Hybrid-driven cardiovascular focus motion trail prediction method and system

Abstract

The invention relates to a hybrid-driven cardiovascular focus motion trail prediction method and system. The method comprises the steps that current motion trail parameters of a cardiovascular focus area of a patient are acquired; inputting the motion trail parameters into a motion trail prediction model to obtain predicted trail data of the cardiovascular lesion area at future moments; wherein the motion track prediction model is obtained by training a target machine learning model based on sample training data; the sample training data comprises historical motion trail data and / or a three-dimensional reconstruction model of the cardiovascular lesion area of the patient in different physiological states and pathological states collected at different time points. According to the scheme disclosed by the invention, accurate prediction of the motion trail of the cardiovascular target region can be realized, so that the radiotherapy diagnosis and treatment instrument adopting the scheme can realize an accurate radiotherapy treatment effect on a patient.

Description

Technical Field

[0001] This disclosure relates to the field of assisted medical automation technology, and in particular to a hybrid-driven method and system for predicting the motion trajectory of cardiovascular lesions. Background Technology

[0002] The development of modern medicine has made it possible for mankind to control and conquer cancer. In the clinical treatment of tumors, surgery, radiotherapy and chemotherapy are the three main treatment methods. Radiotherapy has a wide range of indications and is more selective, so most patients with malignant tumors need to receive radiotherapy at some stage of their treatment.

[0003] The goal of radiotherapy is to concentrate the radiation dose accurately and to the lesion area (target area) to kill tumor cells. However, the cardiovascular target area differs from other solid, static lesions; it is dynamic and constantly changing. Currently, there is a lack of technologies to predict the movement trajectory of cardiovascular lesions, which means that the precision of radiotherapy treatment for patients needs to be improved. Summary of the Invention

[0004] To solve the above-mentioned technical problems, or at least partially solve them, this disclosure provides a method and system for predicting the motion trajectory of cardiovascular lesions.

[0005] In a first aspect, embodiments of this disclosure provide a method for predicting the motion trajectory of cardiovascular lesions, the method comprising:

[0006] Obtain the current motion trajectory parameters of the patient's cardiovascular lesion area;

[0007] The motion trajectory parameters are input into the motion trajectory prediction model to obtain the predicted trajectory data of the cardiovascular lesion area at future times;

[0008] The motion trajectory prediction model is obtained by training the target machine learning model based on sample training data; the sample training data includes historical motion trajectory data of cardiovascular lesion areas and / or three-dimensional reconstruction models collected at different time points under different physiological and pathological conditions of patients.

[0009] In one embodiment, the target machine learning model includes at least a Transformer model; the sample training data also includes labeled data of historical motion trajectory data of the cardiovascular lesion area, the labeled data including discrete target point data of the cardiovascular lesion area.

[0010] In one embodiment, the method further includes:

[0011] Acquire one or more of the patient's cardiovascular system via ultrasound, X-ray, and CT images;

[0012] Based on the ultrasound images, X-ray images, and CT images, a three-dimensional reconstruction model of the cardiovascular lesion area under different physiological and pathological conditions of the patient is generated.

[0013] In one embodiment, generating a three-dimensional reconstruction model of the cardiovascular lesion area under different physiological and pathological states of the patient based on the ultrasound image, X-ray image, and CT image includes:

[0014] Image feature information is extracted from the ultrasound image, X-ray image and CT image, and a three-dimensional reconstruction model is generated based on the image feature information.

[0015] In one embodiment, the method further includes:

[0016] Multiple ultrasound images, multiple X-ray images, and multiple CT images of the patient's cardiovascular system were acquired at different time points;

[0017] The historical motion trajectory data of the cardiovascular lesion area is determined based on the multiple ultrasound images, multiple X-ray images, and multiple CT images.

[0018] In one embodiment, determining the historical motion trajectory data of the cardiovascular lesion area based on the plurality of ultrasound images, plurality of X-ray images, and plurality of CT images includes:

[0019] The first historical motion trajectory data of the cardiovascular lesion area is determined based on the multiple ultrasound images; the second historical motion trajectory data of the cardiovascular lesion area is determined based on the multiple X-ray images; the third historical motion trajectory data of the cardiovascular lesion area is determined based on the multiple CT images; and the historical motion trajectory data of the cardiovascular lesion area is determined by combining the first historical motion trajectory data, the second historical motion trajectory data, and the third historical motion trajectory data.

[0020] In one embodiment, the different physiological states of the patient are determined by different physiological parameters of the patient's body and / or different pathological changes in organs or tissues when the body is ill; the different pathological states of the patient are determined by different pathological changes in organs or tissues when the patient is ill.

[0021] Secondly, embodiments of this disclosure provide a cardiovascular lesion motion trajectory prediction system, comprising:

[0022] The data acquisition module is used to acquire the current motion trajectory parameters of the patient's cardiovascular lesion area;

[0023] The trajectory prediction module is used to input the motion trajectory parameters into the motion trajectory prediction model to obtain the predicted trajectory data of the cardiovascular lesion area at future times; wherein, the motion trajectory prediction model is obtained by training the target machine learning model based on sample training data; the sample training data includes historical motion trajectory data of the cardiovascular lesion area and / or three-dimensional reconstruction model collected at different time points under different physiological and pathological states of the patient.

[0024] Thirdly, embodiments of this disclosure provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the cardiovascular lesion motion trajectory prediction method described in any of the above embodiments.

[0025] Fourthly, embodiments of this disclosure provide an electronic device, including:

[0026] Processor; and

[0027] Memory, used to store computer programs;

[0028] The processor is configured to execute the cardiovascular lesion motion trajectory prediction method described in any of the above embodiments by executing the computer program.

[0029] The technical solution provided in this disclosure has the following advantages compared with the prior art:

[0030] This disclosure provides a method and system for predicting the motion trajectory of cardiovascular lesions. First, the current motion trajectory parameters of the patient's cardiovascular lesion area are obtained. These parameters are then input into a motion trajectory prediction model to obtain predicted trajectory data for the cardiovascular lesion area at future times. The motion trajectory prediction model is trained on a target machine learning model using sample training data. The sample training data includes historical motion trajectory data of the cardiovascular lesion area collected at different time points under different physiological and pathological states of the patient, and / or a three-dimensional reconstruction model. This embodiment fills the gap in predicting the motion trajectory of cardiovascular target areas. Addressing the difficulty in accurately predicting the motion trajectory of cardiovascular target areas, it collects historical motion trajectory data of cardiovascular lesions under different physiological and pathological states at different time points, and combines this data with a three-dimensional reconstruction model as training data to obtain a motion trajectory prediction model. This enables accurate prediction of the motion trajectory of cardiovascular target areas, allowing radiotherapy instruments using this method to achieve precise radiotherapy treatment effects for patients. Attached Figure Description

[0031] The accompanying drawings, which are incorporated in and form a part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure.

[0032] To more clearly illustrate the technical solutions in the embodiments of this disclosure or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0033] Figure 1 This is a flowchart of the cardiovascular lesion motion trajectory prediction method according to an embodiment of the present disclosure;

[0034] Figure 2 This is a flowchart of a method for predicting the motion trajectory of cardiovascular lesions according to another embodiment of this disclosure;

[0035] Figure 3 This is a diagram showing the results of a cardiovascular lesion motion trajectory prediction test according to an embodiment of this disclosure;

[0036] Figure 4 This is a schematic diagram of a hybrid-driven cardiovascular lesion motion trajectory prediction system according to an embodiment of the present disclosure;

[0037] Figure 5 This is a schematic diagram of an electronic device according to an embodiment of the present disclosure. Detailed Implementation

[0038] To better understand the above-mentioned objectives, features, and advantages of this disclosure, the solutions disclosed herein will be further described below. It should be noted that, unless otherwise specified, the embodiments and features described herein can be combined with each other.

[0039] Numerous specific details are set forth in the following description in order to provide a full understanding of this disclosure, but this disclosure may also be implemented in other ways different from those described herein; obviously, the embodiments in the specification are only some, and not all, of the embodiments of this disclosure.

[0040] It should be understood that in the following text, "at least one item" refers to one or more items, and "more than one item" refers to two or more items. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can mean: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one of a, b, or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0041] Unlike other static solid lesions, cardiovascular target areas are affected by factors such as respiration, heartbeat, and the patient's physiological and pathological conditions. Their movement trajectories are highly dynamic and uncertain, with sparsely marked target points, making it difficult to accurately predict the movement trajectory of cardiovascular lesions. This hinders the improvement of precise radiotherapy treatment effects for patients using radiotherapy instruments. Currently, the industry lacks relevant solutions, representing a gap in the field.

[0042] Figure 1 This is a flowchart of a method for predicting the motion trajectory of cardiovascular lesions according to an embodiment of the present disclosure. The method can be executed by a computer device and may specifically include the following steps:

[0043] Step S101: Obtain the current motion trajectory parameters of the patient's cardiovascular lesion area.

[0044] For example, the motion trajectory parameters can be data on the change in amplitude of the pulsation of the cardiovascular lesion area, i.e., the target area, over time. In this embodiment, the motion trajectory parameters of the patient's cardiovascular lesion area at the current time t1 can be obtained. Specifically, as an example, the motion trajectory parameters of the patient's cardiovascular lesion area can be obtained by, for example, existing non-invasive impedance measurement methods. For details, please refer to the existing technology for understanding, which will not be elaborated here. Of course, the acquisition method is not limited to this.

[0045] Step S102: Input the motion trajectory parameters into the motion trajectory prediction model to obtain the predicted trajectory data of the cardiovascular lesion area at future times; wherein, the motion trajectory prediction model is obtained by training the target machine learning model based on sample training data; the sample training data includes historical motion trajectory data of the cardiovascular lesion area and / or three-dimensional reconstruction model collected at different time points under different physiological and pathological states of the patient.

[0046] For example, in one embodiment, the different physiological states of the patient are determined by different physiological parameters of the patient's body and / or different pathological changes in organs or tissues when the body is ill; the different pathological states of the patient are determined by different pathological changes in organs or tissues when the patient is ill.

[0047] For example, physiological parameters may include, but are not limited to, heart rate, respiratory rate, blood pressure, pulse, body temperature, electrocardiogram, and blood oxygen saturation. These parameters can be acquired by relevant medical sensors or devices. A physiological state refers to a clinical manifestation that occurs in the human body under conditions such as strenuous exercise, mental stress, or hunger. A pathological state refers to a clinical manifestation that occurs in the human body when it is ill. For example, different diseases such as stomach disease, liver disease, and diabetes will result in corresponding pathological states. The appearance of a pathological state often indicates that certain organs or tissues in the body have undergone pathological changes, requiring timely treatment. Information on different pathological changes in organs or tissues can be provided in writing by a doctor after diagnosis and treatment, and can be recorded, for example, in an electronic medical record for easy access later.

[0048] In this embodiment, historical motion trajectory data of the cardiovascular lesion area and a three-dimensional (3D) reconstruction model of the cardiovascular lesion area are collected in advance at different time points under different physiological and pathological states of the patient. The three-dimensional reconstruction model contains the motion trajectory change characteristics of the cardiovascular lesion area. The historical motion trajectory data can be regarded as time series data, for example, represented as TX={(T1,X1),(T2,X2),…,(T n ,X n )}, where {T i} represents sampling time i, {X i} represents the motion trajectory parameters of the lesion area at time i. Time series data is used to describe the motion trajectory of the cardiovascular lesion area over time, reflecting the state or degree of change of the cardiovascular lesion area over time, thereby revealing its regularity, such as the change pattern of cardiovascular lesion or target area motion (amplitude, period, etc.) over time. Based on the historical motion trajectory data of the cardiovascular lesion area under different physiological and pathological states of patients at different time points and / or three-dimensional reconstruction models, sample training data is constructed. Based on the sample training data, the target machine learning model, such as a deep convolutional neural network model, is trained to obtain the motion trajectory prediction model. Then, in application, the motion trajectory parameters obtained in step S101 are input into the motion trajectory prediction model to obtain the predicted trajectory data of the cardiovascular lesion area at future times, such as time t2.

[0049] The solution in this embodiment fills the gap in the prediction of cardiovascular target area motion trajectory. Addressing the difficulty in accurately predicting cardiovascular target area motion trajectory, it collects historical motion trajectory data of cardiovascular lesions under different physiological and pathological conditions at different time points. This data, combined with a 3D reconstruction model, is used as training data to train a motion trajectory prediction model. During training, the model utilizes specific, richer, and more comprehensive sample training data to learn the rich and comprehensive changing patterns of cardiovascular target area motion trajectory. This allows the trained motion trajectory prediction model to accurately predict the motion trajectory of cardiovascular target areas, enabling radiotherapy instruments using this solution to achieve precise radiotherapy treatment effects for patients.

[0050] To further and more accurately predict the motion trajectory of cardiovascular lesions, in one embodiment, the target machine learning model includes at least a Transformer model; the sample training data also includes labeled data of historical motion trajectory data of the cardiovascular lesion area, the labeled data including discrete target point data of the cardiovascular lesion area.

[0051] It should be noted that the historical motion trajectory data in the sample training data can be regarded as time series data. Therefore, the Transformer model is specifically used as the target machine learning model for training. In related technologies, neural network models used for time series data prediction are usually based on Long Short-Term Memory Networks (LSTM) and Gated Recurrent Unit Networks (GRU). Among them, GRU is a type of recurrent neural network (RNN). Like LSTM, it suffers from problems such as long memory and gradient vanishing and gradient exploding during backpropagation when dealing with time series data, resulting in poor prediction performance. The Transformer model has a self-attention network. In this embodiment, the motion trajectory prediction model is trained using the Transformer model, which can solve the problems of long memory and gradient vanishing and gradient exploding during backpropagation, resulting in relatively optimal prediction performance. That is, it can achieve more accurate prediction of the motion trajectory of the cardiovascular target area, thereby enabling radiotherapy diagnostic instruments using this approach to achieve precise radiotherapy treatment effects for patients.

[0052] Furthermore, the sparse annotations of target points on the cardiovascular target area's motion trajectory reduce the accuracy of predicting the motion trajectory of cardiovascular lesions. Therefore, in this embodiment, the sample training data may further include annotated data of historical motion trajectory data for the cardiovascular lesion area, i.e., discrete target point data for the cardiovascular lesion area. This allows the model to learn the annotated target point information of the cardiovascular target area's motion trajectory, thereby enabling the trained motion trajectory prediction model to more accurately predict the motion trajectory of cardiovascular lesions.

[0053] Based on any of the above embodiments, in one embodiment, reference is made to Figure 2 As shown, the method may further include the following steps:

[0054] Step S201: Acquire one or more of the patient's cardiovascular ultrasound images, X-ray images, and CT images.

[0055] For example, ultrasound images, X-ray images, and CT images can be obtained in advance through relevant medical equipment, and the electronic data of the relevant images can be stored in a database. When in use, ultrasound images, X-ray images, and CT images of the patient's cardiovascular lesion area can be obtained from the database.

[0056] Step S202: Generate a three-dimensional reconstruction model of the cardiovascular lesion area under different physiological and pathological conditions of the patient based on the ultrasound image, X-ray image and CT image.

[0057] For example, in one embodiment, image feature information is extracted from the ultrasound image, X-ray image and CT image, and a three-dimensional reconstruction model is generated based on the image feature information.

[0058] Since ultrasound images, X-ray images, and CT images typically contain different feature information for the same detection object, such as the cardiovascular system, this embodiment, after acquiring ultrasound, X-ray, and CT images of the patient's cardiovascular lesion area, extracts cardiovascular image feature information from each of the three images—that is, three different cardiovascular image feature information—and then fuses this information to generate a three-dimensional reconstruction model of the cardiovascular lesion area. The specific reconstruction process of the three-dimensional model can refer to existing three-dimensional reconstruction techniques and will not be elaborated here. This yields a relatively accurate three-dimensional reconstruction model of the cardiovascular lesion area, making the variation patterns of the cardiovascular target area's motion trajectory contained in the sample training data built based on this three-dimensional reconstruction model richer and more accurate. Therefore, the model can learn from these rich and accurate variation patterns of the cardiovascular target area's motion trajectory, enabling the trained motion trajectory prediction model to accurately predict the motion trajectory of the cardiovascular target area.

[0059] Based on any of the above embodiments, in one embodiment, the method may further include the following steps: acquiring multiple ultrasound images, multiple X-ray images and multiple CT images of the patient's cardiovascular system at different time points; and determining the historical motion trajectory data of the cardiovascular lesion area based on the multiple ultrasound images, multiple X-ray images and multiple CT images.

[0060] For example, in one embodiment, determining the historical motion trajectory data of the cardiovascular lesion area based on the plurality of ultrasound images, plurality of X-ray images, and plurality of CT images may specifically include the following sub-steps: determining first historical motion trajectory data of the cardiovascular lesion area based on the plurality of ultrasound images; determining second historical motion trajectory data of the cardiovascular lesion area based on the plurality of X-ray images; determining third historical motion trajectory data of the cardiovascular lesion area based on the plurality of CT images; and comprehensively determining the historical motion trajectory data of the cardiovascular lesion area based on the first historical motion trajectory data, the second historical motion trajectory data, and the third historical motion trajectory data.

[0061] For example, multiple ultrasound images, multiple X-ray images, and multiple CT images can be considered as three time-series image datasets. Each time-series image dataset includes different cardiovascular feature information and hidden motion trajectory change features. The first historical motion trajectory data of the cardiovascular lesion area is determined based on the multiple ultrasound images, i.e., the first time-series image dataset. The second historical motion trajectory data of the cardiovascular lesion area is determined based on the multiple X-ray images, i.e., the second time-series image dataset. The third historical motion trajectory data of the cardiovascular lesion area is determined based on the multiple CT images, i.e., the third time-series image dataset. Finally, the historical motion trajectory data of the cardiovascular lesion area is determined by comprehensively fusing the first, second, and third historical motion trajectory data as sample training data. This makes the sample training data constructed based on this historical motion trajectory data contain richer and more accurate features of the cardiovascular target area motion trajectory change patterns. Therefore, the model can learn and extract these rich and accurate features of the cardiovascular target area motion trajectory change patterns, enabling the trained motion trajectory prediction model to accurately predict the motion trajectory of the cardiovascular target area.

[0062] Accurate modeling of cardiovascular target area motion trajectory is fundamental for real-time tracking and precise radiotherapy using diagnostic instruments. Unlike other static lesions, cardiovascular target areas are influenced by factors such as respiration, heartbeat, and the patient's physiological and pathological conditions, resulting in highly dynamic and uncertain motion trajectories. In a specific example, considering the sparse and dynamic nature of manually annotated discrete target points, based on the aforementioned historical motion trajectory data of the cardiovascular target area, a limited amount of labeled data, and a 3D reconstruction model, a Transformer model is employed. Through data-driven and model-driven training, high-precision modeling and trajectory prediction of the cardiovascular target area's motion trajectory can be achieved. An example of trajectory prediction results is shown below. Figure 3 As shown, the actual motion amplitude change curve and the predicted motion amplitude change curve of the cardiovascular target area are quite similar, indicating high accuracy in trajectory prediction.

[0063] It should be noted that although the steps of the method in this disclosure are described in a specific order in the accompanying drawings, this does not require or imply that these steps must be performed in that specific order, or that all the steps shown must be performed to achieve the desired result. Additional or alternative steps may be omitted, multiple steps may be combined into one step, and / or a step may be broken down into multiple steps. Furthermore, it is readily understood that these steps may be executed synchronously or asynchronously, for example, in multiple modules / processes / threads.

[0064] like Figure 4 As shown, this disclosure provides a cardiovascular lesion motion trajectory prediction system, including:

[0065] Data acquisition module 401 is used to acquire the current motion trajectory parameters of the patient's cardiovascular lesion area;

[0066] The trajectory prediction module 402 is used to input the motion trajectory parameters into the motion trajectory prediction model to obtain the predicted trajectory data of the cardiovascular lesion area at future times; wherein, the motion trajectory prediction model is obtained by training the target machine learning model based on sample training data; the sample training data includes historical motion trajectory data of the cardiovascular lesion area and / or three-dimensional reconstruction model collected at different time points under different physiological and pathological states of the patient.

[0067] In one embodiment, the target machine learning model includes at least a Transformer model; the sample training data also includes labeled data of historical motion trajectory data of the cardiovascular lesion area, the labeled data including discrete target point data of the cardiovascular lesion area.

[0068] In one embodiment, the system may further include a model building module for: acquiring one or more of ultrasound images, X-ray images, and CT images of the patient's cardiovascular system; and generating a three-dimensional reconstruction model of the cardiovascular lesion area under different physiological and pathological conditions of the patient based on the ultrasound images, X-ray images, and CT images.

[0069] In one embodiment, the model building module generates a three-dimensional reconstruction model of the cardiovascular lesion area under different physiological and pathological conditions of the patient based on the ultrasound image, X-ray image and CT image. Specifically, it may include: extracting image feature information from the ultrasound image, X-ray image and CT image, and generating a three-dimensional reconstruction model based on the image feature information.

[0070] In one embodiment, the data acquisition module can also be used to: acquire multiple ultrasound images, multiple X-ray images and multiple CT images of the patient's cardiovascular system at different time points; and determine the historical motion trajectory data of the cardiovascular lesion area based on the multiple ultrasound images, multiple X-ray images and multiple CT images.

[0071] In one embodiment, the data acquisition module determines the historical motion trajectory data of the cardiovascular lesion area based on the plurality of ultrasound images, plurality of X-ray images, and plurality of CT images. Specifically, this may include: determining first historical motion trajectory data of the cardiovascular lesion area based on the plurality of ultrasound images; determining second historical motion trajectory data of the cardiovascular lesion area based on the plurality of X-ray images; determining third historical motion trajectory data of the cardiovascular lesion area based on the plurality of CT images; and comprehensively determining the historical motion trajectory data of the cardiovascular lesion area based on the first historical motion trajectory data, the second historical motion trajectory data, and the third historical motion trajectory data.

[0072] In one embodiment, the different physiological states of the patient are determined by different physiological parameters of the patient's body and / or different pathological changes in organs or tissues when the body is ill; the different pathological states of the patient are determined by different pathological changes in organs or tissues when the patient is ill.

[0073] Regarding the system in the above embodiments, the specific methods by which each module performs operations and the corresponding technical effects have been described in detail in the embodiments related to the method, and will not be elaborated here.

[0074] It should be noted that although several modules or units for the device used to perform actions have been mentioned in the detailed description above, this division is not mandatory. In fact, according to embodiments of this disclosure, the features and functions of two or more modules or units described above can be embodied in one module or unit. Conversely, the features and functions of one module or unit described above can be further divided and embodied by multiple modules or units. Components shown as modules or units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of the disclosed solution according to actual needs. Those skilled in the art can understand and implement this without any inventive effort.

[0075] This disclosure also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the cardiovascular lesion motion trajectory prediction method described in any of the above embodiments.

[0076] For example, the readable storage medium may be, but is not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of readable storage media (a non-exhaustive list) include: electrical connections having one or more wires, portable disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0077] The computer-readable storage medium may include data signals propagated in baseband or as part of a carrier wave, carrying readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. The readable storage medium may also be any readable medium other than a readable storage medium, capable of transmitting, propagating, or transmitting programs for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the readable storage medium may be transmitted using any suitable medium, including but not limited to wireless, wired, optical fiber, RF, etc., or any suitable combination thereof.

[0078] This disclosure also provides an electronic device, including a processor and a memory, the memory being used to store a computer program. The processor is configured to execute the cardiovascular lesion motion trajectory prediction method of any of the above embodiments by executing the computer program.

[0079] The following reference Figure 5To describe an electronic device 600 according to this embodiment of the present invention. Figure 5 The electronic device 600 shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of the present invention.

[0080] like Figure 5 As shown, the electronic device 600 is presented in the form of a general-purpose computing device. The components of the electronic device 600 may include, but are not limited to: at least one processing unit 610, at least one storage unit 620, a bus 630 connecting different system components (including storage unit 620 and processing unit 610), a display unit 640, etc.

[0081] The storage unit stores program code that can be executed by the processing unit 610, causing the processing unit 610 to perform the steps described in the above-described method embodiment section of this specification according to various exemplary embodiments of the present invention. For example, the processing unit 610 can perform, as follows: Figure 1 The steps of the method shown are as follows.

[0082] The storage unit 620 may include a readable medium in the form of a volatile storage unit, such as a random access memory unit (RAM) 6201 and / or a cache storage unit 6202, and may further include a read-only memory unit (ROM) 6203.

[0083] The storage unit 620 may also include a program / utility 6204 having a set (at least one) program module 6205, such program module 6205 including but not limited to: an operating system, one or more application programs, other program modules and program data, each or some combination of these examples may include an implementation of a network environment.

[0084] Bus 630 can represent one or more of several types of bus structures, including a memory cell bus or memory cell controller, a peripheral bus, a graphics acceleration port, a processing unit, or a local bus using any of the various bus structures.

[0085] Electronic device 600 can also communicate with one or more external devices 700 (e.g., keyboard, pointing device, Bluetooth device, etc.), and with one or more devices that enable a user to interact with electronic device 600, and / or with any device that enables electronic device 600 to communicate with one or more other computing devices (e.g., router, modem, etc.). This communication can be performed via input / output (I / O) interface 650. Furthermore, electronic device 600 can also communicate with one or more networks (e.g., local area network (LAN), wide area network (WAN), and / or public networks, such as the Internet) via network adapter 660. Network adapter 660 can communicate with other modules of electronic device 600 via bus 630. It should be understood that, although not shown in the figures, other hardware and / or software modules can be used in conjunction with electronic device 600, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0086] From the above description of the embodiments, those skilled in the art will readily understand that the exemplary embodiments described herein can be implemented by software or by combining software with necessary hardware. Therefore, the technical solutions according to the embodiments of this disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, external hard drive, etc.) or on a network, including several instructions to cause a computing device (such as a personal computer, server, or network device, etc.) to execute the method steps of the above embodiments according to the present disclosure.

[0087] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0088] The above description is merely a specific embodiment of this disclosure, enabling those skilled in the art to understand or implement it. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this disclosure. Therefore, this disclosure is not to be limited to the embodiments described herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for predicting the motion trajectory of cardiovascular lesions, characterized in that, The method includes: Obtain the current motion trajectory parameters of the patient's cardiovascular lesion area; The motion trajectory parameters are input into the motion trajectory prediction model to obtain the predicted trajectory data of the cardiovascular lesion area at future times; The motion trajectory prediction model is obtained by training the target machine learning model based on sample training data; the sample training data includes historical motion trajectory data of cardiovascular lesion areas and / or three-dimensional reconstruction models collected at different time points under different physiological and pathological conditions of patients.

2. The method according to claim 1, characterized in that, The target machine learning model includes at least a Transformer model; the sample training data also includes labeled data of historical motion trajectory data of the cardiovascular lesion area, and the labeled data includes discrete target point data of the cardiovascular lesion area.

3. The method according to claim 1 or 2, characterized in that, The method also includes: Acquire one or more of the patient's cardiovascular system via ultrasound, X-ray, and CT images; Based on the ultrasound images, X-ray images, and CT images, a three-dimensional reconstruction model of the cardiovascular lesion area under different physiological and pathological conditions of the patient is generated.

4. The method according to claim 3, characterized in that, The process of generating a three-dimensional reconstruction model of the cardiovascular lesion area under different physiological and pathological conditions of the patient based on the ultrasound images, X-ray images, and CT images includes: Image feature information is extracted from the ultrasound image, X-ray image and CT image, and a three-dimensional reconstruction model is generated based on the image feature information.

5. The method according to claim 1 or 2, characterized in that, The method also includes: Multiple ultrasound images, multiple X-ray images, and multiple CT images of the patient's cardiovascular system were acquired at different time points; The historical motion trajectory data of the cardiovascular lesion area is determined based on the multiple ultrasound images, multiple X-ray images, and multiple CT images.

6. The method according to claim 5, characterized in that, The determination of the historical motion trajectory data of the cardiovascular lesion area based on the multiple ultrasound images, multiple X-ray images, and multiple CT images includes: The first historical motion trajectory data of the cardiovascular lesion area is determined based on the multiple ultrasound images; the second historical motion trajectory data of the cardiovascular lesion area is determined based on the multiple X-ray images; the third historical motion trajectory data of the cardiovascular lesion area is determined based on the multiple CT images; and the historical motion trajectory data of the cardiovascular lesion area is determined by combining the first historical motion trajectory data, the second historical motion trajectory data, and the third historical motion trajectory data.

7. The method according to claim 1 or 2, characterized in that, The different physiological states of the patient are determined by different physiological parameters of the patient's body and / or different pathological changes in organs or tissues when the body is ill; the different pathological states of the patient are determined by different pathological changes in organs or tissues when the patient is ill.

8. A cardiovascular lesion motion trajectory prediction system, characterized in that, include: The data acquisition module is used to acquire the current motion trajectory parameters of the patient's cardiovascular lesion area; The trajectory prediction module is used to input the motion trajectory parameters into the motion trajectory prediction model to obtain the predicted trajectory data of the cardiovascular lesion area at future times; wherein, the motion trajectory prediction model is obtained by training the target machine learning model based on sample training data; the sample training data includes historical motion trajectory data of the cardiovascular lesion area and / or three-dimensional reconstruction model collected at different time points under different physiological and pathological states of the patient.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the cardiovascular lesion motion trajectory prediction method according to any one of claims 1 to 7.

10. An electronic device, characterized in that, include: processor; as well as Memory, used to store computer programs; The processor is configured to execute the cardiovascular lesion motion trajectory prediction method according to any one of claims 1 to 7 by executing the computer program.

Images