Tumor target region pose real-time tracking and prediction method, system, device and medium

Abstract

The application provides a tumor target area pose real-time tracking and prediction method and system, equipment and a medium. The abdominal respiration monitoring device of the system is used for detecting abdominal respiration deformation. The multi-channel array flexible acquisition sensor has deformability and is used for detecting back muscle electric signals and pressure change signals. The body surface information acquisition device is used for acquiring human chest and abdomen point cloud data. The main controller is used for processing data collected by the abdominal respiration monitoring device, the multi-channel array flexible acquisition sensor and the body surface information acquisition device, and obtaining a predicted position of a moving tumor target area. The application adopts back pressure muscle electric information and abdominal respiration deformation information and binocular vision information respiration movement evaluation, constructs a mapping relationship from external movement to internal tumor movement through external multi-mode data and 4D-CT information, combines a big data model, and realizes internal tumor pose prediction and evaluation.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of detection, in particular to a tumor target area pose real-time tracking and prediction method and system, device and medium. BACKGROUND

[0002] The influence of thoraco-abdominal respiratory motion on the shape and position changes of organs will directly affect the accuracy of surgical positioning and tumor radiotherapy target area. The existing tumor target area monitoring means usually relies on real-time CT irradiation, which can provide real-time monitoring data, but will cause the patient to be exposed to a higher radiation dose, and the invasiveness of the gold standard implant monitoring method increases the risk of complications such as pneumothorax in patients. In order to cope with the interference of respiratory motion, surface-based monitoring and prediction technology is particularly important. The current respiratory motion prediction model is divided into two categories: model algorithm and model-free algorithm. The model algorithm predicts through the time series model, but has the problems of limited prediction time and long calculation time, which makes it difficult to effectively compensate for system delay. The model-free algorithm can automatically adjust the parameters according to the different respiratory states of individuals, adapt to non-strict periodic respiratory motion, and has faster update speed, but its prediction accuracy still needs to be improved. In addition, the traditional surface electromyography (sEMG) sensor and pressure sensor have the problems of poor adhesion, easy damage and poor biocompatibility in application, which limits their effective application in tumor tracking. SUMMARY

[0003] In order to achieve the above-mentioned purposes and other advantages of the present application, the first object of the present application is to provide a tumor target area pose real-time tracking and prediction system, comprising an abdominal respiration monitoring device based on a fiber Bragg grating sensor, a multi-channel array flexible acquisition sensor, a surface information acquisition device based on a binocular structured light, and a main controller, wherein the fiber Bragg grating sensor and the multi-channel array flexible acquisition sensor are wearable sensors.

[0004] The abdominal respiration monitoring device is used to detect abdominal respiration deformation.

[0005] The multi-channel array flexible acquisition sensor has deformability and is used to detect back muscle electrical and pressure change signals.

[0006] The surface information acquisition device is used to acquire human thoraco-abdominal point cloud data.

[0007] The main controller is used to process the data collected by the abdominal respiration monitoring device, the multi-channel array flexible acquisition sensor and the surface information acquisition device to obtain the predicted position of the active tumor target area.

[0008] Further, the abdominal respiration monitoring device comprises a light source, a beam expander, a fiber Bragg grating sensor module, a beam combiner, a spectrum analyzer and a computer data processing system.

[0009] The light emitted by the light source is expanded by the beam expander, and the expanded parallel light passes through the fiber Bragg grating sensor module. After wavelength selection, the fiber Bragg grating sensor module transmits light of different wavelengths. The converged light beam is then transmitted to the spectrum analyzer through the beam combiner to analyze the intensity distribution of light beams of different wavelengths. The computer data processing system performs deformation analysis to obtain results.

[0010] Furthermore, the fiber Bragg grating sensor module includes a fiber Bragg grating sensor and silica gel, and the fiber Bragg grating sensor is embedded in the silica gel.

[0011] Furthermore, the fiber Bragg grating sensor is arranged in a twisted manner, and is bent into two layers, both of which are horizontal S-shaped waves, and the S-shapes of the upper and lower layers are opposite, and the upper and lower layers overlap to form a circle.

[0012] Furthermore, the multi-channel array flexible acquisition sensor includes an electromyographic sensor and a pressure sensor. The electromyographic sensor is used to collect electromyographic signals of the back, and the pressure sensor is used to detect stress changes in the back.

[0013] Furthermore, the electromyographic sensor is prepared by a combination of graphene, silk fibroin, and Ca2+.

[0014] Furthermore, the pressure variation sensor is made of graphene material.

[0015] Furthermore, the body surface information acquisition device adopts structured light technology, and projects specially coded structured light onto the surface of the measured object through a projector to generate texture.

[0016] Furthermore, the main controller fuses the data collected by the abdominal respiratory monitoring device, the multi-channel array flexible acquisition sensor and the body surface information acquisition device through a multimodal data fusion model to form a fusion feature, and uses the fusion feature as the input of the tumor target position prediction model for prediction. The predicted information is matched and transformed with the constructed mapping model containing CT image information to finally obtain the predicted position of the active tumor target area.

[0017] A second object of the present invention is to provide a method for real-time tracking and prediction of tumor target position, based on the above-mentioned system, comprising the following steps:

[0018] A multimodal data fusion model is used to fuse back electromyography and pressure change signals, abdominal respiratory deformation, and human chest and abdomen point cloud data to form fusion features;

[0019] Using the fusion features as input data of a tumor target location prediction model for prediction;

[0020] The predicted information is matched and transformed with the constructed mapping model containing CT image information, and finally the predicted position of the active tumor target area is obtained.

[0021] Furthermore, the multimodal data fusion model includes a data input layer, a feature extraction layer, and a feature fusion layer; wherein,

[0022] The data input layer is used to input back electromyography and pressure change signals, abdominal respiratory deformation, and human chest and abdomen point cloud data into the model;

[0023] The feature extraction layer is used to pass the input data through the corresponding feature extraction module and combine it with the self-attention mechanism module to extract useful features of the data;

[0024] The feature fusion layer is used to fuse the features of different input data using a cross-attention mechanism to obtain more comprehensive and consistent environmental perception information.

[0025] Furthermore, the tumor target location prediction model adopts a time series-based prediction model.

[0026] Furthermore, the method further includes the steps of constructing the mapping model:

[0027] Collect CT imaging data and in vitro detection information before surgery and perform data processing and feature extraction;

[0028] Construct multimodal fusion models and prediction models including CT imaging information;

[0029] A mapping relationship between CT images and external detection information is constructed to obtain a mapping model containing CT image information.

[0030] A third object of the present invention is to provide a computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above method when executing the computer program.

[0031] A fourth object of the present invention is to provide a computer-readable storage medium having a computer program stored thereon, wherein the computer program implements the steps of the above method when executed by a processor.

[0032] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0033] To address the problem of respiratory movement during radiotherapy, the present invention uses back pressure electromyography information, abdominal respiratory deformation information, and binocular vision information to evaluate respiratory movement. Through in vitro multimodal data and 4D-CT information, combined with a big data model, a mapping relationship from in vitro movement to in vivo tumor movement is constructed, thereby realizing the prediction and evaluation of in vivo tumor posture.

[0034] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention and to implement it according to the contents of the description, the following preferred embodiments of the present invention are described in detail with reference to the accompanying drawings. The specific implementation methods of the present invention are given in detail by the following embodiments and the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0036] Figure 1 This is a schematic diagram of the real-time tracking and prediction system for the tumor target area posture;

[0037] Figure 2 This is a schematic diagram of an abdominal respiratory monitoring device;

[0038] Figure 3 Schematic diagram of fiber Bragg grating sensor module;

[0039] Figure 4 Schematic diagram of a multi-channel array flexible acquisition sensor;

[0040] Figure 5 A schematic diagram of the attachment position of the wearable sensor;

[0041] Figure 6 A flowchart for real-time tracking and prediction of tumor target position;

[0042] Figure 7 This is a flow chart of the real-time tracking and prediction method for tumor target area posture;

[0043] Figure 8 A schematic diagram of a computer device;

[0044] Figure 9 A schematic diagram of a computer-readable storage medium. DETAILED DESCRIPTION

[0045] Hereinafter, the present application will be further described with reference to the drawings and specific embodiments. Obviously, the described embodiments are only a part of the embodiments of the present application, and are not all the embodiments. It should be noted that, under the condition of no conflict, the embodiments described below or the technical features between the embodiments can be combined to form new embodiments.

[0046] All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present application without creative labor fall within the scope of the present application.

[0047] The figure numbers in the present application are only used to distinguish the steps in the scheme, and are not used to limit the execution order of the steps. The specific execution order is subject to the description in the specification.

[0048] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present application belongs. The terms used in the specification of the present application are only for the purpose of describing the specific embodiments and are not intended to limit the present application.

[0049] Embodiment 1

[0050] A tumor target area pose real-time tracking and prediction system 1, as shown in Figure 1 includes an abdominal respiration monitoring device 100 based on a fiber Bragg grating sensor, a multi-channel array flexible acquisition sensor 110, a body surface information acquisition device 120 based on a binocular structured light, and a main controller 130. The fiber Bragg grating sensor and the multi-channel array flexible acquisition sensor are both wearable sensors. Wherein,

[0051] The abdominal respiration monitoring device is used to detect abdominal respiration deformation;

[0052] The multi-channel array flexible acquisition sensor has deformability and is used to detect back muscle electrical signals and pressure change signals;

[0053] The body surface information acquisition device is used to acquire human chest and abdomen point cloud data;

[0054] The main controller is used to process the data collected by the abdominal respiration monitoring device, the multi-channel array flexible acquisition sensor, and the body surface information acquisition device, to obtain the predicted position of the active tumor target area.

[0055] In some embodiments, as shown in Figure 2 The abdominal respiration monitoring device includes a light source, a beam expander, a fiber Bragg grating sensor module, a beam combiner, a spectrum analyzer, and a computer data processing system.

[0056] The light emitted by the light source is expanded by the beam expander, and the expanded parallel light passes through a transmission fiber Bragg grating sensor module. After wavelength selection, the fiber Bragg grating sensor module transmits light of different wavelengths. The converged light beam is then transmitted to the spectrum analyzer through the beam combiner to analyze the intensity distribution of light beams of different wavelengths. The computer data processing system performs deformation analysis to obtain results.

[0057] Fiber Bragg Grating (FBG) sensors are a type of fiber optic sensor widely used for respiratory monitoring. They work by creating periodic refractive index modulation in the optical fiber. This allows light of a specific wavelength to continue to be transmitted, and the transmission wavelength is directly related to the strain or temperature change in the optical fiber. When the optical fiber is subjected to mechanical stress or temperature changes, the transmission wavelength changes, thereby enabling the measurement of physical quantities. FBG sensors have the advantages of high sensitivity, small size, and strong resistance to electromagnetic interference. In respiratory monitoring, FBG sensors can be installed in chest straps or other wearable devices to measure respiratory rate by detecting the expansion of the chest or abdomen caused by breathing.

[0058] The chest wall movement caused by breathing is an important indicator for monitoring respiratory rate. Considering that the FBG sensor is poorly attached to the human abdomen, the FBG sensor is embedded in the silicone material through a 3D printing mold. Figure 3 As shown, the fiber Bragg grating sensor module includes a fiber Bragg grating sensor 101 and silicone gel 102, wherein the fiber Bragg grating sensor is embedded in the silicone gel. Furthermore, the fiber Bragg grating sensor is arranged in a twisted pattern, and the fiber Bragg grating sensor is bent into two layers, the upper and lower layers, both of which are horizontal S-shaped waves, and the S-shapes of the upper and lower layers are opposite. The upper and lower layers overlap to form a circle. This can increase the evanescent field around the bent structure without destroying the fiber structure, thereby amplifying the effect of human breathing-induced motion on the sensor. For details, see Figure 3 By measuring the changes in FBG wavelength, the FBG sensor can accurately capture the chest and abdominal movements caused by each breath.

[0059] Considering the poor adhesion of traditional electromyographic sensors and pressure sensors, and the inability of a few electrodes to collect complete electrophysiological activity of the back, the rigid pressure sensor needs to be in contact with the back to detect valid data, resulting in a high sensor waste rate. For the acquisition of electromyographic signals from the back, combined with the positional arrangement of the back muscles, this invention designs a multi-channel array flexible acquisition sensor suitable for back electromyographic and pressure variable signal detection, see Figure 4 .

[0060] Specifically, the multi-channel array flexible acquisition sensor includes an electromyographic sensor and a pressure-variable sensor. The electromyographic sensor is used to collect electromyographic signals of the back, and the pressure-variable sensor is used to detect stress changes in the back.

[0061] Taking into account the deviations in the height and body shape of users, the muscle groups of different users are not exactly the same. The multi-channel array flexible acquisition sensor in the present invention is deformable and can be deformed within a certain range, with better adhesion and biocompatibility. In order to ensure normal operation under the CT environment and avoid affecting the CT image acquisition process, the multi-channel array flexible acquisition sensor provided by the present invention adopts non-magnetic materials and is prepared as follows: the electromyographic sensor is prepared using a combination of graphene / silk fibroin / Ca2+ (Gr / SF / Ca2+), and the silk fibroin material has the property of swelling when it encounters water. After being damaged, only a small amount of water needs to be added to complete self-repair in a short time, thereby improving the utilization rate and service life. The pressure sensor is also made of graphene material. A high signal-to-noise ratio, long-term and stable detection of human physiological signals is achieved.

[0062] The muscle group collection positions in this embodiment are as follows Figure 5 As shown in (a), the final results are the back electromyographic signal diagram and back stress change diagram. Figure 5 Middle (b) is an abdominal respiratory monitoring device based on FBG sensors, which is attached to the abdomen in a distributed manner and uses multiple optical fibers in an integrated manner to better detect respiratory movements.

[0063] The traditional binocular vision ranging system consists only of a binocular camera. By capturing the left and right views of a spatial object, it can quickly calculate the depth information of the object in the image. However, it is easily affected by the texture information of the object being measured. When the texture information is weak, it increases the difficulty of stereo matching and easily leads to matching errors. The present invention adopts structured light technology. A projector projects specially encoded structured light onto the surface of the measured object to generate texture. With the help of a customized texture method, the difficulty of stereo matching is reduced, and the speed and accuracy of measurement are improved. The calibration method uses a motion platform to drive the camera along a known trajectory, while controlling the camera to collect multiple sets of images. The camera parameters are solved by establishing a correspondence between image information and known displacements.

[0064] In some embodiments, as Figure 6 As shown, the main controller fuses the data collected by the abdominal respiratory monitoring device, the multi-channel array flexible acquisition sensor and the body surface information acquisition device through a multimodal data fusion model to form a fusion feature, and uses the fusion feature as the input of the tumor target position prediction model for prediction. The predicted information is matched and transformed with the constructed mapping model containing CT image information, and finally the predicted position of the active tumor target is obtained.

[0065] The multimodal data fusion model fuses the back array pressure change and electromyography monitoring images, the abdominal respiratory deformation based on the FBG sensor, and the human chest and abdomen point cloud based on the body surface information acquisition device based on binocular structured light to form fusion features, specifically including:

[0066] Data input layer: inputs data from multiple sensors into the model.

[0067] Feature extraction layer: In the feature extraction layer, the data of each sensor passes through the corresponding feature extraction module and is combined with the self-attention mechanism module to extract useful features of the sensor data, including image edges and corners, and point cloud features of the human body.

[0068] Feature Fusion Layer: At this layer, a cross-attention mechanism is used to fuse features from different sensors to obtain more comprehensive and consistent environmental perception information. To better align features, the cross-attention mechanism is used to dynamically capture the correlations between multiple modalities.

[0069] Internal tumor target location prediction must account for time delays, requiring tumor location prediction and pre-determined delivery of the target location. The prediction model needs to predict the tumor's location at different times over an entire timeframe (e.g., 0-1000ms), necessitating the use of a time series model for location prediction.

[0070] Currently, the classic method for modeling sequence information is recurrent neural networks (RNNs) and their various variants. However, RNN training is difficult to parallelize, and modeling long sequences faces problems such as vanishing and exploding gradients. This paper uses an informer to predict tumor target locations. This method uses real-time monitoring of chest and abdominal surface point cloud information, back compression deformation, and electromyographic array images. Abdominal respiratory deformation is then fed into a multimodal fusion model, and the fused features are used as input for prediction.

[0071] Then, a multimodal fusion model and prediction model including CT image information are constructed to build a mapping relationship from CT images to external detection information.

[0072] Finally, the information obtained during the prediction is matched and transformed with the constructed mapping model containing CT image information, and the predicted position of the active tumor target area is finally obtained. Among them, the mapping model containing CT image information is constructed before the radiotherapy operation. The collected information includes in vitro detection information and CT images. During the operation, the in vitro detection information is collected and the position of the active target area is predicted by the tumor target area position prediction model. For details, see Figure 6 .

[0073] Example 2

[0074] A tumor target area pose real-time tracking and prediction method, based on the above system, the detailed description of the system can refer to the corresponding description in the above system embodiment, which will not be repeated here. As shown in Figure 6 、 Figure 7 The method comprises the following steps:

[0075] S1, a multi-modal data fusion model is used to fuse the back electromyography and pressure change signal, the abdominal respiratory deformation, and the human chest and abdominal point cloud data to form a fusion feature;

[0076] The multi-modal data fusion model comprises a data input layer, a feature extraction layer, and a feature fusion layer; wherein,

[0077] The data input layer is used to input the back electromyography and pressure change signal, the abdominal respiratory deformation, and the human chest and abdominal point cloud data into the model;

[0078] The feature extraction layer is used to extract useful features of the input data through corresponding feature extraction modules combined with a self-attention mechanism module, including edges, corner points of images, point cloud features of the human body, etc.

[0079] The feature fusion layer is used to fuse the features of different input data by using a cross-attention mechanism to obtain more comprehensive and consistent environmental perception information. In order to better align the features, the cross-attention mechanism is used to dynamically capture the correlation between multiple modalities.

[0080] S2, the fusion feature is used as input data of a tumor target area position prediction model for prediction;

[0081] The internal tumor target area position prediction needs to consider the time delay, and the tumor position needs to be predicted and the target position needs to be issued in advance. The prediction model needs to predict the position of the tumor at different times in the future (such as 0-1000 ms), so a time series model needs to be used for position prediction, that is, the tumor target area position prediction model uses a prediction model based on time series.

[0082] At present, the classical method for modeling sequence information is recurrent neural network (RNN) and its various variants, but the training of RNN is difficult to parallelize, and there are problems such as gradient disappearance and gradient explosion in modeling long sequences. The present application uses Informer to predict the position of the tumor target area, through real-time monitoring of the surface point cloud information of the human chest and abdomen, the back pressure change and electromyography array image, and the abdominal respiratory deformation, the fusion feature is used as input data for prediction after multi-modal fusion model.

[0083] S3, the predicted information is matched with the mapping model containing CT image information, and the relationship is transformed, and finally the predicted position of the moving tumor target area is obtained.

[0084] In some embodiments, the method further comprises a step of constructing the mapping model:

[0085] The CT image data and the external detection information are collected before the surgery, and data processing and feature extraction are performed, as described in detail in Figure 6 ;

[0086] A multi-modal fusion model and a prediction model are constructed, which contain CT image information.

[0087] A mapping relationship between the CT image and the external detection information is constructed, and a mapping model containing CT image information is obtained.

[0088] The mapping model containing CT image information is constructed before the radiotherapy surgery, and the external detection information is collected during the surgery implementation stage, and the position of the moving target area is predicted by the tumor target area position prediction model, as described in detail in Figure 6 .

[0089] Embodiment 3

[0090] A computer device 200, as shown in Figure 8 , includes a memory 210, a processor 220, and a computer program 230 stored in the memory and executable on the processor, and the processor executes the computer program to implement the steps of a tumor target area pose real-time tracking and prediction method. For detailed description of the method, please refer to the corresponding description in the above method embodiment, which will not be repeated here.

[0091] Embodiment 4

[0092] A computer readable storage medium, as shown in Figure 9 , has a computer program stored thereon, and the computer program is executed by a processor to implement the steps of a tumor target area pose real-time tracking and prediction method. For detailed description of the method, please refer to the corresponding description in the above method embodiment, which will not be repeated here.

[0093] Embodiment 5

[0094] A computer program product, the computer program product comprising a computer program, the computer program being executed by a processor to implement the steps of a tumor target area pose real-time tracking and prediction method. For detailed description of the method, please refer to the corresponding description in the above method embodiment, which will not be repeated here.

[0095] The number of devices and the scale of processing described herein are used to simplify the description of the present application. Applications, modifications and variations of the present application are obvious to those skilled in the art.

[0096] Although the embodiments of the present invention have been disclosed above, they are not limited to the applications listed in the description and implementation methods. They can be fully applied to various fields suitable for the present invention. For those familiar with the art, additional modifications can be easily implemented. Therefore, without departing from the general concept defined by the claims and the scope of equivalents, the present invention is not limited to the specific details and illustrations shown and described herein.

[0097] The apparatus, computer device, non-volatile computer storage medium, and method provided in the embodiments of this specification correspond to each other. Therefore, the apparatus, computer device, and non-volatile computer storage medium also have similar beneficial technical effects as the corresponding method. Since the beneficial technical effects of the method have been described in detail above, the beneficial technical effects of the corresponding apparatus, computer device, and non-volatile computer storage medium will not be repeated here.

[0098] Those skilled in the art will also appreciate that, in addition to implementing the controller in pure computer-readable program code, it is entirely possible to implement the same functionality by programming the method steps logically, such as through logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers. Therefore, such a controller can be considered a hardware component, and the devices included therein for implementing various functions can also be considered structures within the hardware component. Alternatively, the devices for implementing various functions can be considered both software units implementing the method and structures within the hardware component.

[0099] The systems, devices, or units described in the above embodiments can be implemented by computer chips or physical devices, or by products with certain functions. For ease of description, the above devices are described separately by function, with each unit described separately. Of course, when implementing one or more embodiments of this specification, the functions of each unit can be implemented in the same or multiple software and / or hardware components.

[0100] Those skilled in the art will appreciate that the embodiments of this specification may be provided as methods, systems, or computer program products. Therefore, the embodiments of this specification may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the embodiments of this specification may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0101] This specification is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of this specification. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0102] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0103] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 The steps for the function specified in one or more boxes.

[0104] It should also be noted that the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, commodity, or apparatus that includes a series of elements includes not only those elements but also other elements not explicitly listed, or includes elements inherent to such process, method, commodity, or apparatus. In the absence of further limitations, an element defined by the phrase "comprises a ..." does not exclude the presence of other identical elements in the process, method, commodity, or apparatus that includes the element.

[0105] This specification may be described in the general context of computer-executable instructions executed by a computer, such as program units. Generally, program units include routines, programs, objects, components, data structures, and the like that perform specific tasks or implement specific abstract data types. The specification may also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communications network. In a distributed computing environment, program units may be located in local and remote computer storage media, including storage devices.

[0106] The various embodiments in this specification are described in a progressive manner. Similar parts between the various embodiments can be referred to in conjunction with each other. Each embodiment focuses on the differences between the other embodiments. In particular, the system embodiments are generally similar to the method embodiments, so the description is relatively simple. For relevant parts, refer to the description of the method embodiments.

[0107] The foregoing is merely an example of the present invention and is not intended to limit the present invention to one or more embodiments. It will be apparent to those skilled in the art that various modifications and variations may be made to the present invention to one or more embodiments. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention to one or more embodiments shall be included within the scope of the claims of the present invention to one or more embodiments.

Claims

1. A real-time tracking and prediction system for tumor target position, characterized by: It includes an abdominal respiratory monitoring device based on a fiber Bragg grating sensor, a multi-channel array flexible acquisition sensor, a body surface information acquisition device based on binocular structured light, and a main controller. The fiber Bragg grating sensor and the multi-channel array flexible acquisition sensor are both wearable sensors; wherein, The abdominal respiratory monitoring device is used to detect abdominal respiratory deformation; The multi-channel array flexible acquisition sensor has deformability and is used to detect back electromyography and pressure change signals; The body surface information acquisition device is used to collect point cloud data of the chest and abdomen of the human body; The main controller is used to process the data collected by the abdominal respiratory monitoring device, the multi-channel array flexible acquisition sensor and the body surface information acquisition device to obtain the predicted position of the active tumor target area; The fiber Bragg grating sensor is arranged in a twisted pattern and is bent into two layers, upper and lower layers. Both layers are horizontal S-shaped, and the S-shapes of the upper and lower layers are opposite. The upper and lower layers overlap to form a circle.

2. The real-time tracking and prediction system for tumor target position according to claim 1, characterized in that: The abdominal respiratory monitoring device includes a light source, a beam expander, a fiber Bragg grating sensor module, a beam combiner, a spectrum analyzer, and a computer data processing system; The light emitted by the light source is expanded by the beam expander, and the expanded parallel light passes through the fiber Bragg grating sensor module. After wavelength selection, the fiber Bragg grating sensor module transmits light of different wavelengths. The converged light beam is then transmitted to the spectrum analyzer through the beam combiner to analyze the intensity distribution of light beams of different wavelengths. The computer data processing system performs deformation analysis to obtain results.

3. The real-time tracking and prediction system for tumor target position according to claim 2, characterized in that: The fiber Bragg grating sensor module comprises a fiber Bragg grating sensor and silica gel, wherein the fiber Bragg grating sensor is embedded in the silica gel.

4. The real-time tracking and prediction system for tumor target position according to claim 1, characterized in that: The multi-channel array flexible acquisition sensor includes an electromyographic sensor and a pressure-variable sensor. The electromyographic sensor is used to collect electromyographic signals of the back, and the pressure-variable sensor is used to detect stress changes in the back.

5. The real-time tracking and prediction system for tumor target position according to claim 4, characterized in that: The electromyographic sensor is prepared by combining graphene, silk fibroin and Ca2+.

6. The real-time tracking and prediction system for tumor target position according to claim 4, characterized in that: The pressure variation sensor is made of graphene material.

7. The real-time tracking and prediction system for tumor target position according to claim 1, characterized in that: The body surface information acquisition device adopts structured light technology, and projects specially coded structured light onto the surface of the measured object through a projector to generate texture.

8. The real-time tracking and prediction system for tumor target position according to claim 1, characterized in that: The main controller fuses the data collected by the abdominal respiratory monitoring device, the multi-channel array flexible acquisition sensor and the body surface information acquisition device through a multimodal data fusion model to form a fusion feature, uses the fusion feature as the input of the tumor target position prediction model for prediction, matches and transforms the predicted information with the constructed mapping model containing CT image information, and finally obtains the predicted position of the active tumor target area.

9. A method for real-time tracking and prediction of tumor target position, based on the system according to any one of claims 1 to 8, characterized in that: The following steps are involved: A multimodal data fusion model is used to fuse back electromyography and pressure change signals, abdominal respiratory deformation, and human chest and abdomen point cloud data to form fusion features; Using the fusion features as input data of a tumor target location prediction model for prediction; The predicted information is matched and transformed with the constructed mapping model containing CT image information, and finally the predicted position of the active tumor target area is obtained.

10. The method for real-time tracking and prediction of tumor target position according to claim 9, characterized in that: The multimodal data fusion model includes a data input layer, a feature extraction layer, and a feature fusion layer; wherein, The data input layer is used to input back electromyography and pressure change signals, abdominal respiratory deformation, and human chest and abdomen point cloud data into the model; The feature extraction layer is used to pass the input data through the corresponding feature extraction module and combine it with the self-attention mechanism module to extract useful features of the data; The feature fusion layer is used to fuse the features of different input data using a cross-attention mechanism to obtain more comprehensive and consistent environmental perception information.

11. The method for real-time tracking and prediction of tumor target position according to claim 9, characterized in that: The tumor target area position prediction model adopts a time series-based prediction model.

12. The method for real-time tracking and prediction of tumor target position according to claim 9, wherein: The steps for constructing the mapping model are also included: Collect CT imaging data and in vitro detection information before surgery and perform data processing and feature extraction; Construct multimodal fusion models and prediction models including CT imaging information; A mapping relationship between CT images and external detection information is constructed to obtain a mapping model containing CT image information.

13. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 9 to 12 are implemented.

14. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 9 to 12 are implemented.

Images