Training method and device of detection result correction model applied to radiotherapy

Abstract

The invention discloses a training method and device for a detection result correction model applied to radiotherapy, and relates to the technical field of radiotherapy. In the process that a plurality of detectors are irradiated by radiated light for n times, detector parameters and irradiation dose values of each detector after each time of irradiation can be collected. And determining correction coefficient data according to the plurality of irradiation dose values. Performing data preprocessing on the plurality of detector parameters and dose measurement values, and determining a training sample; and inputting the correction coefficient data and the training sample into the detection result correction model, and training the model until the loss function of the detection result correction model converges. Therefore, the long-term and cumulative degradation effect caused by ionizing radiation damage of the detector body after the detector is irradiated for multiple times is considered, and training of the detection result correction model is realized.

Description

Technical Field

[0001] This specification relates to the field of radiotherapy technology, and in particular to a training method and apparatus for a detection result correction model applied to radiotherapy. Background Technology

[0002] In modern radiotherapy, the accuracy of dose measurement is crucial for ensuring treatment effectiveness and patient safety. To achieve high-precision dose verification and treatment quality control, detectors such as X-ray flat panel detectors (FPDs), ionization chamber arrays, and diode arrays are typically used for dose verification.

[0003] However, in long-term, high-dose clinical applications, the detector itself, as a radiation-sensitive device, will inevitably suffer cumulative damage to its electronic components and materials from ionizing radiation. This affects the detector's response accuracy to X-ray signals, leading to systematic drift or error accumulation in dose measurement results.

[0004] Currently, common methods to compensate for deviations in detector measurement results mainly focus on correcting the effects of ambient temperature and humidity. However, these methods neglect the long-term, cumulative degradation effects caused by ionizing radiation damage to the detector itself, making it difficult to meet the dose measurement accuracy requirements under high-dose, long-term operating environments. Summary of the Invention

[0005] This specification provides a training method and apparatus for a detection result correction model applied to radiotherapy, so as to at least partially solve the above-mentioned problems existing in the prior art.

[0006] The following technical solution is adopted in this specification: This specification provides a training method for a detection result correction model applied to radiotherapy, including: During the process of irradiating multiple detectors with radiation light n times, the detector parameters and radiation dose values ​​are collected after each irradiation. The detector parameters include breakdown voltage and / or series resistance and / or quenching resistance and / or parasitic resistance and / or pixel capacitive reactance and / or discharge time constant and / or gain and / or dark count rate and / or photon detection efficiency. The radiation dose values ​​include standard dose values ​​and dose measurements collected by the detectors. Based on the multiple irradiation dose values, determine the correction factor data; Data preprocessing is performed on multiple detector parameters and dose measurements to determine training samples; The correction coefficient data and the training samples are input into the detection result correction model, and the detection result correction model is trained until the loss function of the detection result correction model converges.

[0007] Preferably, before irradiating the multiple detectors n times, the method further includes: For any given detector, determine the maximum radiation dose corresponding to that detector based on its usage scenario; Based on the maximum radiation dose corresponding to the detector, determine the single radiation dose when the detector is irradiated n times.

[0008] Preferably, determining the correction coefficient data based on the plurality of irradiation dose values ​​includes: For any given irradiation dose value, the correction factor value corresponding to that irradiation dose value is determined using the following formula; in, This is the correction coefficient value corresponding to the i-th detector after being illuminated for the nth time; This represents the standard dose value corresponding to the i-th detector after being irradiated for the nth time; This represents the dose measurement value collected by the i-th detector after it has been irradiated for the nth time. Based on the multiple correction coefficient values, the correction coefficient data is determined.

[0009] Preferably, data preprocessing is performed on multiple detector parameters and dose measurements to determine training samples, including: The correlation between detector parameters and dose measurements was established using the least squares method; the correlation is as follows: in, The breakdown voltage; The series resistor; The quenching resistance; The parasitic resistance; The pixel capacitance; G is the discharge time constant; DCR is the dark count rate; PDE is the photon detection efficiency; and D is the dose measurement value. The target algorithm is used to solve the above-mentioned correlations and determine the feature vector matrix; Training samples are determined based on the feature vector, the multiple detector parameters, and the dose measurement values.

[0010] Preferably, the number of detectors is i; Both D and are i×n matrices, and the data at the same position correspond to each other; The step of solving the aforementioned correlations using a target algorithm to determine training samples includes: Through feature extraction methods Dimensionality reduction is performed by using principal component analysis to solve for the eigenvectors of the sample space covariance matrix; the sample space is... The element in the s-th row and t-th column of the covariance matrix is ​​defined as follows: in, For the sample space, Let f be the f-th sample; any sample is a flattened i×10 dimension vector. It is the covariance matrix; This represents the value of the f-th sample in the s-th dimension; The average value of the samples in the s-th dimension; Perform eigenvalue decomposition on the covariance matrix: in, Ω is a diagonal matrix, where each diagonal element is an eigenvalue of the covariance matrix M and they are all distinct; Ω is the eigenvector matrix; each column of Ω is the eigenvector corresponding to the covariance matrix. Training samples are determined based on the feature vector, the multiple detector parameters, and the dose measurement values.

[0011] Preferred, It can be determined by the following formula: in, This represents the value of the f-th sample in the s-th dimension.

[0012] On the other hand, this specification also provides a method for correcting detection results applied to radiotherapy, which utilizes the detection result correction model for radiotherapy trained in the above-mentioned aspect, including: Acquire the dose measurements collected by the detector, as well as the breakdown voltage and / or series resistance and / or quenching resistance and / or parasitic resistance and / or pixel capacitive reactance and / or discharge time constant and / or gain and / or dark count rate and / or photon detection efficiency; The dose measurement value, as well as the breakdown voltage and / or series resistance and / or quenching resistance and / or parasitic resistance and / or pixel capacitance and / or discharge time constant and / or gain and / or dark count rate and / or photon detection efficiency, are input into the detection result correction model applied to radiotherapy to determine the corrected dose value output by the detection result correction model applied to radiotherapy.

[0013] On the other hand, this specification provides a training device for a detection result correction model applied to radiotherapy, comprising: The acquisition unit is used to acquire detector parameters and radiation dose values ​​after each irradiation during the process of irradiating multiple detectors with radiation light n times. The detector parameters include breakdown voltage and / or series resistance and / or quenching resistance and / or parasitic resistance and / or pixel capacitance and / or discharge time constant and / or gain and / or dark count rate and / or photon detection efficiency. The radiation dose values ​​include standard dose values ​​and dose measurements acquired by the detectors. A determining unit is used to determine correction coefficient data based on the plurality of irradiation dose values; The processing unit is used to preprocess multiple detector parameters and dose measurements to determine training samples. The training unit is used to input the correction coefficient data and the feature vector into the detection result correction model, and train the detection result correction model until the loss function of the detection result correction model converges.

[0014] On the other hand, this specification provides a detection result correction device for use in radiotherapy, comprising: The acquisition unit is used to acquire the dose measurement value collected by the detector, as well as the breakdown voltage and / or series resistance and / or quenching resistance and / or parasitic resistance and / or pixel capacitive reactance and / or discharge time constant and / or gain and / or dark count rate and / or photon detection efficiency. A correction unit is used to input the dose measurement value, as well as the breakdown voltage and / or series resistance and / or quenching resistance and / or parasitic resistance and / or pixel capacitance and / or discharge time constant and / or gain and / or dark count rate and / or photon detection efficiency into the detection result correction model applied to radiotherapy, and determine the corrected dose value output by the detection result correction model applied to radiotherapy; the detection result correction model is the detection result correction model applied to radiotherapy trained by the above-mentioned aspect.

[0015] On the other hand, the computer-readable storage medium provided in this specification stores a computer program that, when executed by a processor, implements the training method for the detection result correction model applied to radiotherapy or the detection result correction method applied to radiotherapy provided in one aspect above.

[0016] On the other hand, this specification provides an electronic device including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the training method for a detection result correction model applied to radiotherapy or the detection result correction method applied to radiotherapy provided in the above-mentioned aspect.

[0017] On the other hand, this specification provides a computer program product in which the instructions are executed by the processor of an electronic device, causing the electronic device to implement the training method for the detection result correction model applied to radiotherapy or the detection result correction method applied to radiotherapy provided in the above-mentioned aspect.

[0018] The above-mentioned technical solutions adopted in this specification can achieve the following beneficial effects: In the training method for the detection result correction model applied to radiotherapy provided in this specification, during the process of irradiating multiple detectors with radiation light n times, the detector parameters and radiation dose values ​​of each detector are collected after each irradiation. Correction coefficient data is then determined based on these multiple radiation dose values. Next, the multiple detector parameters and dose measurements are preprocessed to determine training samples. The correction coefficient data and the training samples are then input into the detection result correction model to train the model until its loss function converges.

[0019] As can be seen from the above method, the long-term, cumulative degradation effect caused by ionizing radiation damage to the detector body after multiple irradiations is taken into account, and the training of the detection result correction model is realized. Attached Figure Description

[0020] The accompanying drawings, which are included to provide a further understanding of this specification and form part of this specification, illustrate exemplary embodiments and are used to explain this specification, but do not constitute an undue limitation thereof. In the drawings: Figure 1 A flowchart illustrating a training method for a detection result correction model applied to radiotherapy, provided as an embodiment of this specification; Figure 2 A flowchart illustrating a method for correcting detection results applied to radiotherapy, provided as an embodiment of this specification; Figure 3 A schematic diagram of a training device for a detection result correction model applied to radiotherapy, provided as an embodiment of this specification; Figure 4 A schematic diagram of a detection result correction device for radiotherapy provided as an embodiment of this specification; Figure 5 This is a schematic diagram of the structure of an electronic device provided as an embodiment of this specification. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of this specification clearer, the technical solutions of this specification will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this specification, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments in this specification without creative effort are within the scope of protection of this application.

[0022] In the description of this invention, it should be noted that the term "or" is generally used to include the meaning of "and / or" unless otherwise expressly stated in the content.

[0023] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. Furthermore, in the description of this application, the terms "first," "second," etc., are used only for distinguishing descriptions and should not be construed as indicating or implying relative importance.

[0024] Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0025] The technical solutions provided in the various embodiments of this specification are described in detail below with reference to the accompanying drawings.

[0026] Figure 1 A flowchart illustrating a training method for a detection result correction model applied to radiotherapy, as provided in one embodiment of this specification, is shown below. Figure 1 As shown, the training method for the detection result correction model applied to radiotherapy specifically includes the following steps: S100: During the process of irradiating multiple detectors with radiation light n times, the detector parameters and radiation dose values ​​of each detector are collected after each irradiation.

[0027] Preferably, the training method for the detection result correction model applied to radiotherapy can be executed by an electronic device. This electronic device can be a computer, server, etc., and this specification does not limit its use.

[0028] Preferably, the detector parameters include breakdown voltage and / or series resistance and / or quenching resistance and / or parasitic resistance and / or pixel capacitive reactance and / or discharge time constant and / or gain and / or dark count rate and / or photon detection efficiency, etc.

[0029] Preferably, the irradiation dose value includes a standard dose value and a dose measurement value collected by the detector. The standard dose value is the absorbed dose value measured under known radiation conditions by a standard measuring device or reference instrument (such as a calibrated ionization chamber) and used as the basis for dose measurement calibration.

[0030] Preferably, the detector parameters and the radiation dose value can be collected by the electronic device in response to the user's input operation, or they can be collected by the sensors configured in the electronic device. This specification does not limit this.

[0031] Preferably, the detector needs to be disassembled during the process of collecting the detector parameters.

[0032] Using the above method, the electronic device acquires the detector parameters and irradiation dose values ​​under different cumulative dose conditions.

[0033] Preferably, this step aims to systematically collect data on how the physical performance parameters of the detector change with cumulative radiation damage by simulating the long-term use of the detector. The effect is to establish a dynamic dataset that reflects the detector's performance degradation process, providing the necessary time / dose-related sample basis for subsequently training a corrective model that can compensate for this long-term cumulative effect.

[0034] S102: Determine correction coefficient data based on the multiple irradiation dose values.

[0035] Preferably, the electronic device can determine the correction factor value corresponding to any given irradiation dose value using the following formula; in, This is the correction coefficient value corresponding to the i-th detector after being illuminated for the nth time; This represents the standard dose value corresponding to the i-th detector after being irradiated for the nth time; This represents the dose measurement value collected by the i-th detector after the n-th irradiation.

[0036] Preferably, the collected multiple correction coefficient values ​​can form an i×n matrix.

[0037] Further preferably, the electronic device can determine the correction coefficient data based on the plurality of correction coefficient values. Specifically, the electronic device can directly determine the correction coefficient data as an i×n matrix formed by multiple dose measurements. Alternatively, the electronic device can determine the average correction coefficient value corresponding to the same n value using the arithmetic mean method for correction coefficient values, and then, based on the plurality of average correction coefficient values ​​and their corresponding n values ​​and / or standard dose values, fit a first relationship function between n and the average correction coefficient values, or fit a second relationship function between the standard dose value and the average correction coefficient values, and finally determine the first relationship function or the second relationship function as the correction coefficient data. Of course, the above is merely an illustrative example, and the electronic device can also use other methods to determine the correction coefficient data, which is not limited herein.

[0038] Preferably, this step aims to quantify the detector's measurement error at different stages of cumulative damage by calculating the ratio of the standard dose value to the detector's measurement value. The effect is to generate a set of correction coefficients for subsequent model training, allowing the model to learn how to predict and compensate for actual measurement deviations based on the detector's state.

[0039] S104: Perform data preprocessing on multiple detector parameters and dose measurements to determine the eigenvector matrix.

[0040] Preferably, the electronic device can perform data preprocessing on multiple detector parameters and dose measurements to determine the feature vector matrix.

[0041] Specifically, the electronic device can preferentially establish the correlation between detector parameters and dose measurements using the least squares method; the correlation is as follows: in, This is the breakdown voltage; This is the series resistor; This is the quenching resistor; This is the parasitic resistance; This is the capacitance of the pixel; G is the discharge time constant; DCR is the dark count rate; PDE is the photon detection efficiency; and D is the dose measurement value.

[0042] Preferably, the number of detectors is i.

[0043] Further preferred, Both D and are i×n matrices, and the data at the same position correspond to each other.

[0044] Then, the electronic device can use feature extraction methods to... Dimensionality reduction is performed, and principal component analysis is used to solve for the eigenvectors of the sample space covariance matrix.

[0045] Wherein, the sample space is .in, For the sample space, Let f be the f-th sample; any sample is a flattened vector of dimensions i×10. 10, as mentioned above. The number of D elements.

[0046] The element in the s-th row and t-th column of the covariance matrix is ​​defined as follows: in, It is the covariance matrix; This is the element in the s-th row and t-th column of the covariance matrix; This represents the value of the f-th sample in the s-th dimension; is the sample mean in the s-th dimension.

[0047] Furthermore, preferably, the electronic device can perform eigenvalue decomposition on the covariance matrix: in, is a diagonal matrix, where each diagonal element is an eigenvalue of the covariance matrix M, and all diagonal elements are distinct; Ω is the eigenvector matrix, Ω d = 10 × i; each column of Ω is the eigenvector corresponding to the covariance matrix M. The eigenvectors of M are the principal components of the samples, arranged in descending order of eigenvalues, which yields the 1st, 2nd, ..., nth principal components of Ψ: PC1, PC2, ..., PC... n These n principal components ultimately form a complete description of Ψ.

[0048] Further preferably, the electronic device can determine training samples based on the feature vector, the multiple detector parameters, and the dose measurement. Specifically, firstly, the electronic device can determine the original dataset. Where n is the number of samples; d = i × 10, d represents the dimension of each sample. i is the number of detectors, and 10 is the aforementioned... The number of elements in D. Here, each row of the sample matrix X represents a sample, and each column represents a feature. It is important to emphasize that in determining the training samples, only the aforementioned elements may be used. One or more of the following, that is, d=i×2, d=i×3, d=i×4, d=i×5, d=i×6, d=i×7, d=i×8, and d=i×9 are all possible. Similarly, in other parts of this specification, Of the nine elements, only one element may be used.

[0049] Secondly, the electronic device can determine the mean vector of each feature, and then determine the centralized matrix according to the sample matrix and the mean vector through the following formula: where, is the centralized matrix; X is the original data set; 1 is a column vector of all 1s; is the mean vector of each feature; the meaning of the superscript T is transpose, that is, transposing a row vector into a column vector.

[0050] Then, the electronic device can select the first k principal component feature vectors from the feature vector matrix Ω , where, is the k-th feature vector, and k << d. k is the number of principal components. The electronic device can determine the value of k based on methods such as the cumulative variance contribution rate or fixed dimension, etc., which is not limited in this specification.

[0051] Finally, the electronic device can project the centralized matrix into the principal component space through the following formula to determine the training samples: where, ; Z is the training sample after dimensionality reduction. [[ID=二十五]] [[ID=二十六]]

[0052] Of course, the content of determining the training samples based on the feature vector, the multiple detector parameters, and the dose measurement value above is only an exemplary illustration, and the electronic device can also use other methods to determine, which is not limited in this specification.

[0053] Preferably, this step aims to perform dimensionality reduction and feature extraction on high-dimensional detector data containing multiple physical parameters through methods such as principal component analysis. The effect is to remove redundant information in the original data and extract the core features that can most effectively characterize the aging state of the detector due to radiation damage, thereby simplifying the subsequent model training process and improving its correction performance.

[0054] S106: Input the correction coefficient data and the feature vector into the detection result correction model, and train the detection result correction model until the loss function of the detection result correction model converges; the input of the detection result correction model is the dose measurement value of the detector, and the output is the corrected dose value. This step is the core learning link of the entire method, aiming to train the correction model using the training samples representing the physical state of the detector and the correction coefficients representing its measurement errors. The final effect is to generate a trained intelligent model that can understand and master the performance degradation law of the detector due to long-term radiation damage, so that it can accurately predict and correct its dose measurement results according to the current physical parameters of the detector.

[0055] Based on Figure 1 The method shown describes a training method for a detection result correction model applied to radiotherapy. The electronic device collects detector parameters and radiation dose values ​​for each detector after each irradiation as multiple detectors are irradiated n times by radiation. Correction coefficients are determined based on these multiple radiation dose values. Then, the detector parameters and dose measurements are preprocessed to determine training samples. These correction coefficients and training samples are then input into the detection result correction model for training until the model's loss function converges.

[0056] As can be seen from the above method, the electronic device takes into account the long-term, cumulative degradation effect caused by ionizing radiation damage to the detector body after multiple irradiations, and realizes the training of the detection result correction model.

[0057] Preferably, the electronic device can also irradiate multiple detectors n times with radiated light.

[0058] More preferably, before performing step S100, the electronic device can also acquire the detector parameters and the irradiation dose value before irradiation.

[0059] Those skilled in the art will understand that detectors can be used in various scenarios, and the acceptable radiation dose for a detector varies depending on the scenario. Therefore, before executing step S100, the electronic device can determine the maximum radiation dose corresponding to any detector based on the detector's usage scenario. Then, based on the maximum radiation dose corresponding to the detector, the single radiation dose for n irradiations of the detector can be determined.

[0060] More preferably, for each detector, the single radiation dose is the same during n irradiations.

[0061] Preferably, during step S104, the electronic device can be determined using the following formula. : in, This represents the value of the f-th sample in the s-th dimension.

[0062] More preferably, before performing step S106, the electronic device can construct a multi-layer backpropagation neural network (BP) model. The BP model includes, but is not limited to, a fully connected BP model. This description uses a fully connected BP model with one input layer, three hidden layers, and one output layer as an example.

[0063] Furthermore, the electronic device can also determine the activation function of the BP model.

[0064] Further preferably, the derivative of the activation function should be continuously differentiable within the domain of the input signal, and the derivative value should change significantly to meet the requirements for gradient sensitivity during error descent.

[0065] Preferably, the activation function is a sigmoid function, which includes a sigmoid function with a range of (0,1) and a tangent function with a range of (-1,1). Of course, the electronic device may also choose other activation functions, and this specification does not impose any restrictions.

[0066] More preferably, the training function for this BP model is the Levenberg-Marquardt (LM) algorithm, which solves a nonlinear least squares problem. Of course, other training functions can also be selected for this electronic device, and this specification does not impose any restrictions on them.

[0067] More preferably, after completing step S106, the electronic device can be tested in the following manner.

[0068] Specifically, firstly, the electronic device acquires dose measurements and standard dose values ​​when the detector is in a Co-60 radiation field with total irradiation doses of 5000 Gy, 10000 Gy, and 100000 Gy, respectively. It also acquires breakdown voltage and / or series resistance and / or quenching resistance and / or parasitic resistance and / or pixel capacitive reactance and / or discharge time constant and / or gain and / or dark count rate and / or photon detection efficiency, determined using shift testing of the equivalent electrical parameters and dynamic functional parameters of the test sample. This data is then input into the detection result correction model to determine the corrected dose value, which is then compared and verified with the standard dose value. If the error is less than a preset error parameter, the model training is considered complete. If the error is greater than or equal to the error parameter, the process returns to step S100 to retrain the model until the error is less than the preset error parameter. Of course, the total irradiation dose mentioned above is only an example; other values ​​can also be used.

[0069] Table 1

[0070] Table 1 is a comparison table of dose measurement values ​​and standard dose values ​​provided in one embodiment of this specification when the total dose is 2000, 4000, and 8000 Gy, respectively. As shown in Table 1, the error is greatly reduced.

[0071] Based on the above-mentioned dose measurement result correction method and steps, the error in radiation dose measurement results caused by ionizing radiation damage to the X-ray imaging plate can be corrected. This can achieve relatively accurate measurement of the radiation dose of the device under different irradiation conditions, and make up for the lack of susceptibility of X-ray imaging plate devices to radiation damage in strong radiation environments.

[0072] Preferably, the input to the detection result correction model is the dose measurement value, as well as the breakdown voltage and / or series resistance and / or quenching resistance and / or parasitic resistance and / or pixel capacitive reactance and / or discharge time constant and / or gain and / or dark count rate and / or photon detection efficiency. The input to the detection result correction model is the corrected dose value.

[0073] The above are training methods for detection result correction models applied to radiotherapy provided by one or more embodiments of this specification. Based on the same idea, this specification also provides a detection result correction method applied to radiotherapy.

[0074] Figure 2 A flowchart illustrating a detection result correction method for radiotherapy, provided as an embodiment of this specification, is shown below. Figure 2 As shown, the method includes the following steps.

[0075] S200: Acquire the dose measurement values ​​collected by the detector, as well as the breakdown voltage and / or series resistance and / or quenching resistance and / or parasitic resistance and / or pixel capacitive reactance and / or discharge time constant and / or gain and / or dark count rate and / or photon detection efficiency.

[0076] Preferably, the method can be executed by an electronic device, such as a computer, server, etc., which is not limited herein.

[0077] Preferably, the electronic device stores the aforementioned Figure 1 The trained model for correcting detection results applied to radiotherapy.

[0078] Preferably, the electronic device can acquire dose measurements collected by the detector, as well as breakdown voltage and / or series resistance and / or quenching resistance and / or parasitic resistance and / or pixel capacitive reactance and / or discharge time constant and / or gain and / or dark count rate and / or photon detection efficiency.

[0079] S202: Input the dose measurement value, as well as the breakdown voltage and / or series resistance and / or quenching resistance and / or parasitic resistance and / or pixel capacitance and / or discharge time constant and / or gain and / or dark count rate and / or photon detection efficiency into the detection result correction model applied to radiotherapy, and determine the corrected dose value output by the detection result correction model applied to radiotherapy.

[0080] It should be noted that all actions involving the acquisition of signals, information, or data in this application are carried out in compliance with the relevant data protection laws and policies of the country where the application is located, and with the authorization granted by the owner of the relevant device.

[0081] The above describes a training method and a method for correcting detection results applied to radiotherapy, provided by one or more embodiments of this specification. Based on the same idea, this specification also provides a corresponding training device for correcting detection results applied to radiotherapy, such as... Figure 3 As shown.

[0082] Figure 3 A schematic diagram of a training device for a detection result correction model applied to radiotherapy, provided as an embodiment of this specification, is shown below. Figure 3 As shown, the training device for the detection result correction model applied to radiotherapy specifically includes: The acquisition unit 300 is used to irradiate multiple detectors n times with radiated light and to acquire detector parameters and irradiation dose values ​​after each irradiation; the detector parameters include breakdown voltage and / or series resistance and / or quenching resistance and / or parasitic resistance and / or pixel capacitive reactance and / or discharge time constant and / or gain and / or dark count rate and / or photon detection efficiency; the irradiation dose values ​​include standard dose values ​​and dose measurement values ​​acquired by the detectors. The determining unit 302 is used to determine correction coefficient data based on the plurality of irradiation dose values; Processing unit 304 is used to preprocess multiple detector parameters and dose measurement values ​​to determine the feature vector matrix; Training unit 306 is used to input the correction coefficient data and the feature vector into the detection result correction model, and train the detection result correction model until the loss function of the detection result correction model converges.

[0083] Preferably, the acquisition unit 300 is further configured to determine the maximum radiation dose corresponding to any detector based on the usage scenario of the detector; and to determine the single radiation dose when the detector is irradiated n times based on the maximum radiation dose corresponding to the detector.

[0084] Preferably, the determining unit 302 is further configured to determine the correction coefficient value corresponding to any given irradiation dose value using the following formula; in, This is the correction coefficient value corresponding to the i-th detector after being illuminated for the nth time; This represents the standard dose value corresponding to the i-th detector after being irradiated for the nth time; The dose measurement value collected by the i-th detector after the n-th irradiation; the correction coefficient data is determined based on the multiple correction coefficient values.

[0085] Preferably, the processing unit 304 is further configured to establish a correlation between detector parameters and dose measurements using the least squares method; the correlation is as follows: in, The breakdown voltage; The series resistor; The quenching resistance; The parasitic resistance; The pixel capacitance; G is the discharge time constant; G is the gain; DCR is the dark count rate; PDE is the photon detection efficiency; D is the dose measurement value; the above correlations are solved by the target algorithm to determine the feature vector matrix.

[0086] Preferably, the number of detectors is i; Both D and are i×n matrices, and data at the same position correspond to each other; this processing unit 304 is also used to extract features using a feature extraction method. Dimensionality reduction is performed by using principal component analysis to solve for the eigenvectors of the sample space covariance matrix; the sample space is... The element in the s-th row and t-th column of the covariance matrix is ​​defined as follows: in, For the sample space, Let f be the f-th sample; any sample is a flattened i×10 dimension vector. It is the covariance matrix; This represents the value of the f-th sample in the s-th dimension; The covariance matrix is ​​the sample mean in the s-th dimension; eigenvalue decomposition is performed on the covariance matrix: in, It is a diagonal matrix, where each diagonal element is an eigenvalue of the covariance matrix M and they are all different; Ω is an eigenvector matrix; each column of Ω is an eigenvector corresponding to the covariance matrix; training samples are determined based on the eigenvectors, the multiple detector parameters, and the dose measurement values.

[0087] Preferably, the processing unit 304 is further configured to determine by the following formula : in, This represents the value of the f-th sample in the s-th dimension.

[0088] The above describes a training method and apparatus for a detection result correction model applied to radiotherapy, and a method for correcting detection results applied to radiotherapy, provided by one or more embodiments of this specification. Based on the same idea, this specification also provides a corresponding detection result correction apparatus applied to radiotherapy, such as... Figure 4 As shown.

[0089] Figure 4 A schematic diagram of a detection result correction device for radiotherapy provided as an embodiment of this specification, as shown below. Figure 4 As shown, the detection result correction device applied to radiotherapy specifically includes: Acquisition unit 400 is used to acquire dose measurement values ​​collected by the detector; The correction unit 402 is used to input the dose measurement value into the detection result correction model applied to radiotherapy, and determine the corrected dose value output by the detection result correction model applied to radiotherapy; the detection result correction model applied to radiotherapy is the detection result correction model applied to radiotherapy trained in any of the foregoing embodiments.

[0090] This specification also provides a computer-readable storage medium storing a computer program that can be used to execute the above-described... Figure 1 The provided training method or method for a detection result correction model applied to radiotherapy Figure 2 A method for correcting detection results applied to radiotherapy is provided.

[0091] This specification also provides a computer program product in which instructions, when executed by the processor of an electronic device, cause the electronic device to perform the above-described functions. Figure 1 The provided training method or method for a detection result correction model applied to radiotherapy Figure 2 A method for correcting detection results applied to radiotherapy is provided.

[0092] Figure 5 This is a schematic diagram of the structure of an electronic device provided as an embodiment of this specification. Figure 5 As shown, at the hardware level, this electronic device includes a processor, internal bus, network interface, memory, and non-volatile memory, and may also include other hardware required for business operations. The processor reads the corresponding computer program from the non-volatile memory into memory and then runs it to achieve the above. Figure 1 The training method for the detection result correction model applied to radiotherapy or Figure 2 This document provides a method for correcting detection results used in radiotherapy. Of course, besides software implementation, this specification does not exclude other implementation methods, such as logic devices or a combination of hardware and software. In other words, the execution entity of the following processing flow is not limited to individual logic units, but can also be hardware or logic devices.

[0093] In the 1990s, improvements to a technology could be clearly distinguished as either hardware improvements (e.g., improvements to the circuit structure of diodes, transistors, switches, etc.) or software improvements (improvements to the methodology). However, with technological advancements, many methodological improvements today can be considered direct improvements to the hardware circuit structure. Designers almost always obtain the corresponding hardware circuit structure by programming the improved methodology into the hardware circuit. Therefore, it cannot be said that a methodological improvement cannot be implemented using hardware physical modules. For example, a Programmable Logic Device (PLD) (such as a Field Programmable Gate Array (FPGA)) is such an integrated circuit whose logic function is determined by the user programming the device. Designers can program and "integrate" a digital system onto a PLD themselves, without needing chip manufacturers to design and manufacture dedicated integrated circuit chips. Furthermore, nowadays, instead of manually manufacturing integrated circuit chips, this programming is mostly implemented using "logic compiler" software. Similar to the software compiler used in program development, the original code before compilation must be written in a specific programming language, called a Hardware Description Language (HDL). There are many HDLs, such as ABEL (Advanced Boolean Expression Language), AHDL (Altera Hardware Description Language), Confluence, CUPL (Cornell University Programming Language), HDCal, JHDL (Java Hardware Description Language), Lava, Lola, MyHDL, PALASM, and RHDL (Ruby Hardware Description Language). Currently, VHDL (Very-High-Speed ​​Integrated Circuit Hardware Description Language) and Verilog are the most commonly used. Those skilled in the art should also understand that by simply performing some logic programming on the method flow using one of these hardware description languages ​​and programming it into an integrated circuit, the hardware circuit implementing the logical method flow can be easily obtained.

[0094] The controller can be implemented in any suitable manner. For example, it can take the form of a microprocessor or processor and a computer-readable medium storing computer-readable program code (e.g., software or firmware) executable by the (micro)processor, logic gates, switches, application-specific integrated circuits (ASICs), programmable logic controllers, and embedded microcontrollers. Examples of controllers include, but are not limited to, the following microcontrollers: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20, and Silicon Labs C8051F320. A memory controller can also be implemented as part of the control logic of the memory. Those skilled in the art will also recognize that, in addition to implementing the controller in purely computer-readable program code form, the same functionality can be achieved by logically programming the method steps to make the controller take the form of 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 means included therein for implementing various functions can also be considered as structures within the hardware component. Alternatively, the means for implementing various functions can be considered as both software modules implementing the method and structures within the hardware component.

[0095] The systems, devices, modules, or units described in the above embodiments can be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, a computer can be, for example, a personal computer, laptop computer, cellular phone, camera phone, smartphone, personal digital assistant, media player, navigation device, email device, game console, tablet computer, wearable device, or any combination of these devices.

[0096] For ease of description, the above devices are described in terms of function, divided into various units. Of course, in implementing this specification, the functions of each unit can be implemented in one or more software and / or hardware components.

[0097] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0098] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0099] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0100] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0101] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0102] Memory may include non-persistent storage in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0103] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0104] It should also be noted that 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 process, method, article, or apparatus. Unless otherwise specified, 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 that element.

[0105] Those skilled in the art will understand that the embodiments of this specification can be provided as methods, systems, or computer program products. Therefore, this specification may take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this specification may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0106] This specification can be described in the general context of computer-executable instructions that are executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform a specific task or implement a specific abstract data type. This specification can also be practiced in distributed computing environments, where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0107] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the system embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments.

[0108] The above are merely embodiments of this specification and are not intended to limit this specification. Various modifications and variations can be made to this specification by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this specification should be included within the scope of the claims of this application.

Claims

1. A training method for a detection result correction model applied to radiotherapy, characterized in that, include: During the process of irradiating multiple detectors with radiation light n times, the detector parameters and radiation dose values ​​are collected after each irradiation. The detector parameters include breakdown voltage and / or series resistance and / or quenching resistance and / or parasitic resistance and / or pixel capacitive reactance and / or discharge time constant and / or gain and / or dark count rate and / or photon detection efficiency; the irradiation dose value includes the standard dose value and the dose measurement value collected by the detector. Based on the multiple irradiation dose values, determine the correction factor data; Data preprocessing is performed on multiple detector parameters and dose measurements to determine training samples; The correction coefficient data and the training samples are input into the detection result correction model, and the detection result correction model is trained until the loss function of the detection result correction model converges.

2. The training method for the detection result correction model applied to radiotherapy according to claim 1, characterized in that, Before subjecting the multiple detectors to n illuminations respectively, the method further includes: For any given detector, determine the maximum radiation dose corresponding to that detector based on its usage scenario; Based on the maximum radiation dose corresponding to the detector, determine the single radiation dose when the detector is irradiated n times.

3. The training method for the detection result correction model applied to radiotherapy according to claim 1, characterized in that, The step of determining the correction coefficient data based on the plurality of irradiation dose values ​​includes: For any given irradiation dose value, the correction factor value corresponding to that irradiation dose value is determined using the following formula; in, This is the correction coefficient value corresponding to the i-th detector after being illuminated for the nth time; This represents the standard dose value corresponding to the i-th detector after being irradiated for the nth time; This represents the dose measurement value collected by the i-th detector after it has been irradiated for the nth time. Based on the multiple correction coefficient values, the correction coefficient data is determined.

4. The training method for the detection result correction model applied to radiotherapy according to claim 1, characterized in that, Data preprocessing was performed on multiple detector parameters and dose measurements to determine training samples, including: The correlation between detector parameters and dose measurements was established using the least squares method; the correlation is as follows: in, The breakdown voltage; The series resistor; The quenching resistance; The parasitic resistance; The pixel capacitance; G is the discharge time constant; DCR is the dark count rate; PDE is the photon detection efficiency; and D is the dose measurement value. The target algorithm is used to solve the above-mentioned correlations and determine the feature vector matrix; Training samples are determined based on the feature vector, the multiple detector parameters, and the dose measurement values.

5. The training method for the detection result correction model applied to radiotherapy according to claim 4, characterized in that, The number of detectors is i; Both D and are i×n matrices, and the data at the same position correspond to each other; The step of solving the aforementioned correlations using a target algorithm to determine training samples includes: Through feature extraction methods Dimensionality reduction is performed by using principal component analysis to solve for the eigenvectors of the sample space covariance matrix; the sample space is... The element in the s-th row and t-th column of the covariance matrix is ​​defined as follows: in, For the sample space, Let f be the f-th sample; any sample is a flattened i×10 dimension vector. It is the covariance matrix; This represents the value of the f-th sample in the s-th dimension; The average value of the samples in the s-th dimension; Perform eigenvalue decomposition on the covariance matrix: in, Ω is a diagonal matrix, where each diagonal element is an eigenvalue of the covariance matrix M and they are all distinct; Ω is the eigenvector matrix; each column of Ω is the eigenvector corresponding to the covariance matrix. Training samples are determined based on the feature vector, the multiple detector parameters, and the dose measurement values.

6. The training method for the detection result correction model applied to radiotherapy according to claim 5, characterized in that, It can be determined by the following formula: in, This represents the value of the f-th sample in the s-th dimension.

7. A method for correcting detection results applied to radiotherapy, characterized in that, A detection result correction model for radiotherapy trained using any one of claims 1-6, comprising: Acquire the dose measurements collected by the detector, as well as the breakdown voltage and / or series resistance and / or quenching resistance and / or parasitic resistance and / or pixel capacitive reactance and / or discharge time constant and / or gain and / or dark count rate and / or photon detection efficiency; The dose measurement value, as well as the breakdown voltage and / or series resistance and / or quenching resistance and / or parasitic resistance and / or pixel capacitance and / or discharge time constant and / or gain and / or dark count rate and / or photon detection efficiency, are input into the detection result correction model applied to radiotherapy to determine the corrected dose value output by the detection result correction model applied to radiotherapy.

8. A training device for a detection result correction model applied to radiotherapy, characterized in that, include: The acquisition unit is used to acquire detector parameters and radiation dose values ​​after each irradiation during the process of irradiating multiple detectors with radiation light n times. The detector parameters include breakdown voltage and / or series resistance and / or quenching resistance and / or parasitic resistance and / or pixel capacitance and / or discharge time constant and / or gain and / or dark count rate and / or photon detection efficiency. The radiation dose values ​​include standard dose values ​​and dose measurements acquired by the detectors. A determining unit is used to determine correction coefficient data based on the plurality of irradiation dose values; The processing unit is used to preprocess multiple detector parameters and dose measurements to determine training samples. The training unit is used to input the correction coefficient data and the feature vector into the detection result correction model, and train the detection result correction model until the loss function of the detection result correction model converges.

9. A detection result correction device for use in radiotherapy, characterized in that, include: The acquisition unit is used to acquire the dose measurement value collected by the detector, as well as the breakdown voltage and / or series resistance and / or quenching resistance and / or parasitic resistance and / or pixel capacitive reactance and / or discharge time constant and / or gain and / or dark count rate and / or photon detection efficiency. A correction unit is configured to input the dose measurement value, as well as the breakdown voltage and / or series resistance and / or quenching resistance and / or parasitic resistance and / or pixel capacitance and / or discharge time constant and / or gain and / or dark count rate and / or photon detection efficiency, into the detection result correction model applied to radiotherapy, and determine the corrected dose value output by the detection result correction model applied to radiotherapy; the detection result correction model is the detection result correction model applied to radiotherapy trained according to any one of claims 1-6.

10. A computer-readable storage medium, characterized in that, The storage medium stores a computer program, which, when executed by a processor, implements the method described in any one of claims 1-7.

11. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the method described in any one of claims 1-7.

12. A computer program product, characterized in that, When the instructions in the computer program product are executed by the processor of the electronic device, the electronic device causes the electronic device to perform the method as described in any one of claims 1-7.

Images