result to which the marker placement region B1, B3, B4, G1, Y2, Y3 of patient C belongs is 0.27, the x-axis displacement amount is 0.15, the y-axis displacement amount is -0.22, and the z-axis displacement amount is 0.02. The absolute displacement amount corresponding to the clustering result to which the marker placement region B6, G4 of patient C belongs is 1.24, the x-axis displacement amount is 0.14, the y-axis displacement amount is 0.95, and the z-axis displacement amount is -0.79. The absolute displacement amount corresponding to the clustering result to which the marker placement region B2, G2, G3 of patient C belongs is 1.98, the x-axis displacement amount is -0.32, the y-axis displacement amount is 0.22, and the z-axis displacement amount is -1.94. The absolute displacement amount corresponding to the clustering result to which the marker placement region Y1, Y4, L1 of patient C belongs is 2.44, the x-axis displacement amount is 0.02, the y-axis displacement amount is -2.30, and the z-axis displacement amount is -0.80. The absolute displacement amount corresponding to the clustering result to which the marker placement region L2, L3, L4 of patient C belongs is 5.09, the x-axis displacement amount is -0.37, the y-axis displacement amount is -3.97, and the z-axis displacement amount is -3.17. The marker placement region B5, G5 of patient C belongs to the empty clustering result, and the corresponding empty value.

[0127] In one embodiment, the displacement amount is taken as the absolute displacement amount of the cluster center position. From the data analysis of the above example, it can be known (see Figure 5 and Figure 6 shown):

[0128] The displacement amount corresponding to the clustering result to which the marker placement region B4, G1, G2, G3, G5, Y1 of patient A belongs is ranked 2, the displacement amount corresponding to the clustering result to which the marker placement region B1, B3 of patient A belongs is ranked 1, the displacement amount corresponding to the clustering result to which the marker placement region L2, L3, L4 of patient A belongs is ranked 5, the displacement amount corresponding to the clustering result to which the marker placement region B2, B5, Y2 of patient A belongs is ranked 3, the displacement amount corresponding to the clustering result to which the marker placement region B6, G4, Y3 of patient A belongs is ranked 4, and the displacement amount corresponding to the clustering result to which the marker placement region L1, Y4 of patient A belongs is ranked as an empty value.

[0129] The displacement amount corresponding to the clustering result to which the marker placement region B2, B5, G1 of patient B belongs is ranked 3. The displacement amount corresponding to the clustering result to which the marker placement region B4, G2, Y1, Y2, L2 of patient B belongs is ranked 1. The displacement amount corresponding to the clustering result to which the marker placement region B1, G5, Y4, L3, L4 of patient B belongs is ranked 2. The displacement amount corresponding to the clustering result to which the marker placement region B3, B6, G4, Y3 of patient B belongs is ranked 5. The displacement amount corresponding to the clustering result to which the marker placement region G3, L1 of patient B belongs is ranked 4.

[0130] The displacement amount corresponding to the cluster result to which the marker placement region B1, B3, B4, G1, Y2, Y3 of patient C belongs is ranked 1. The displacement amount corresponding to the cluster result to which the marker placement region B6, G4 of patient C belongs is ranked 2. The displacement amount corresponding to the cluster result to which the marker placement region B2, G2, G3 of patient C belongs is ranked 3. The displacement amount corresponding to the cluster result to which the marker placement region Y1, Y4, L1 of patient C belongs is ranked 4. The displacement amount corresponding to the cluster result to which the marker placement region L2, L3, L4 of patient C belongs is ranked 5. The displacement amount corresponding to the cluster result to which the marker placement region B5, G5 of patient C belongs is ranked as a null value.

[0131] Referring to Table 1, Table 1 is a table of displacement amount ranking based on the average of displacement amount ranking provided by the embodiment of the present application. The first column "displacement amount ranking" in Table 1 is the ranking result of the sixth column "average of displacement amount ranking" from large to small, the second column is the "region code" of 19 marker placements, and the third to fifth columns are the displacement ranking values corresponding to the cluster results of each marker placement region of three patients. Tables 2-4 have the same structure as Table 1.

[0132] Table 1

[0133]

[0134] It can be seen from Table 1 that if M=6 marker placement regions are to be selected from N=19 marker placement regions, the marker placement regions L1, L2, L3, L4, G4, B6 with the top 6 displacement amount rankings can be selected as the M marker placement regions. By placing the body surface markers in the six marker placement regions L1, L2, L3, L4, G4, B6, a respiratory motion correlation model with higher precision and stronger stability can be obtained in the process of establishing the respiratory motion correlation model based on the displacement amount of the body surface markers placed in the M marker placement regions, thereby improving the quality of the respiratory motion correlation model established based on the motion of the body surface markers.

[0135] In one embodiment, the displacement amount is taken as an example of x-axis displacement amount. From the data analysis of the above example, it can be known that:

[0136] The displacement amount corresponding to the clustering result to which the marker placement region B4, G1, G2, G3, G5 and Y1 of patient A belongs is ranked as 1, the displacement amount corresponding to the clustering result to which the marker placement region B1 and B3 of patient A belongs is ranked as 4, the displacement amount corresponding to the clustering result to which the marker placement region L2, L3 and L4 of patient A belongs is ranked as 3, the displacement amount corresponding to the clustering result to which the marker placement region B2, B5 and Y2 of patient A belongs is ranked as 5, the displacement amount corresponding to the clustering result to which the marker placement region B6, G4 and Y3 of patient A belongs is ranked as 2, and the displacement amount corresponding to the clustering result to which the marker placement region L1 and Y4 of patient A belongs is ranked as a null value.

[0137] The displacement amount corresponding to the clustering result to which the marker placement region B2, B5 and G1 of patient B belongs is ranked as 3. The displacement amount corresponding to the clustering result to which the marker placement region B4, G2, Y1, Y2 and L2 of patient B belongs is ranked as 1. The displacement amount corresponding to the clustering result to which the marker placement region B1, G5, Y4, L3 and L4 of patient B belongs is ranked as 2. The displacement amount corresponding to the clustering result to which the marker placement region B3, B6, G4 and Y3 of patient B belongs is ranked as 5. The displacement amount corresponding to the clustering result to which the marker placement region G3 and L1 of patient B belongs is ranked as 4.

[0138] The displacement amount corresponding to the clustering result to which the marker placement region B1, B3, B4, G1, Y2 and Y3 of patient C belongs is ranked as 3. The displacement amount corresponding to the clustering result to which the marker placement region B6 and G4 of patient C belongs is ranked as 2. The displacement amount corresponding to the clustering result to which the marker placement region B2, G2 and G3 of patient C belongs is ranked as 4. The displacement amount corresponding to the clustering result to which the marker placement region Y1, Y4 and L1 of patient C belongs is ranked as 1. The displacement amount corresponding to the clustering result to which the marker placement region L2, L3 and L4 of patient C belongs is ranked as 5. The displacement amount corresponding to the clustering result to which the marker placement region B5 and G5 of patient C belongs is ranked as a null value.

[0139] Referring to Table 2, Table 2 is a table of displacement amount ranking based on x-axis displacement amount ranking mean value provided by the embodiment of the present application.

[0140] Table 2

[0141]

[0142] It can be seen from Table 2 that if M=6 marker placement regions need to be selected from N=19 marker placement regions, the marker placement regions B2, B3, B5, L3, L4 and Y3 ranked first, second, third in displacement amount can be selected as the M marker placement regions. By placing body surface markers in the six marker placement regions B2, B3, B5, L3, L4 and Y3, a respiratory motion correlation model with higher precision and stronger stability can be obtained in the process of establishing a respiratory motion correlation model based on the displacement amount of the body surface markers placed in the M marker placement regions, thereby improving the quality of the respiratory motion correlation model established based on the movement of the body surface markers.

[0143] In one embodiment, the displacement amount is taken as an example of y-axis displacement amount. From the data analysis of the above example, it can be seen that:

[0144] The cluster result corresponding to the displacement amount of the marker placement regions B4, G1, G2, G3, G5 and Y1 of patient A is ranked 3. The cluster result corresponding to the displacement amount of the marker placement regions B1 and B3 of patient A is ranked 1. The cluster result corresponding to the displacement amount of the marker placement regions L2, L3 and L4 of patient A is ranked 5. The cluster result corresponding to the displacement amount of the marker placement regions B2, B5 and Y2 of patient A is ranked 2. The cluster result corresponding to the displacement amount of the marker placement regions B6, G4 and Y3 of patient A is ranked 4. The cluster result corresponding to the displacement amount of the marker placement regions L1 and Y4 of patient A is empty.

[0145] The cluster result corresponding to the displacement amount of the marker placement regions B2, B5 and G1 of patient B is ranked 4. The cluster result corresponding to the displacement amount of the marker placement regions B4, G2, Y1, Y2 and L2 of patient B is ranked 3. The cluster result corresponding to the displacement amount of the marker placement regions B1, G5, Y4, L3 and L4 of patient B is ranked 5. The cluster result corresponding to the displacement amount of the marker placement regions B3, B6, G4 and Y3 of patient B is ranked 1. The cluster result corresponding to the displacement amount of the marker placement regions G3 and L1 of patient B is ranked 2.

[0146] The displacement quantity corresponding to the clustering result to which the marker placement region B1, B3, B4, G1, Y2, Y3 of patient C belongs is ranked 2. The displacement quantity corresponding to the clustering result to which the marker placement region B6, G4 of patient C belongs is ranked 3. The displacement quantity corresponding to the clustering result to which the marker placement region B2, G2, G3 of patient C belongs is ranked 1. The displacement quantity corresponding to the clustering result to which the marker placement region Y1, Y4, L1 of patient C belongs is ranked 4. The displacement quantity corresponding to the clustering result to which the marker placement region L2, L3, L4 of patient C belongs is ranked 5. The displacement quantity corresponding to the clustering result to which the marker placement region B5, G5 of patient C belongs is ranked as a null value.

[0147] Referring to Table 3, Table 3 is a table of displacement quantity ranking based on y-axis displacement quantity ranking mean provided by the embodiment of the present application.

[0148] Table 3

[0149]

[0150]

[0151] It can be seen from Table 3 that if M=6 marker placement regions are to be selected from N=19 marker placement regions, the marker placement regions L2, L3, L4, Y1, Y4, G5 ranked first 6 in displacement quantity can be selected as the M marker placement regions. By placing the body surface markers in the 6 marker placement regions L2, L3, L4, Y1, Y4, G5, a respiratory motion correlation model with higher precision and stronger stability can be obtained in the process of establishing the respiratory motion correlation model based on the displacement quantity of the body surface markers placed in the M marker placement regions, thereby improving the quality of the respiratory motion correlation model established based on the motion of the body surface markers.

[0152] In one embodiment, the displacement quantity is taken as an example of z-axis displacement quantity. It can be known from the data analysis of the above example that:

[0153] The displacement quantity corresponding to the clustering result to which the marker placement region B4, G1, G2, G3, G5, Y1 of patient A belongs is ranked 2. The displacement quantity corresponding to the clustering result to which the marker placement region B1, B3 of patient A belongs is ranked 1. The displacement quantity corresponding to the clustering result to which the marker placement region L2, L3, L4 of patient A belongs is ranked 4. The displacement quantity corresponding to the clustering result to which the marker placement region B2, B5, Y2 of patient A belongs is ranked 5. The displacement quantity corresponding to the clustering result to which the marker placement region B6, G4, Y3 of patient A belongs is ranked 3. The displacement quantity corresponding to the clustering result to which the marker placement region L1, Y4 of patient A belongs is ranked as a null value.

[0154] The displacement amount corresponding to the clustering result to which the marker placement region B2, B5, G1 of patient B belongs is ranked 4. The displacement amount corresponding to the clustering result to which the marker placement region B4, G2, Y1, Y2, L2 of patient B belongs is ranked 3. The displacement amount corresponding to the clustering result to which the marker placement region B1, G5, Y4, L3, L4 of patient B belongs is ranked 1. The displacement amount corresponding to the clustering result to which the marker placement region B3, B6, G4, Y3 of patient B belongs is ranked 2. The displacement amount corresponding to the clustering result to which the marker placement region G3, L1 of patient B belongs is ranked 5.

[0155] The displacement amount corresponding to the clustering result to which the marker placement region B1, B3, B4, G1, Y2, Y3 of patient C belongs is ranked 1. The displacement amount corresponding to the clustering result to which the marker placement region B6, G4 of patient C belongs is ranked 2. The displacement amount corresponding to the clustering result to which the marker placement region B2, G2, G3 of patient C belongs is ranked 4. The displacement amount corresponding to the clustering result to which the marker placement region Y1, Y4, L1 of patient C belongs is ranked 3. The displacement amount corresponding to the clustering result to which the marker placement region L2, L3, L4 of patient C belongs is ranked 5. The displacement amount corresponding to the clustering result to which the marker placement region B5, G5 of patient C belongs is ranked as a null value.

[0156] Referring to Table 4, Table 4 is a table of displacement amount ranking based on z-axis displacement amount ranking mean value provided by the embodiment of the present application.

[0157] Table 4

[0158]

[0159]

[0160] It can be seen from Table 4 that if M=6 marker placement regions are to be selected from N=19 marker placement regions, the first 6 marker placement regions B2, B5, L1, L2, G3, G1 (or L3 or L4) in displacement amount ranking can be selected as the M marker placement regions. In the process of establishing a respiratory motion correlation model based on the displacement amount of the body surface markers placed in the 6 marker placement regions B2, B5, L1, L2, G3, G1, a respiratory motion correlation model with higher precision and stronger stability can be obtained, thereby improving the quality of the respiratory motion correlation model established based on the motion of the body surface markers.

[0161] It should be noted that since the number of samples selected in the above example is only 3, the accuracy, universality and robustness of the ranking result still have room for improvement.

[0162] The M marker placement regions can be ranked according to any one of the absolute displacement amount, the x-axis displacement amount, the y-axis displacement amount, or the z-axis displacement amount.

[0163] It should be noted that the M marker placement regions ranked according to the absolute displacement amount, the M marker placement regions ranked according to the x-axis displacement amount, the M marker placement regions ranked according to the y-axis displacement amount, and the M marker placement regions ranked according to the z-axis displacement amount are not necessarily completely the same.

[0164] Specifically, which index of the absolute displacement amount, the x-axis displacement amount, the y-axis displacement amount, or the z-axis displacement amount is selected for ranking can be selected from the mean values of the indexes of multiple patients, such as the absolute displacement amount or the y-axis displacement amount. The selection can also be made according to clinical needs (lesion position, patient individual characteristics, surgical planning, operating room conditions, etc.).

[0165] The x-axis is the left-right direction of the patient (the positive direction of the x-axis is the direction in which the left shoulder points to the right shoulder), the y-axis is the front-back direction of the patient (the positive direction of the y-axis is the direction in which the front chest points to the back), and the z-axis is the head-to-foot direction of the patient (the direction in which the head points to the foot).

[0166] In the embodiments of the present application, the N marker placement regions are ranked by the displacement amount ranking mean value corresponding to the clustering result to which each marker placement region belongs, without directly ranking according to the mean value of the displacement amount, so that the influence of the difference in respiratory motion between patients on the ranking can be avoided, and the universality of the ranking of the N marker placement regions is improved.

[0167] Optionally, the displacement amount includes an absolute displacement amount, an x-axis displacement amount, a y-axis displacement amount, and a z-axis displacement amount.

[0168] In step 105, the electronic device obtains the N marker placement regions corresponding to the displacement amount obtained from the L-class clustering results of the body surface motion of the multiple patients, and obtains M marker placement regions with larger displacement amounts, including:

[0169] (31) The electronic device obtains the displacement amount ranking of the first displacement amount corresponding to the clustering center of the clustering result to which each marker placement region of each patient belongs according to the first displacement amount; the first displacement amount is any one of the absolute displacement amount, the x-axis displacement amount, the y-axis displacement amount, and the z-axis displacement amount; the displacement amount ranking of the first displacement amount includes: the first displacement amount corresponding to the clustering result to which each marker placement region of each patient belongs, in all L clustering results corresponding to the first displacement amount, ranking from small to large;

[0170] (32) The electronic device ranks the first displacement amount corresponding to the clustering result of each marker placement area according to the displacement amount of a plurality of patients, to obtain a first displacement amount ranking mean value of the clustering result of each marker placement area;

[0171] (33) The electronic device sorts the first displacement amount ranking mean value of the clustering result of each marker placement area in descending order, to obtain a sorting of the first displacement amount ranking mean value of the clustering result of each marker placement area;

[0172] (34) The electronic device performs weighted summation on the sorting of the absolute displacement amount ranking mean value, the sorting of the x-axis displacement amount ranking mean value, the sorting of the y-axis displacement amount ranking mean value, and the sorting of the z-axis displacement amount ranking mean value corresponding to the clustering result of each marker placement area, to obtain a weighted sum value of the displacement amount ranking mean value sorting of the clustering result of each marker placement area;

[0173] (35) The electronic device sorts the weighted sum value of the displacement amount ranking mean value sorting of the clustering result of the N marker placement areas in ascending order, to obtain the top M marker placement areas.

[0174] Optionally, after step (35) is performed, step (36) can also be performed, wherein:

[0175] (36) According to the physiological structure of the human body, the body surface physiological feature position contained in each of the M marker placement areas is selected, and the body surface physiological feature position contained in each of the M marker placement areas is taken as the position description of the M marker placement areas on the body surface.

[0176] In the embodiments of the present application, N=19 (the region codes of the 19 marker placement areas are B1-B6, G1-G5, Y1-Y4, and L1-L4, respectively), as shown in Figure 6 , L=5, the plurality of patients are Q patients, and Q=9 is taken as an example, because of the length, Figure 4 and Figure 5 Sample 4 (patient A), sample 5 (patient B), and sample 7 (patient C) are taken as examples. The L-class clustering results include 5-class clustering results. Each class of clustering results corresponds to an absolute displacement amount, an x-axis displacement amount, a y-axis displacement amount, and a z-axis displacement amount. For example (see Figure 5 , Figure 6The absolute displacement amount corresponding to the clustering result to which the marker placement region B4, G1, G2, G3, G5, Y1 of patient A belongs is 1.83, the x-axis displacement amount is 0.03, the y-axis displacement amount is 1.82, and the z-axis displacement amount is -0.19. The absolute displacement amount corresponding to the clustering result to which the marker placement region B1, B3 of patient A belongs is 0.48, the x-axis displacement amount is -0.42, the y-axis displacement amount is 0.21, and the z-axis displacement amount is -0.12. The absolute displacement amount corresponding to the clustering result to which the marker placement region L2, L3, L4 of patient A belongs is 4.51, the x-axis displacement amount is 0.37, the y-axis displacement amount is 4.23, and the z-axis displacement amount is -1.53. The absolute displacement amount corresponding to the clustering result to which the marker placement region B2, B5, Y2 of patient A belongs is 2.53, the x-axis displacement amount is -0.46, the y-axis displacement amount is 1.79, and the z-axis displacement amount is -1.73. The absolute displacement amount corresponding to the clustering result to which the marker placement region B6, G4, Y3 of patient A belongs is 2.78, the x-axis displacement amount is 0.27, the y-axis displacement amount is 2.38, and the z-axis displacement amount is -1.42. The marker placement region L1, Y4 of patient A belongs to the empty clustering result, and the corresponding empty value is obtained.

[0177] The absolute displacement amount corresponding to the clustering result to which the marker placement region B2, B5, G1 of patient B belongs is 0.93, the x-axis displacement amount is 0.35, the y-axis displacement amount is -0.64, and the z-axis displacement amount is -0.58. The absolute displacement amount corresponding to the clustering result to which the marker placement region B4, G2, Y1, Y2, L2 of patient B belongs is 0.66, the x-axis displacement amount is -0.23, the y-axis displacement amount is -0.31, and the z-axis displacement amount is 0.54. The absolute displacement amount corresponding to the clustering result to which the marker placement region B1, G5, Y4, L3, L4 of patient B belongs is 0.85, the x-axis displacement amount is -0.25, the y-axis displacement amount is 0.79, and the z-axis displacement amount is -0.21. The absolute displacement amount corresponding to the clustering result to which the marker placement region B3, B6, G4, Y3 of patient B belongs is 1.27, the x-axis displacement amount is 1.19, the y-axis displacement amount is 0.11, and the z-axis displacement amount is -0.43. The absolute displacement amount corresponding to the clustering result to which the marker placement region G3, L1 of patient B belongs is 1.11, the x-axis displacement amount is 0.83, the y-axis displacement amount is -0.13, and the z-axis displacement amount is 0.72.

[0178] The absolute displacement amount corresponding to the clustering result to which the marker placement region B1, B3, B4, G1, Y2, Y3 of patient C belongs is 0.27, the x-axis displacement amount is 0.15, the y-axis displacement amount is -0.22, and the z-axis displacement amount is 0.02. The absolute displacement amount corresponding to the clustering result to which the marker placement region B6, G4 of patient C belongs is 1.24, the x-axis displacement amount is 0.14, the y-axis displacement amount is 0.95, and the z-axis displacement amount is -0.79. The absolute displacement amount corresponding to the clustering result to which the marker placement region B2, G2, G3 of patient C belongs is 1.98, the x-axis displacement amount is -0.32, the y-axis displacement amount is 0.22, and the z-axis displacement amount is -1.94. The absolute displacement amount corresponding to the clustering result to which the marker placement region Y1, Y4, L1 of patient C belongs is 2.44, the x-axis displacement amount is 0.02, the y-axis displacement amount is -2.30, and the z-axis displacement amount is -0.80. The absolute displacement amount corresponding to the clustering result to which the marker placement region L2, L3, L4 of patient C belongs is 5.09, the x-axis displacement amount is -0.37, the y-axis displacement amount is -3.97, and the z-axis displacement amount is -3.17. The marker placement region B5, G5 of patient C belongs to the empty clustering result, corresponding to the empty value.

[0179] From the data analysis of the above example (see Table 1), it can be seen that:

[0180] The displacement quantity ranking of the absolute displacement quantity corresponding to the clustering result to which the marker placement region B4, G1, G2, G3, G5, Y1 of patient A belongs is 2, the displacement quantity ranking of the absolute displacement quantity corresponding to the clustering result to which the marker placement region B1, B3 of patient A belongs is 1, the displacement quantity ranking of the absolute displacement quantity corresponding to the clustering result to which the marker placement region L2, L3, L4 of patient A belongs is 5, the displacement quantity ranking of the absolute displacement quantity corresponding to the clustering result to which the marker placement region B2, B5, Y2 of patient A belongs is 3, the displacement quantity ranking of the absolute displacement quantity corresponding to the clustering result to which the marker placement region B6, G4, Y3 of patient A belongs is 4, and the displacement quantity ranking of the absolute displacement quantity corresponding to the clustering result to which the marker placement region L1, Y4 of patient A belongs is null. The displacement quantity ranking of the absolute displacement quantity corresponding to the clustering result to which the marker placement region B2, B5, G1 of patient B belongs is 3. The displacement quantity ranking of the absolute displacement quantity corresponding to the clustering result to which the marker placement region B4, G2, Y1, Y2, L2 of patient B belongs is 1. The displacement quantity ranking of the absolute displacement quantity corresponding to the clustering result to which the marker placement region B1, G5, Y4, L3, L4 of patient B belongs is 2. The displacement quantity ranking of the absolute displacement quantity corresponding to the clustering result to which the marker placement region B3, B6, G4, Y3 of patient B belongs is 5. The displacement quantity ranking of the absolute displacement quantity corresponding to the clustering result to which the marker placement region G3, L1 of patient B belongs is 4. The displacement quantity ranking of the absolute displacement quantity corresponding to the clustering result to which the marker placement region B1, B3, B4, G1, Y2, Y3 of patient C belongs is 1. The displacement quantity ranking of the absolute displacement quantity corresponding to the clustering result to which the marker placement region B6, G4 of patient C belongs is 2. The displacement quantity ranking of the absolute displacement quantity corresponding to the clustering result to which the marker placement region B2, G2, G3 of patient C belongs is 3. The displacement quantity ranking of the absolute displacement quantity corresponding to the clustering result to which the marker placement region Y1, Y4, L1 of patient C belongs is 4. The displacement quantity ranking of the absolute displacement quantity corresponding to the clustering result to which the marker placement region L2, L3, L4 of patient C belongs is 5. The displacement quantity ranking of the absolute displacement quantity corresponding to the clustering result to which the marker placement region B5, G5 of patient C belongs is null.

[0181] As can be seen from Table 1, the rightmost column of Table 1 represents the average of the absolute displacement quantity ranking corresponding to the clustering result to which each marker placement region belongs, and the leftmost column of Table 1 represents the ranking of the average of the absolute displacement quantity ranking corresponding to the clustering result to which each marker placement region belongs.

[0182] Referring to Table 2, the displacement amount ordering of the x-axis displacement amount corresponding to the clustering result to which the marker placement region B4, G1, G2, G3, G5, Y1 of the patient A belongs is 1, the displacement amount ordering of the x-axis displacement amount corresponding to the clustering result to which the marker placement region B1, B3 of the patient A belongs is 4, the displacement amount ordering of the x-axis displacement amount corresponding to the clustering result to which the marker placement region L2, L3, L4 of the patient A belongs is 3, the displacement amount ordering of the x-axis displacement amount corresponding to the clustering result to which the marker placement region B2, B5, Y2 of the patient A belongs is 5, the displacement amount ordering of the x-axis displacement amount corresponding to the clustering result to which the marker placement region B6, G4, Y3 of the patient A belongs is 2, and the displacement amount ordering of the x-axis displacement amount corresponding to the clustering result to which the marker placement region L1, Y4 of the patient A belongs is null. The displacement amount ordering of the x-axis displacement amount corresponding to the clustering result to which the marker placement region B2, B5, G1 of the patient B belongs is 3. The displacement amount ordering of the x-axis displacement amount corresponding to the clustering result to which the marker placement region B4, G2, Y1, Y2, L2 of the patient B belongs is 1. The displacement amount ordering of the x-axis displacement amount corresponding to the clustering result to which the marker placement region B1, G5, Y4, L3, L4 of the patient B belongs is 2. The displacement amount ordering of the x-axis displacement amount corresponding to the clustering result to which the marker placement region B3, B6, G4, Y3 of the patient B belongs is 5. The displacement amount ordering of the x-axis displacement amount corresponding to the clustering result to which the marker placement region G3, L1 of the patient B belongs is 4. The displacement amount ordering of the x-axis displacement amount corresponding to the clustering result to which the marker placement region B1, B3, B4, G1, Y2, Y3 of the patient C belongs is 3. The displacement amount ordering of the x-axis displacement amount corresponding to the clustering result to which the marker placement region B6, G4 of the patient C belongs is 2. The displacement amount ordering of the x-axis displacement amount corresponding to the clustering result to which the marker placement region B2, G2, G3 of the patient C belongs is 4. The displacement amount ordering of the x-axis displacement amount corresponding to the clustering result to which the marker placement region Y1, Y4, L1 of the patient C belongs is 1. The displacement amount ordering of the x-axis displacement amount corresponding to the clustering result to which the marker placement region L2, L3, L4 of the patient C belongs is 5. The displacement amount ordering of the x-axis displacement amount corresponding to the clustering result to which the marker placement region B5, G5 of the patient C belongs is null.

[0183] As can be seen from Table 2, the rightmost column of Table 2 indicates the average of the x-axis displacement amount ordering corresponding to the clustering result to which each marker placement region belongs, and the leftmost column of Table 2 indicates the ordering of the average of the x-axis displacement amount ordering corresponding to the clustering result to which each marker placement region belongs.

[0184] Referring to Table 3, the displacement amount ordering of the y-axis displacement amount corresponding to the clustering result to which the marker placement region B4, G1, G2, G3, G5, Y1 of the patient A belongs is 3, the displacement amount ordering of the y-axis displacement amount corresponding to the clustering result to which the marker placement region B1, B3 of the patient A belongs is 1, the displacement amount ordering of the y-axis displacement amount corresponding to the clustering result to which the marker placement region L2, L3, L4 of the patient A belongs is 5, the displacement amount ordering of the y-axis displacement amount corresponding to the clustering result to which the marker placement region B2, B5, Y2 of the patient A belongs is 2, the displacement amount ordering of the y-axis displacement amount corresponding to the clustering result to which the marker placement region B6, G4, Y3 of the patient A belongs is 4, and the displacement amount ordering of the y-axis displacement amount corresponding to the clustering result to which the marker placement region L1, Y4 of the patient A belongs is null. The displacement amount ordering of the y-axis displacement amount corresponding to the clustering result to which the marker placement region B2, B5, G1 of the patient B belongs is 4. The displacement amount ordering of the y-axis displacement amount corresponding to the clustering result to which the marker placement region B4, G2, Y1, Y2, L2 of the patient B belongs is 3. The displacement amount ordering of the y-axis displacement amount corresponding to the clustering result to which the marker placement region B1, G5, Y4, L3, L4 of the patient B belongs is 5. The displacement amount ordering of the y-axis displacement amount corresponding to the clustering result to which the marker placement region B3, B6, G4, Y3 of the patient B belongs is 1. The displacement amount ordering of the y-axis displacement amount corresponding to the clustering result to which the marker placement region G3, L1 of the patient B belongs is 2. The displacement amount ordering of the y-axis displacement amount corresponding to the clustering result to which the marker placement region B1, B3, B4, G1, Y2, Y3 of the patient C belongs is 2. The displacement amount ordering of the y-axis displacement amount corresponding to the clustering result to which the marker placement region B6, G4 of the patient C belongs is 3. The displacement amount ordering of the y-axis displacement amount corresponding to the clustering result to which the marker placement region B2, G2, G3 of the patient C belongs is 1. The displacement amount ordering of the y-axis displacement amount corresponding to the clustering result to which the marker placement region Y1, Y4, L1 of the patient C belongs is 4. The displacement amount ordering of the y-axis displacement amount corresponding to the clustering result to which the marker placement region L2, L3, L4 of the patient C belongs is 5. The displacement amount ordering of the y-axis displacement amount corresponding to the clustering result to which the marker placement region B5, G5 of the patient C belongs is null.

[0185] As can be seen from Table 3, the rightmost column of Table 3 indicates the average of the ordering of the y-axis displacement amount corresponding to the clustering result to which each marker placement region belongs, and the leftmost column of Table 3 indicates the ordering of the average of the ordering of the y-axis displacement amount corresponding to the clustering result to which each marker placement region belongs.

[0186] Referring to Table 4, the displacement amount ordering of the z-axis displacement amount corresponding to the clustering result to which the marker placement region B4, G1, G2, G3, G5, Y1 of the patient A belongs is 2, the displacement amount ordering of the z-axis displacement amount corresponding to the clustering result to which the marker placement region B1, B3 of the patient A belongs is 1, the displacement amount ordering of the z-axis displacement amount corresponding to the clustering result to which the marker placement region L2, L3, L4 of the patient A belongs is 4, the displacement amount ordering of the z-axis displacement amount corresponding to the clustering result to which the marker placement region B2, B5, Y2 of the patient A belongs is 5, the displacement amount ordering of the z-axis displacement amount corresponding to the clustering result to which the marker placement region B6, G4, Y3 of the patient A belongs is 3, and the displacement amount ordering of the z-axis displacement amount corresponding to the clustering result to which the marker placement region L1, Y4 of the patient A belongs is null. The displacement amount ordering of the z-axis displacement amount corresponding to the clustering result to which the marker placement region B2, B5, G1 of the patient B belongs is 4. The displacement amount ordering of the z-axis displacement amount corresponding to the clustering result to which the marker placement region B4, G2, Y1, Y2, L2 of the patient B belongs is 3. The displacement amount ordering of the z-axis displacement amount corresponding to the clustering result to which the marker placement region B1, G5, Y4, L3, L4 of the patient B belongs is 1. The displacement amount ordering of the z-axis displacement amount corresponding to the clustering result to which the marker placement region B3, B6, G4, Y3 of the patient B belongs is 2. The displacement amount ordering of the z-axis displacement amount corresponding to the clustering result to which the marker placement region G3, L1 of the patient B belongs is 5. The displacement amount ordering of the z-axis displacement amount corresponding to the clustering result to which the marker placement region B1, B3, B4, G1, Y2, Y3 of the patient C belongs is 1. The displacement amount ordering of the z-axis displacement amount corresponding to the clustering result to which the marker placement region B6, G4 of the patient C belongs is 2. The displacement amount ordering of the z-axis displacement amount corresponding to the clustering result to which the marker placement region B2, G2, G3 of the patient C belongs is 4. The displacement amount ordering of the z-axis displacement amount corresponding to the clustering result to which the marker placement region Y1, Y4, L1 of the patient C belongs is 3. The displacement amount ordering of the z-axis displacement amount corresponding to the clustering result to which the marker placement region L2, L3, L4 of the patient C belongs is 5. The displacement amount ordering of the z-axis displacement amount corresponding to the clustering result to which the marker placement region B5, G5 of the patient C belongs is null.

[0187] As can be seen from Table 4, the rightmost column of Table 4 indicates the average of the z-axis displacement amount ordering corresponding to the clustering result to which each marker placement region belongs, and the leftmost column of Table 4 indicates the ordering of the average of the z-axis displacement amount ordering corresponding to the clustering result to which each marker placement region belongs.

[0188] Based on the above example, the sorting of the absolute displacement sorting mean value corresponding to the clustering result to which each marker placement region belongs, the sorting of the x-axis displacement sorting mean value, the sorting of the y-axis displacement sorting mean value, and the sorting of the z-axis displacement sorting mean value are obtained. The sorting of the absolute displacement sorting mean value corresponding to the clustering result to which each marker placement region belongs, the sorting of the x-axis displacement sorting mean value, the sorting of the y-axis displacement sorting mean value, and the sorting of the z-axis displacement sorting mean value are weighted and summed to obtain a weighted sum value of the displacement sorting mean value sorting corresponding to the clustering result to which each marker placement region belongs.

[0189] For example, the sorting of the absolute displacement sorting mean value, the sorting of the x-axis displacement sorting mean value, the sorting of the y-axis displacement sorting mean value, and the sorting of the z-axis displacement sorting mean value can be respectively assigned weights of 0.4, 0.2, 0.2, and 0.2.

[0190] The x-axis is the left-right direction of the patient (the positive direction of the x-axis is the direction in which the left shoulder points to the right shoulder), the y-axis is the front-back direction of the patient (the positive direction of the y-axis is the direction in which the front chest points to the back), and the z-axis is the head-to-foot direction of the patient (the direction in which the head points to the foot).

[0191] In the embodiments of the present application, the N marker placement regions are sorted according to the weighted sum value of the displacement sorting mean value sorting corresponding to the clustering result to which each marker placement region belongs, without directly sorting according to the mean value of the displacement, so that the influence of the respiratory motion difference between patients on the sorting can be avoided, and the universality of the sorting of the N marker placement regions is improved. In addition, the sorting of the N marker placement regions considers the absolute displacement, the x-axis displacement, the y-axis displacement, and the z-axis displacement, so that the robustness of the sorting of the N marker placement regions is improved.

[0192] In the embodiments of the present application, the M marker placement regions with larger displacement in the process of respiratory motion can be screened out from the N marker placement regions by analyzing the clustering result to which the body surface respiratory motion change amount belongs, so that the M marker placement regions with larger displacement in the process of respiratory motion are obtained. In the process of establishing the respiratory motion correlation model based on the displacement of the body surface markers placed in the M marker placement regions, a respiratory motion correlation model with higher accuracy and stronger stability can be obtained, and the quality of the respiratory motion correlation model established based on the body surface marker motion is improved.

[0193] Please refer to Figure 2 , Figure 2 is a flowchart of another method for determining a body surface marker placement region provided by the embodiments of the present application. As shown in Figure 2 , the method can include the following steps.

[0194] 201, the electronic device obtains sample data of a plurality of patients, the sample data comprising two medical image data at two different respiratory phases.

[0195] 202, the electronic device extracts, according to the two medical image data of each patient, two body surface contours of each patient, and calculates a displacement change amount of each pixel point of the body surface of each patient at the two different respiratory states according to the two body surface contours of each patient.

[0196] 203, the electronic device performs clustering processing according to the displacement change amount of each pixel point of the body surface of each patient at the two different respiratory states, to obtain L-class clustering results of the body surface motion of each patient.

[0197] The specific implementation of steps 201 to 203 can be referred to the above Figure 1 steps 101 to 103, which will not be described here.

[0198] 204, the electronic device divides the body surface contour according to the physiological structure of the human body to obtain K physiological structure regions, K being a positive integer.

[0199] In the embodiments of the present application, the body surface contour can be divided according to the physiological structure of the human body. For example, the front chest body surface is divided, the horizontal direction can be divided along the horizontal position of the rib of the human body (for example, the horizontal direction is divided along the lower edge of the 2nd, 4th, 6th, 8th and 10th rib, respectively), and the vertical direction (vertical direction) can be divided according to the vertical line of the chest body surface, such as the anterior median line, the clavicular midline, the sternum line, the parasternal line, the anterior axillary line, etc. For example, the electronic device can identify each rib lower edge from the CT image of the patient, and determine the anterior median line, the clavicular midline, the sternum line, the parasternal line, and the anterior axillary line according to the CT image.

[0200] The body surface of each patient can be divided into K physiological structure regions according to the same rule.

[0201] The appearance of each patient is different, and the common human structure features such as the manubrium, the xiphoid, the nipple, the rib, the navel, the anterior median line, etc. Since the physiological structure of most people includes the manubrium, the xiphoid, the nipple, the rib, the navel, the anterior median line, etc. Therefore, by dividing the K physiological structure regions through the common human structure features, and taking the body surface physiological structure features contained in each region of the M marker placement regions in the N marker placement regions as the position description of the region on the body surface for placing the body surface marker, the influence of individual differences on the modeling of the respiratory motion correlation model and the application of the model can be maximized.

[0202] The step 204 needs to be performed before the step 205. For example, the step 204 can be performed simultaneously with the step 201 or the step 202 or the step 203, or can be performed before the step 201 or the step 202 or the step 203, or can be performed after the step 201 or the step 202 or the step 203.

[0203] 205, the electronic device divides the patient's body surface into N marker placement regions according to the L-class clustering results of the body surface movements of the plurality of patients and the K physiological structure regions, maps the L-class clustering results of the body surface movements of each patient to the N marker placement regions, and each marker placement region of each patient belongs to one class of clustering results or an empty clustering result.

[0204] In the embodiments of the present application, the body surface contour of each patient can be divided according to the physiological structure of the human body to obtain K physiological structure regions, K is a positive integer, and the patient's body surface can be divided into N marker placement regions according to the L-class clustering results of the body surface movements of the plurality of patients and the K physiological structure regions.

[0205] The K physiological structure regions can be divided according to the common characteristics of the physiological structure of the human body for the body surface region where the body surface marker needs to be placed. The patient's body surface can be divided into N marker placement regions according to the L-class clustering results of the body surface movements of the plurality of patients and the K physiological structure regions. Specifically, the L-class clustering results of the body surface movements of each patient can be normalized by means of the K physiological structure regions and then mapped to the N marker placement regions. The body surface of each patient is divided into N marker placement regions in the same position, so that each marker placement region of each patient belongs to one class of clustering results as much as possible and belongs to an empty clustering result as little as possible, and the N marker placement regions are associated with the K physiological structure regions, so that the marker placement personnel can find the N marker placement regions according to the physiological structure of the human body.

[0206] 206, the electronic device obtains the displacement amount corresponding to the N marker placement regions obtained according to the L-class clustering results of the body surface movements of the plurality of patients, and obtains M marker placement regions with larger displacement amounts, M is a positive integer less than or equal to N.

[0207] 207, the electronic device selects the body surface physiological feature positions contained in each region of the M marker placement regions according to the physiological structure of the human body and in combination with the K physiological structure regions, and takes the body surface physiological feature positions contained in each region of the M marker placement regions as the position description of the M marker placement regions on the body surface.

[0208] In the embodiments of the present application, in order to accurately and quickly position M marker placement regions on different patient body surfaces, the body surface physiological feature positions included in each of the M marker placement regions are selected according to the human physiological structure, the body surface physiological feature positions included in each of the M marker placement regions are taken as the position description of the M marker placement regions on the body surface, and the marker placement personnel can quickly and accurately place the markers in the corresponding regions according to the position description of each marker placement region. The position of the M marker placement regions on the body surface is described according to the human physiological structure, so that the uniformity of marker placement can be ensured.

[0209] The specific implementation of step 206 can be referred to step 105 in the above Figure 1 , which will not be described here again.

[0210] The above method can be applied to a respiratory motion correlation model based on a body surface marker, so as to improve the quality of the respiratory motion correlation model based on the body surface marker. The respiratory motion correlation model based on the body surface marker can include any one of a neural network model, a support vector machine model, a linear regression model, and a model established by a statistical method based on a probability distribution.

[0211] The embodiments of the present application provide a specific implementation step, method and application process for determining a lung body surface marker placement scheme.

[0212] 1) Data acquisition: In this example, two full lung CT data (scanning layer thickness 1.5 mm) of 9 patients are collected, which basically meet the requirements of the sample in the embodiments of the present application, and the patients are in supine position. The body posture and body position of each patient are unchanged when acquiring two full lung CT data.

[0213] 2) Displacement calculation: The patient body surface contour is manually labeled and extracted from the CT image by using a medical image analysis software (such as 3D Slicer software), and the body surface displacement change vector of the CT image of each patient in two different breathing states is calculated by using a registration algorithm (such as B-Spline algorithm in Elastix toolbox).

[0214] 3) Body surface partition: According to the human physiological structure, the lung body surface is divided into regions, which are divided along the lower edges of the 2nd, 4th, 6th, 8th and 10th ribs in the horizontal direction, and are divided according to the anterior median line, the midclavicular line, the parasternal line and the anterior axillary line in the vertical direction. Please refer to Figure 3 , Figure 3 is a schematic diagram provided by the embodiments of the present application for dividing the lung body surface into regions according to the human physiological structure. As shown in Figure 3 .The left side is a human structure schematic diagram and the corresponding body surface division, and the right side is the lung body surface morphology of a patient extracted from a certain sample CT image and the corresponding body surface division. Figure 3

[0215] 4) Displacement clustering: obtain the body surface regions with similar respiratory motion changes using a fuzzy clustering algorithm (e.g., Fuzzy C means), the clustering feature parameters in this example are the motion displacement amounts in the three directions of the Cartesian coordinate system (x, y, z), and the body surface motion changes are clustered into 5 categories. Figure 4 and Figure 5 The clustering results of three samples (sample 4, sample 5, and sample 7) are shown. Figure 4 is a schematic diagram of the clustering results of the motion of each point on the body surface of three samples in a three-dimensional space obtained by a clustering algorithm, Figure 5 is a schematic diagram of mapping the clustering results of the body surface motion of 3 samples to the body surface. Figure 5 The first row of is the position of each clustering center, Figure 5 The second row of is the mapping of the clustering results to the body surface, Figure 4 as shown in the schematic diagram, Figure 5 The third row of is the sample body surface partition diagram extracted from the CT image on the left, and the absolute displacement values of each clustering center sorted in descending order on the right. Figure 4 and Figure 5 In, the gray scale of the clustering area corresponds to the gray scale of the displacement data of the area.

[0216] 5) Result analysis: according to the clustering feature parameters, four motion feature indexes are set in this example, namely the three-direction displacements in the Cartesian coordinate system (x, y, z) and the Euclidean absolute displacement, hereinafter referred to as displacement index.

[0217] According to the body surface clustering results, the body surface respiratory motion displacement is used as the sorting standard, wherein the weight values of the three-direction displacements (x, y, z) are all 0.2, and the weight of the absolute displacement is larger, which is 0.4. According to this rule, the priority selection order of the optimal placement area of the lung body surface marker is calculated.

[0218] (a) Please refer to Figure 6 , Figure 6 is a schematic diagram provided by an embodiment of the present application for setting the candidate placement area of the body surface marker and encoding each area. As shown in Figure 6 , according to the common and similar features of the clustering results of the body surface motion of all samples and the division of the body surface area, 19 candidate placement areas of the body surface marker are set in this example, which are divided into 4 groups: 6 in the upper and middle lung field, 5 in the lower lung field, and 4 in each of the two abdominal regions.

[0219] As can be seen from Figure 5 , the body surface motion changes are clustered into 5 categories: R, G, B, Y, and C, Figure 5 The first row of shows the coordinates of the clustering centers of each category, which shows the displacement amounts of the clustering centers of each category in the x, y, and z directions. Figure 5The second row of (a) shows the mapping of the 5-class clustering results of each sample on the CT image, Figure 5 The asterisks in the second row of (a) represent N(N=19) marker placement regions on the CT image, Figure 5 The left side of the third row of (a) is a schematic diagram of the mapping of the 5-class clustering results on the CT image to N marker placement regions, and each circle represents one marker placement region. The letters in each circle represent the clustering result to which the marker placement region belongs. Figure 5 The hollow circles in the third row of (a) indicate that the marker placement region does not belong to a single cluster. Such marker placement regions are often located at the boundary of multiple clusters (see Figure 4 ), and in this example, such marker placement regions are ignored, or they can be assigned to one of the candidate regions according to the distribution pattern of the clusters.

[0220] Special notes: Figure 6 , Figure 10 , Figure 11 The gray scale of the circles in (a) corresponds to the grouping of the 19 candidate placement regions, Figure 4 , Figure 5 , Figure 7 The gray scale of the circles corresponds to the sample clustering region, and the two meanings are different and unrelated.

[0221] (b) According to the position of each sample clustering center, sort the displacement and Euclidean absolute displacement in the x, y, and z directions from small to large. Figure 7 Taking sample 5 as an example, the displacement sorting values of the four displacement indicators in each marker placement region are shown. In this way, each clustering region of sample 5 obtains the sorting values of the four displacement indicators; Figure 7 is a schematic diagram of the sorting values of the four displacement indicators of each clustering region of sample 5 provided by the embodiment of the present application.

[0222] (c) Calculate the mean value of the displacement sorting values of the same clustering region of all samples, see Figure 8 the last column “displacement sorting mean value” of the four displacement indicator table in (a). Figure 8 is a schematic diagram of calculating the displacement sorting mean value of the clustering region and sorting according to the displacement sorting mean value.

[0223] Special notes: In this example, the displacement sorting mean value of region L3 is not calculated because the 9 samples used are all clustered in the same class, so the displacement sorting mean value of L3 is the same as that of L4. In actual operation and use, regions with adjacent spatial positions and the same motion characteristics can be combined into a larger marker placement region.

[0224] (d) Sort the candidate placement region displacement sorting mean values calculated in step (c) from large to small, seeFigure 8 The first column of the four displacement index tables is "displacement ranking".

[0225] The ranking of any one displacement index can be used alone as a priority selection criterion for the placement of lung surface markers. Figure 10 In the second to fifth surface CT images, according to Figure 8 The "displacement ranking" of the four displacement indexes shown in the table is obtained. Figure 10 is a schematic diagram of the priority selection order of a lung surface marker candidate placement area provided by the embodiment of the present application.

[0226] (e) The four displacement indexes, i.e., the "displacement ranking" values of the three direction displacements (x, y, z) and the absolute displacement (z) are respectively assigned weights (0.2, 0.2, 0.2, 0.4), and after weighted addition, they are again ranked from small to large to obtain the comprehensive priority selection order of each marker placement area calculated based on the displacement size of the surface respiratory motion. Figure 8 The table in the above (e) shows the calculation process from left to right, and the conclusion is summarized in the "total ranking" column of the rightmost table, which is mapped to Figure 9 the first surface CT image in the upper left corner. Figure 10 is a schematic diagram of the priority order of a lung surface marker candidate placement area calculated according to respiratory motion displacement provided by the embodiment of the present application. Figure 9

[0227] 6) Marker positioning: Assuming that the clinical requirement is that the patient's surface markers do not exceed 10, please refer to Figure 11 , Figure 11 shows a marker placement scheme that integrates the design of four displacement indexes of the lung surface. Figure 11 is a schematic diagram of a marker placement scheme and a description of the placement position of each marker provided by the embodiment of the present application.

[0228] Using the physiological structure characteristics of the human body, the specific placement position of the marker in each selected placement area is described, so that the operator (medical staff) can quickly find the exact position of the placement of these markers on the surface of most patients during the sample collection and clinical treatment process.

[0229] The base area of the surface marker actually used is generally less than 5-6 / 1000 of the area of the circle in Figure 11 During the sampling and treatment process, the marker placement position can be appropriately adjusted as needed, for example, to avoid the position where the operation is performed, the special situation of the patient's surface, to avoid artifacts generated by the marker tracking instrument, and the like. As long as it is within the placement area, the model accuracy of the sample data on the respiratory motion correlation model and the applicability of the implementation model to most patients are not greatly affected.​

[0230] The embodiment of the present application has the following advantages:

[0231] The cluster method is used to comprehensively quantify the respiratory motion state of each part of the body surface, and the motion law is analyzed to provide a scientific basis for the marker placement position, which can improve the scientific nature of the marker placement position.

[0232] Based on the clustering results of the body surface respiratory motion, the motion similar regions are divided, which ensures the integrity, accuracy, specificity of the placement position, and the uniformity of the region.

[0233] Based on the sorting and weighted combination of different displacement indicators of different body surface regions, the priority selection order of the marker placement region is obtained. When formulating the marker placement scheme, the best placement position can be selected according to actual needs. For example, M (M < N) regions are selected from N marker placement regions, and the selected regions require that the respiratory motion change is most significant, does not affect the implementation of the operation, is closely related to the lesion position, and is uniformly distributed on the body surface, etc.

[0234] The marker placement is standardized. The standardized marker placement scheme not only makes the marker placement operation simple and efficient, but also ensures the quality of the sample data;

[0235] The stability of the respiratory motion correlation model established based on the body surface marker motion is improved. The embodiment of the present application reveals the body surface respiratory motion law, sets the body surface marker placement position, and ensures that the deviation of each sample marker placement position has the least influence on the model when data sampling is performed for model establishment;

[0236] When the model is used, the placement position of the body surface marker of each patient is most suitable for the requirements of the model;

[0237] The compatibility of the respiratory motion correlation model established based on the body surface marker motion is improved. The appearance of each patient is different. The common structural features of the human body, such as the manubrium, xiphoid, nipple, rib, navel, and anterior median line, are used as reference marks for the actual placement position of the body surface marker, which maximally eliminates the influence of individual differences on modeling and model application;

[0238] The precision of the respiratory motion correlation model established based on the body surface marker motion is improved. The stability and compatibility of the model are the premise and guarantee of the effectiveness and accuracy of the model;

[0239] Simple and easy to implement, the embodiment of the present application is flexible in strategy, clear and simple in operation, wide in application, has no special hardware requirements, has no influence on treatment activities, and has no additional harm to patients and medical staff.

[0240] The above describes the scheme of the embodiments of the present application from the perspective of the method side. It can be understood that the electronic device includes hardware structures and / or software modules corresponding to the execution of each function to achieve the above functions. Those skilled in the art should easily realize that, in combination with the unit and algorithm steps of each example described in the embodiments provided herein, the present application can be realized in the form of hardware or a combination of hardware and computer software. Whether a certain function is executed in the form of hardware or computer software driven hardware depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0241] The embodiments of the present application can divide the functional units of the electronic device according to the above method examples. For example, each functional unit can be divided according to each function, or two or more functions can be integrated into one processing unit. The integrated unit can be realized in the form of hardware or software functional unit. It should be noted that the division of units in the embodiments of the present application is illustrative and is only a logical functional division. Actual implementation can have another division method.

[0242] Please refer to Figure 12 , Figure 12 is a structural schematic diagram of a device for determining a body surface marker placement area provided by the embodiments of the present application. The device for determining a body surface marker placement area 1200 is applied to an electronic device. The device for determining a body surface marker placement area 1200 can include an acquisition unit 1201, a calculation unit 1202, a clustering unit 1203, a mapping unit 1204, and a region determination unit 1205, wherein:

[0243] The acquisition unit 1201 is configured to acquire sample data of a plurality of patients, the sample data including two medical image data at two different respiratory phases; wherein the patient's posture and body position are unchanged when acquiring the two medical image data;

[0244] The calculation unit 1202 is configured to extract two body surface contours of each patient according to the two medical image data of each patient, and calculate a displacement change amount of each pixel point of the body surface of each patient at the two different respiratory states according to the two body surface contours of each patient.

[0245] The clustering unit 1203 is configured to perform clustering processing according to the displacement change amount of each pixel point of the body surface of each patient at the two different respiratory states, to obtain an L-class clustering result of the body surface motion of each patient, L being a positive integer greater than or equal to 2.

[0246] The mapping unit 1204 is configured to divide a patient's body surface into N marker placement regions according to the distribution of the L-class clustering results of the body surface movements of the plurality of patients, N being a positive integer greater than or equal to 2; and map the L-class clustering results of the body surface movements of each patient to the N marker placement regions, each marker placement region of each patient belonging to one clustering result or an empty clustering result.

[0247] The region determining unit 1205 is configured to obtain M marker placement regions with larger displacement amounts from the N marker placement regions corresponding to the clustering results of the L-class clustering results of the body surface movements of the plurality of patients, M being a positive integer less than or equal to N.

[0248] Optionally, the displacement amount includes any one of an absolute displacement amount, an x-axis displacement amount, a y-axis displacement amount, and a z-axis displacement amount.

[0249] The region determining unit 1205 obtains M marker placement regions with larger displacement amounts from the N marker placement regions corresponding to the clustering results of the L-class clustering results of the body surface movements of the plurality of patients according to the displacement amounts of the N marker placement regions, and the method includes the following steps: obtaining a displacement amount sorting corresponding to each clustering result of each marker placement region of each patient according to the displacement amount corresponding to each clustering result of each marker placement region of each patient; the displacement amount sorting includes: the displacement amount corresponding to each clustering result of each marker placement region of each patient is sorted in ascending order among all displacement amounts corresponding to the L clustering results; calculating a clustering result displacement amount sorting average of each marker placement region according to the displacement amount sorting corresponding to the clustering result of each marker placement region of the plurality of patients; and sorting the clustering result displacement amount sorting averages of the N marker placement regions in descending order to obtain M marker placement regions ranked in the top M.

[0250] Optionally, the displacement amount includes an absolute displacement amount, an x-axis displacement amount, a y-axis displacement amount, and a z-axis displacement amount; the region determination unit 1205 obtains a displacement amount corresponding to an N marker placement region from an L-class clustering result of body surface movement of the plurality of patients, and obtains M marker placement regions with larger displacement amounts, including: obtaining a displacement amount sorting of a first displacement amount corresponding to a clustering result of each marker placement region of each patient according to the first displacement amount corresponding to the clustering result of each marker placement region of each patient; the first displacement amount is any one of the absolute displacement amount, the x-axis displacement amount, the y-axis displacement amount, and the z-axis displacement amount; the displacement amount sorting of the first displacement amount includes: the first displacement amount corresponding to the clustering result of each marker placement region of each patient, in all L clustering results, in ascending order; obtaining a first displacement amount sorting mean of each marker placement region from the displacement amount sorting of the first displacement amount corresponding to the clustering result of each marker placement region of the plurality of patients; sorting the first displacement amount sorting mean of each marker placement region in descending order to obtain a sorting of the first displacement amount sorting mean of each marker placement region; obtaining a weighted sum value of the sorting of the clustering result corresponding to each marker placement region from the sorting of the absolute displacement amount sorting mean, the x-axis displacement amount sorting mean, the y-axis displacement amount sorting mean, and the z-axis displacement amount sorting mean; sorting the weighted sum value of the displacement amount sorting mean of the clustering result of the N marker placement regions in ascending order to obtain the top M marker placement regions.

[0251] Optionally, the first time body surface contour of each patient includes a first set of body surface pixels, and the second time body surface contour of each patient includes a second set of body surface pixels; the calculation unit 1202 calculates a displacement change amount of each pixel point of the body surface of each patient in the two different breathing states according to the two body surface contours of each patient, including: calculating a displacement change amount of the corresponding pixel points of each patient in the two different breathing states according to the coordinates of the corresponding pixel points on the two body surface contours of each patient.

[0252] Optionally, the clustering unit 1203 performs clustering processing on the displacement change amount of each pixel point of the body surface of each patient in the two different breathing states to obtain an L-class clustering result of the body surface movement of each patient, including: performing fuzzy clustering processing on the displacement change amount of each pixel point of the body surface of each patient in the two different breathing states to obtain an L-class clustering result of the body surface movement of each patient.

[0253] Optionally, the mapping unit 1204 maps the L-class clustering result of the body surface movement of each patient to the N marker placement regions, including:

[0254] mapping the L-class clustering result of the body surface movement of each patient to the N marker placement regions to obtain an initial clustering result of the N marker placement regions of each patient; in a case where the initial clustering result of a first marker placement region completely belongs to one of the L classes, taking the one class clustering result as the clustering result to which the first marker placement region belongs; the first marker placement region is any one of the N marker placement regions; in a case where the initial clustering result of the first marker placement region contains at least two class clustering results in the L classes, determining one class clustering result from the at least two class clustering results as the clustering result to which the first marker placement region belongs, or attributing the clustering result to which the first marker placement region belongs to an empty clustering result.

[0255] Optionally, the device 1200 for determining a body surface marker placement region further includes a division unit 1206.

[0256] The division unit 1206 is configured to divide the body surface contour according to a human physiological structure to obtain K physiological structure regions, K being a positive integer.

[0257] The mapping unit 1204 divides the body surface of a patient into N marker placement regions according to the distribution of the L-class clustering result of the body surface movement of a plurality of patients on the body surface contour, including: dividing the body surface of a patient into N marker placement regions according to the distribution of the L-class clustering result of the body surface movement of a plurality of patients on the body surface contour and the K physiological structure regions.

[0258] Optionally, the device 1200 for determining a body surface marker placement region further includes a region position description unit 1207.

[0259] The region position description unit 1207 is configured to select, according to a human physiological structure and in combination with the K physiological structure regions, a body surface physiological feature position contained in each of the M marker placement regions, and take the body surface physiological feature position contained in each of the M marker placement regions as a position description of the M marker placement regions on the body surface.

[0260] In the embodiments of the present application, the acquisition unit 1201, the calculation unit 1202, the clustering unit 1203, the mapping unit 1204, the region determination unit 1205, the division unit 1206, and the region position description unit 1207 can be processors in electronic devices.

[0261] In the embodiments of the present application, the L-class clustering results of the body surface movements of multiple patients can be analyzed, and M marker placement regions with larger displacement amounts of the N marker placement regions on the body surface in two different respiratory phases can be analyzed from the sample data of the multiple patients, so as to screen M marker placement regions with larger displacement amounts in the process of respiratory movement, and place the body surface markers in the M marker placement regions. In the process of establishing the respiratory movement correlation model based on the displacement amounts of the body surface markers placed in the M marker placement regions, a respiratory movement correlation model with higher precision and stronger stability can be obtained, and the quality of the respiratory movement correlation model established based on the movement of the body surface markers can be improved.

[0262] Please refer to Figure 13 , Figure 13 is a structural schematic diagram of an electronic device provided by the embodiments of the present application, as shown in Figure 13 , the electronic device 1300 includes a processor 1301 and a memory 1302, and the processor 1301 and the memory 1302 can be connected to each other through a communication bus 1303. The communication bus 1303 can be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The communication bus 1303 can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 13 , only one thick line is used, but it does not mean that there is only one bus or only one type of bus. The memory 1302 is used to store a computer program, and the computer program includes program instructions, and the processor 1301 is configured to invoke the program instructions, and the above program includes part or all of the steps of the method contained in the above program. Figures 1-2

[0263] The processor 1301 can be a general central processing unit (CPU), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits for controlling the execution of the above program.

[0264] ​The memory 1302 can be read-only memory (ROM) or other type of static storage devices that can store static information and instructions, random access memory (RAM), or other type of dynamic storage device that can store information and instructions, electrically erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disk storage, magnetic disk storage or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and that can be accessed by a computer, but is not limited to this. The memory can exist independently, and is connected with the processor through the bus. The memory can also be integrated with the processor.

[0265] The electronic device 1300 can also include a display 1304. The display 1304 can include any of a liquid crystal display, an LED display, an OLED display.

[0266] In addition, the electronic device 1300 can also include a communication interface (such as a USB interface, a microphone interface, etc.), an antenna, and other general-purpose components, which are not described here in detail.

[0267] The embodiments of the present application also provide a computer readable storage medium, wherein the computer readable storage medium stores a computer program for electronic data exchange, and the computer program causes a computer to execute part or all steps of any one of the methods for determining a body surface marker placement area as described in the above method embodiments.

[0268] It should be noted that, for the above-mentioned method embodiments, in order to simply describe, they are all expressed as a combination of a series of actions, but those skilled in the art should know that the present application is not limited to the action order described, because according to the present application, certain steps can be performed in other order or simultaneously. Secondly, those skilled in the art should know that the embodiments described in the specification all belong to preferred embodiments, and the actions and modules involved are not necessarily necessary for the present application.

[0269] In the above embodiments, the description of each embodiment has its own focus, and the parts not described in detail in a certain embodiment can be referred to the related description of other embodiments.

[0270] In several embodiments provided in the present application, it should be understood that the disclosed apparatus can be implemented in other manners. For example, the division of the apparatus embodiments described above is merely an example, and the division can be other forms. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the displayed or discussed mutual couplings or direct couplings or communication connections can be indirect couplings or communication connections through some interfaces, devices or units, and can be in electrical, mechanical or other forms.

[0271] The units described as separate components can or can not be physically separate, and the components shown as units can or can not be physical units, i.e., can be located in one place, or can be distributed on multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the embodiment.

[0272] In addition, the functional units in each embodiment of the application can be integrated into a processing unit, or each unit can be physically present separately, or two or more units can be integrated into one unit. The integrated unit can be realized in the form of hardware or in the form of a software program module.

[0273] When the integrated unit is realized in the form of a software program module and sold or used as an independent product, it can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application essentially or the part that contributes to the prior art or the whole or part of the technical solutions can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for making a computer device (which can be a personal computer, a server or a network device, etc.) execute all or part of the steps of the methods described in the embodiments of the present application. The aforementioned storage medium includes: a U disk, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk, and various media that can store program codes.

[0274] A person of ordinary skill in the art can understand that all or part of the steps of the various methods in the above embodiments can be completed by a program instructing relevant hardware, and the program can be stored in a computer readable storage medium, which can include a flash disk, a read-only memory, a random access memory, a magnetic disk or an optical disk, etc.

[0275] The above has carried out the detailed introduction to the embodiment of the application, the principle and implementation mode of the application have been described by applying specific examples in this paper, the above embodiment explanation is only used for helping understanding the method of the application and its core idea; at the same time, for the general technical personnel in the art, according to the idea of the application, the specific implementation mode and application range will have the change, and the above is described, the content of the specification should not be understood as the limitation of the application.

Claims

1. A method for determining the placement area of ​​surface markers, characterized in that, The method is applied to an electronic device, and the method includes: Acquire sample data from multiple patients, including two medical imaging data at two different respiratory phases; Two body surface contours of each patient are extracted based on two medical imaging data of each patient, and the displacement change of each pixel point on the body surface of each patient is calculated based on the two body surface contours of each patient under the two different breathing states. Clustering is performed on each patient's body surface pixels under the two different breathing states to obtain L-class clustering results for each patient's body surface movement, where L is a positive integer greater than or equal to 2. Based on the distribution of L-class clustering results of body surface movements of multiple patients on the body surface contour, the patient's body surface is divided into N marker placement areas, where N is a positive integer greater than 2; The L-class clustering results of the body surface movements of each patient are mapped to the N marker placement areas. Each marker placement area of ​​each patient belongs to one clustering result or an empty clustering result. Based on the displacement amount corresponding to the N marker placement areas obtained from the L-class clustering results of the body surface movements of the multiple patients, the M marker placement areas with larger displacement amounts are obtained, where M is a positive integer less than or equal to N.

2. The method according to claim 1, characterized in that, After obtaining the M marker placement areas with large displacements, the method further includes: Based on the human physiological structure, the physiological features of the body surface contained in each of the M marker placement areas are selected, and the physiological features of the body surface contained in each of the M marker placement areas are used as the location description of the M marker placement areas on the body surface.

3. The method according to claim 1, characterized in that, The displacement includes any one of absolute displacement, x-axis displacement, y-axis displacement, and z-axis displacement; The displacement amounts corresponding to the N marker placement areas obtained based on the L-class clustering results of the body surface movements of the multiple patients are used to obtain the M marker placement areas with larger displacement amounts, including: Based on the displacement amount corresponding to the cluster result of each marker placement area for each patient, the displacement amount corresponding to the cluster result of each marker placement area for each patient is sorted in ascending order. The displacement amount sorting includes: the displacement amount corresponding to the cluster result of each marker placement area for each patient is sorted in ascending order among the displacement amounts corresponding to all L cluster results. Based on the displacement ranking of the cluster results corresponding to each marker placement area of ​​multiple patients, calculate the average displacement ranking of the cluster results corresponding to each marker placement area; The average displacement values ​​corresponding to the clustering results of the N marker placement areas are sorted from largest to smallest to obtain the top M marker placement areas.

4. The method according to claim 1, characterized in that, The displacement includes absolute displacement, x-axis displacement, y-axis displacement, and z-axis displacement; The displacement amounts corresponding to the N marker placement areas obtained based on the L-class clustering results of the body surface movements of the multiple patients are used to obtain the M marker placement areas with larger displacement amounts, including: Based on the first displacement amount corresponding to the cluster result of each marker placement area for each patient, the displacement amount ranking of the first displacement amount corresponding to the cluster result of each marker placement area for each patient is obtained; the first displacement amount is any one of the absolute displacement amount, the x-axis displacement amount, the y-axis displacement amount, and the z-axis displacement amount; the displacement amount ranking of the first displacement amount includes: the first displacement amount corresponding to the cluster result of each marker placement area for each patient is sorted in ascending order among the first displacement amounts corresponding to all L cluster results; Based on the displacement of the first displacement corresponding to the cluster result of each marker placement area of ​​multiple patients, the average value of the first displacement corresponding to the cluster result of each marker placement area is obtained. Sort the first displacement mean of the clustering results for each marker placement area in descending order to obtain the sorting of the first displacement mean of the clustering results for each marker placement area. The weighted sum of the average absolute displacement, average x-axis displacement, average y-axis displacement, and average z-axis displacement corresponding to the clustering results of each marker placement area is obtained by performing a weighted sum. The weighted sum of the displacement values ​​corresponding to the clustering results of the N marker placement areas is sorted from smallest to largest to obtain the top M marker placement areas.

5. The method according to claim 1, characterized in that, The first body surface contour of each patient includes a first body surface pixel set, and the second body surface contour of each patient includes a second body surface pixel set. The calculation of the displacement change of each pixel point on the body surface of each patient under the two different respiratory states based on the two body surface contours of each patient includes: Based on the coordinates of the corresponding pixels on the body surface contour of each patient in the two different breathing states, calculate the displacement change of the corresponding pixels of each patient in the two different breathing states.

6. The method according to claim 1, characterized in that, The step of mapping the L-class clustering results of each patient's body surface movement to the N marker placement areas includes: The L-class clustering results of the body surface movements of each patient are mapped to the N marker placement areas to obtain the initial clustering results of the N marker placement areas for each patient; If the initial clustering result of the first marker placement area completely belongs to one of the clustering results in class L, then the clustering result in class L is taken as the clustering result to which the first marker placement area belongs; the first marker placement area is any one of the N marker placement areas; If the initial clustering result of the first marker placement area contains at least two clustering results in L classes, one clustering result is determined from the at least two clustering results as the clustering result to which the first marker placement area belongs, or the clustering result to which the first marker placement area belongs is assigned to the empty clustering result.

7. The method according to any one of claims 1 to 6, characterized in that, Before mapping the L-class clustering results of each patient's body surface movement to the N marker placement areas, the method further includes: The body surface contour is divided according to the human physiological structure to obtain K physiological structure regions, where K is a positive integer; The distribution of L-class clustering results of body surface movements from multiple patients onto the body surface contour divides the patient's body surface into N marker placement areas, including: Based on the L-class clustering results of multiple patients' body surface movements, the distribution of body surface contours, and the K physiological structural regions, the patient's body surface is divided into N marker placement areas.

8. A device for determining the placement area of ​​a body surface marker, characterized in that, The device is used in an electronic device, and the device includes: An acquisition unit is used to acquire sample data from multiple patients, the sample data including two medical imaging data at two different respiratory phases; The calculation unit is used to extract two body surface contours of each patient based on two medical imaging data of each patient, and to calculate the displacement change of each pixel point on the body surface of each patient under the two different breathing states based on the two body surface contours of each patient. The clustering unit is used to perform clustering processing based on the displacement change of each pixel point on the body surface of each patient in the two different breathing states, to obtain the L-class clustering result of the body surface movement of each patient, where L is a positive integer greater than or equal to 2. The mapping unit is used to divide the patient's body surface into N marker placement areas based on the distribution of L-class clustering results of body surface movements of multiple patients on the body surface contour, where N is a positive integer greater than or equal to 2; and to map the L-class clustering result of body surface movements of each patient to the N marker placement areas, where each marker placement area of ​​each patient belongs to one clustering result or an empty clustering result. The region determination unit obtains the M marker placement regions with larger displacements based on the displacement amounts corresponding to the N marker placement regions clustering results obtained from the L-class clustering results of the body surface movements of the multiple patients, where M is a positive integer less than or equal to N.

9. An electronic device, characterized in that, The device includes a processor and a memory, the memory being used to store a computer program, the computer program including program instructions, and the processor being configured to invoke the program instructions to perform the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, the computer program including program instructions that, when executed by a processor, cause the processor to perform the method as described in any one of claims 1 to 7.

Images