A patient tumor monitoring system based on stereo image reconstruction

Abstract

The present invention discloses a patient tumor monitoring system based on three-dimensional image reconstruction, which relates to the technical field of tumor monitoring. The system discloses a three-dimensional image data analysis module, a monitoring actual impact analysis module, a monitoring simulation impact analysis module, and a patient tumor monitoring adjustment module. The three-dimensional image data analysis module and the monitoring actual impact analysis module are set to analyze various types of three-dimensional image data of the patient's tumor, and the patient's tumor is accurately evaluated by combining deep learning. The monitoring simulation impact analysis module and the patient tumor monitoring adjustment module are set. By constructing a three-dimensional tumor image simulation model, the patient's tumor is simulated and analyzed, and the impact degree of the patient's tumor on the monitoring standard is comprehensively analyzed from both the actual impact and the simulation impact, so as to accurately quantify and adjust the monitoring strategy for the patient's tumor.

Description

Technical Field

[0001] The present invention relates to the technical field of tumor monitoring, and more specifically, it relates to a patient tumor monitoring system based on three-dimensional image reconstruction. Background Art

[0002] In the field of tumor treatment, continuous and accurate monitoring of a patient's tumor is crucial for ensuring the treatment effect and the quality of life of the patient. Traditional tumor monitoring methods mainly rely on doctors' clinical experience, regular imaging examinations, and the detection of tumor markers. However, these methods have many limitations. For example, doctors' clinical experience may be affected by subjective factors, regular imaging examinations may not be able to capture the subtle changes in tumor growth, and the detection of tumor markers may be interfered by various physiological and pathological factors.

[0003] With the rapid development of medical imaging technology, three-dimensional imaging technologies such as CT (Computed Tomography), MRI (Magnetic Resonance Imaging), and PET (Positron Emission Tomography) have become important tools for tumor diagnosis and treatment. These technologies can provide three-dimensional structural and functional information of tumors, providing doctors with more intuitive and accurate tumor images. However, there are still certain challenges in relying solely on these three-dimensional imaging data for tumor monitoring because the growth and spread of tumors are complex processes that require considering multiple factors.

[0004] Therefore, in order to overcome the limitations of traditional tumor monitoring methods and improve the accuracy and efficiency of tumor monitoring, the present invention proposes a patient tumor monitoring system based on three-dimensional image reconstruction. Summary of the Invention

[0005] Aiming at the deficiencies of the existing technology, the purpose of the present invention is to provide a patient tumor monitoring system based on three-dimensional image reconstruction.

[0006] To achieve the above purpose, the present invention provides the following technical solutions:

[0007] A patient tumor monitoring system based on three-dimensional image reconstruction, including a three-dimensional image data analysis module, a monitoring actual impact analysis module, a monitoring simulation impact analysis module, and a patient tumor monitoring adjustment module;

[0008] Every time a tumor image reconstruction duration T elapses, the three-dimensional image data analysis module collects various types of three-dimensional image data of the patient's tumor, obtains the basic data values of various types of three-dimensional image data, and determines whether to mark the three-dimensional image data as abnormal image data based on the comparison result between the basic data value and the basic data threshold;

[0009] The monitoring actual impact analysis module obtains a monitoring actual impact value based on the abnormal image data;

[0010] The monitoring simulation impact analysis module constructs a three-dimensional tumor image simulation model based on various types of three-dimensional image data of the patient's tumor and the historical data of the patient's tumor, and obtains the monitoring simulation impact value based on the three-dimensional tumor image simulation model;

[0011] The patient tumor monitoring adjustment module obtains a monitoring adjustment value based on the actual monitoring impact value and the monitoring simulation impact value, and determines whether to adjust the monitoring standard for the patient's tumor based on the comparison result between the monitoring adjustment value and the monitoring adjustment threshold.

[0012] Further, the method for obtaining the basic data value of each type of three-dimensional image data is as follows: obtain the basic data model corresponding to each type of three-dimensional image data, perform data feature extraction on each type of three-dimensional image data, obtain the feature data of each type of three-dimensional image data after extraction, and input the feature data of each type of three-dimensional image data into the corresponding basic data model respectively, and then output the basic data value of each type of three-dimensional image data.

[0013] Further, the actual monitoring impact value is obtained based on the abnormal image data. Specifically: mark the total number of abnormal image data as Hrw, perform pairwise matching processing on all abnormal image data, sum the basic data values of the two abnormal image data after matching to obtain the abnormal data accumulation value, set the abnormal data accumulation threshold, when the abnormal data accumulation value is greater than or equal to the abnormal data accumulation threshold, increase the abnormal accumulation count by one, mark the abnormal accumulation count as Zry, and use the formula to obtain the actual monitoring impact value DPX, where b1 is the abnormal data quantity coefficient and b2 is the abnormal accumulation count coefficient.

[0014] Further, the monitoring simulation impact value is obtained based on the three-dimensional tumor image simulation model. Specifically: perform a simulated change in the tumor image reconstruction duration T of the three-dimensional tumor image simulation model, set J tumor simulation time points at equal time intervals within the tumor image reconstruction duration T, obtain the simulated performance value Sjk of each tumor simulation time point, j = 1, 2,..., J, j is the serial number of the tumor simulation time point, set the simulated performance coefficient as Bk, obtain the number Rgs of simulated missing periods and the number Lzp of simulated performance fluctuations, and use the formula to obtain the monitoring simulation impact value MZE.

[0015] Further, the method for obtaining the number Rgs of simulated loss periods and the number Lzp of simulated performance fluctuations is as follows: Sort all the simulated performance values in the order of the tumor simulation time points. Sum the two adjacent simulated performance values after sorting to obtain the period performance value. Set the period performance threshold. When the period performance value is greater than or equal to the period performance threshold, increase the number of simulated loss periods by one. Mark the number of simulated loss periods as Rgs. Calculate the difference between the two adjacent simulated performance values after sorting and take the absolute value to obtain the simulated performance fluctuation value. Set the simulated performance fluctuation threshold. When the simulated performance fluctuation value is greater than or equal to the simulated performance fluctuation threshold, increase the number of simulated performance fluctuations by one. Mark the number of simulated performance fluctuations as Lzp.

[0016] Further, the method for obtaining the simulated performance value at the tumor simulation time point is as follows: Collect the basic data values of various types of stereoscopic image data when the tumor stereoscopic image simulation model is at the tumor simulation time point. Sum and average the basic data values of all types of stereoscopic image data to obtain the simulated performance value at the tumor simulation time point.

[0017] Further, obtain the monitoring adjustment value based on the monitored actual influence value and the monitored simulated influence value. Specifically: Use the formula to obtain the monitoring adjustment value SFUP, where wa is the actual influence coefficient and wb is the simulated influence coefficient.

[0018] Further, based on the comparison result between the monitoring adjustment value and the monitoring adjustment threshold, determine whether to adjust the monitoring standard for the patient's tumor. Specifically: Set the monitoring adjustment threshold as YZUP. When the monitoring adjustment value SFUP is greater than the monitoring adjustment threshold YZUP, after the tumor image reconstruction duration collect various types of stereoscopic image data of the patient's tumor next time. When the monitoring adjustment value SFUP is less than or equal to the monitoring adjustment threshold YZUP, maintain the tumor image reconstruction duration T and collect various types of stereoscopic image data of the patient's tumor next time.

[0019] Compared with the prior art, the present invention has the following beneficial effects:

[0020] 1. Set up a stereoscopic image data analysis module and a monitored actual influence analysis module to analyze various types of stereoscopic image data of the patient's tumor, and combine deep learning to accurately evaluate the patient's tumor;

[0021] 2. Set up a monitored simulated influence analysis module and a patient's tumor monitoring adjustment module. Through constructing a tumor stereoscopic image simulation model, conduct simulation analysis on the patient's tumor, comprehensively analyze the influence degree of the patient's tumor on the monitoring standard from both the actual influence and the simulated influence aspects, and then accurately quantify and adjust the monitoring strategy for the patient's tumor. Description of the Drawings

[0022] Figure 1 It is a system module diagram of a patient tumor monitoring system based on three-dimensional image reconstruction;

[0023] Figure 2 It is a system operation flowchart of a patient tumor monitoring system based on three-dimensional image reconstruction;

[0024] Figure 3 It is a flowchart for obtaining the basic data values of three-dimensional image data. Detailed implementation manner

[0025] Refer to Figures 1 - 3 , a patient tumor monitoring system based on three-dimensional image reconstruction, including a three-dimensional image data analysis module, a monitoring actual impact analysis module, a monitoring simulation impact analysis module, and a patient tumor monitoring adjustment module.

[0026] Three-dimensional image data analysis module: Set the tumor image reconstruction duration as T (the length of the tumor image reconstruction duration T is set according to the requirements of the monitoring system). Every time a tumor image reconstruction duration T passes, collect various types of three-dimensional image data of the patient's tumor (the types of three-dimensional image data of the patient's tumor include CT data, MRI data, MRA data, MRV data, PET data, SPEC data, etc.), obtain the basic data values of various types of three-dimensional image data, set the basic data threshold (the basic data threshold is a preset value of the system). When the basic data value of the three-dimensional image data is greater than or equal to the basic data threshold, mark the three-dimensional image data as abnormal image data. When the basic data value of the three-dimensional image data is less than the basic data threshold, no processing is performed.

[0027] The methods for obtaining the basic data values of various types of three-dimensional image data are as follows: Obtain the corresponding basic data models for various types of three-dimensional image data, extract the data features of various types of three-dimensional image data. After extraction, obtain the characteristic data of various types of three-dimensional image data, and input the characteristic data of various types of three-dimensional image data into the corresponding basic data models respectively, and then output to obtain the basic data values of various types of three-dimensional image data.

[0028] Different types of stereoscopic image data correspond to different basic data models. For example, CT data and MRI data each correspond to a basic data model. Each basic data model is constructed based on a deep learning model. The difference in the construction process of different basic data models lies in the different training data. The following will introduce the construction method of the basic data model for CT data: Collect n groups of CT data, extract data features from each group of CT data, and then obtain the feature data of n groups of CT data. Construct a deep learning model, use the feature data of CT data as the training data of the deep learning model, assign a basic data value to each training data, and the value range of the basic data value is (3.0 - 6.0). The larger the value of the basic data value, the more abnormal the CT data of the patient's tumor; the smaller the value of the basic data value, the more normal the CT data of the patient's tumor. Divide the training data into a training set, a validation set, and a test set according to the set ratio of 3:1:1, and train the training set, the validation set, and the test set. After the training is completed, the basic data model of CT data is obtained.

[0029] Monitoring actual impact analysis module: Mark the total number of abnormal image data as Hrw, perform pairwise matching processing on all abnormal image data, sum the basic data values of the two abnormal image data after matching to obtain the abnormal data accumulation value. Set the abnormal data accumulation threshold (the abnormal data accumulation threshold is a system preset value). When the abnormal data accumulation value is greater than or equal to the abnormal data accumulation threshold, increase the abnormal accumulation count by one. When the abnormal data accumulation value is less than the abnormal data accumulation threshold, do nothing. Mark the abnormal accumulation count as Zry, and use the formula to obtain the monitoring actual impact value DPX, where b1 is the abnormal data quantity coefficient, b2 is the abnormal accumulation count coefficient, the value of b1 is 0.92, and the value of b2 is 0.88.

[0030] Set up a stereoscopic image data analysis module and a monitoring actual impact analysis module to analyze various types of stereoscopic image data of the patient's tumor, and combine deep learning methods to accurately evaluate the patient's tumor.

[0031] Monitoring simulation impact analysis module: Based on various types of three-dimensional image data of the patient's tumor and the historical data of the patient's tumor (the historical data of the patient's tumor is the historical data of various types of three-dimensional image data of the patient's tumor), a three-dimensional tumor image simulation model is constructed. The three-dimensional tumor image simulation model simulates the change in the duration T of tumor image reconstruction. J tumor simulation time points at equal time intervals are set within the duration T of tumor image reconstruction, and the simulated performance values Sjk at each tumor simulation time point are obtained, where j = 1, 2,..., J, and j is the serial number of the tumor simulation time point. The simulated performance coefficient is set as Bk, where k = 1, 2,..., k, and B1 < B2 < B3 <... < Bk. A range of simulated performance values is set for each simulated performance coefficient. The ranges of simulated performance values include (0, Sj1], (Sj1, Sj2],..., (Sjk-1, Sjk]. When Sjk ∈ (Sj1, Sj2], the simulated performance coefficient is B2. All simulated performance values are sorted in the order of the tumor simulation time points, and the adjacent two simulated performance values after sorting are summed to obtain the period performance value. A period performance threshold is set (the period performance threshold is a system preset value). When the period performance value is greater than or equal to the period performance threshold, the number of simulated loss periods is increased by one. When the period performance value is less than the period performance threshold, no processing is performed. The number of simulated loss periods is marked as Rgs. The difference between the adjacent two simulated performance values after sorting is calculated and the absolute value is taken to obtain the simulated performance fluctuation value. A simulated performance fluctuation threshold is set (the simulated performance fluctuation threshold is a system preset value). When the simulated performance fluctuation value is greater than or equal to the simulated performance fluctuation threshold, the number of simulated performance fluctuations is increased by one. When the simulated performance fluctuation value is less than the simulated performance fluctuation threshold, no processing is performed. The number of simulated performance fluctuations is marked as Lzp. Using the formula The monitoring simulation impact value MZE is obtained.

[0032] The method for obtaining the simulated performance value at the tumor simulation time point is as follows: Collect the basic data values of various types of three-dimensional image data when the three-dimensional tumor image simulation model is at the tumor simulation time point, sum up and take the average of the basic data values of all types of three-dimensional image data to obtain the simulated performance value at the tumor simulation time point.

[0033] The method for constructing the three-dimensional tumor image simulation model is as follows: Select simulation software such as Materialise Mimics and Amira, create a new tumor model in the simulation software based on various types of three-dimensional image data of the patient's tumor, optimize the various parameters of the tumor model, determine the growth trend and diffusion speed of the tumor model based on the historical data of the patient's tumor, determine the change direction of the tumor model based on the growth trend and diffusion speed, and then construct the three-dimensional tumor image simulation model. The three-dimensional tumor image simulation model will simulate the change in the change direction.

[0034] Patient tumor monitoring adjustment module: Using the formula to obtain the monitoring adjustment value SFUP. Here, wa is the actual influence coefficient, wb is the simulation influence coefficient. The value of wa is 0.52, and the value of wb is 0.43. Set the monitoring adjustment threshold as YZUP. When the monitoring adjustment value SFUP is greater than the monitoring adjustment threshold YZUP, after the tumor image reconstruction duration collect the next set of three-dimensional image data of the patient's tumor in various types. When the monitoring adjustment value SFUP is less than or equal to the monitoring adjustment threshold YZUP, maintain the tumor image reconstruction duration T and collect the next set of three-dimensional image data of the patient's tumor in various types.

[0035] Set up the monitoring simulation influence analysis module and the patient tumor monitoring adjustment module. By constructing a three-dimensional tumor image simulation model, perform a simulation analysis on the patient's tumor, comprehensively analyze the influence degree of the patient's tumor on the monitoring standard from both the actual influence and the simulation influence aspects, and then accurately quantify and adjust the monitoring strategy for the patient's tumor.

[0036] All the above formulas are dimensionless and take their numerical calculations. The formula is a formula obtained by software simulation through collecting a large amount of data to approximate the real situation. The preset parameters in the formula are set by those skilled in the art according to the actual situation.

[0037] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the processes or functions described in the embodiments of the present application are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center in a wired or wireless (such as infrared, wireless, microwave, etc.) manner. The computer-readable storage medium can be any available medium that the computer can access or a data storage device such as a server or data center that contains one or more sets of available media. The available media can be magnetic media (such as floppy disks, hard disks, magnetic tapes), optical media (such as DVDs), or semiconductor media. The semiconductor media can be a solid-state drive.

[0038] It should be understood that in various embodiments of the present application, the magnitudes of the sequence numbers of the above processes do not imply the order of execution, and the order of execution of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present application.

[0039] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods for each specific application to implement the described functions, but such implementation should not be considered to exceed the scope of the present application.

[0040] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the systems, devices, and units described above can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated herein.

[0041] In several embodiments provided by the present application, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is only a logical function division, and there may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings, direct couplings, or communication connections to each other can be through some interfaces, and the indirect couplings or communication connections of the devices or units can be in an electrical, mechanical, or other forms.

[0042] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present application. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs that can store program codes.

[0043] As described above, it is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present application can easily think of changes or substitutions, which should all be covered within the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims described above.

Claims

1. A patient tumor monitoring system based on stereoscopic image reconstruction, characterized in that, It includes a stereoscopic image data analysis module, a monitoring actual impact analysis module, a monitoring simulation impact analysis module, and a patient tumor monitoring adjustment module; For each tumor image reconstruction duration, the three-dimensional image data analysis module collects various types of three-dimensional image data of the patient's tumor, obtains the basic data values of the various types of three-dimensional image data, and determines whether to mark the three-dimensional image data as abnormal image data based on the comparison result between the basic data value and the basic data threshold; The monitoring actual impact analysis module obtains the monitoring actual impact value based on the abnormal image data; Mark the total number of abnormal image data as Hrw. Pairwise match all the abnormal image data, sum the basic data values of the two abnormal image data after matching to obtain the abnormal data cumulative value. Set the abnormal data cumulative threshold. When the abnormal data cumulative value is greater than or equal to the abnormal data cumulative threshold, increase the abnormal cumulative count by one. Mark the abnormal cumulative count as Zry, and use the formula to obtain the actual monitoring impact value DPX, where b1 is the abnormal data quantity coefficient and b2 is the abnormal cumulative count coefficient; The monitoring simulation impact analysis module constructs a tumor three-dimensional image simulation model based on various types of three-dimensional image data of the patient's tumor and the historical data of the patient's tumor. The duration of tumor three-dimensional image reconstruction in the tumor three-dimensional image simulation model of the simulation change, within the duration of tumor image reconstruction set J tumor simulation time points at equal time intervals, obtain the simulation performance values Sjk at each tumor simulation time point, j = 1, 2, …, J, j is the serial number of the tumor simulation time point, set the simulation performance coefficient as Bk, sort all the simulation performance values in the order of the tumor simulation time points, perform a summation process on two adjacent simulation performance values after sorting to obtain the period performance value, set the period performance threshold. When the period performance value is greater than or equal to the period performance threshold, increase the number of simulation loss periods by one, mark the number of simulation loss periods as Rgs, calculate the difference between two adjacent simulation performance values after sorting and take the absolute value to obtain the simulation performance fluctuation value, set the simulation performance fluctuation threshold. When the simulation performance fluctuation value is greater than or equal to the simulation performance fluctuation threshold, increase the simulation performance fluctuation times by one, mark the simulation performance fluctuation times as Lzp, and use the formula to obtain the monitoring simulation impact value MZE; The patient tumor monitoring adjustment module uses the formula to obtain the monitoring adjustment value SFUP. Among them, wa is the actual influence coefficient, wb is the simulation influence coefficient, and the monitoring adjustment threshold is set as YZUP. When the monitoring adjustment value SFUP is greater than the monitoring adjustment threshold YZUP, after the tumor image reconstruction duration , collect the three-dimensional image data of various types of the patient's tumor next time. When the monitoring adjustment value SFUP is less than or equal to the monitoring adjustment threshold YZUP, maintain the tumor image reconstruction duration , and collect the three-dimensional image data of various types of the patient's tumor next time.

2. The patient tumor monitoring system based on stereoscopic image reconstruction according to claim 1, characterized in that The method for obtaining the basic data value of each type of stereoscopic image data is as follows: Obtain the basic data model corresponding to each type of stereoscopic image data, extract the data features of each type of stereoscopic image data, and after extraction, obtain the feature data of each type of stereoscopic image data. Input the feature data of each type of stereoscopic image data into the corresponding basic data model respectively, and then output the basic data value of each type of stereoscopic image data.

3. A patient tumor monitoring system based on stereoscopic image reconstruction according to claim 1, characterized in that, The method for obtaining the simulation performance value at the tumor simulation time point is as follows: Collect the basic data values of each type of stereoscopic image data when the tumor stereoscopic image simulation model is at the tumor simulation time point, sum up the basic data values of all types of stereoscopic image data and take the average value to obtain the simulation performance value at the tumor simulation time point.

Images