Medical electron linear accelerator mechanical parameter verification method and system

Abstract

The invention belongs to the technical field of parameter verification, and discloses a medical electron linear accelerator mechanical parameter verification method and system, and the method comprises the steps: obtaining a log file of the single use of a linear accelerator, extracting the actually executed mechanical parameters, and forming an actual execution sequence set; obtaining a plan file, extracting expected mechanical parameters, and forming an expected sequence set; obtaining a historical use record, and generating a self-adaptive tolerance standard; judging whether each mechanical parameter is qualified or not by applying a self-adaptive tolerance standard; calculating an actual execution dose distribution when at least one mechanical parameter is unqualified; and comparing the actually executed dose distribution difference with the original planned dose distribution difference, and generating a dose deviation evaluation report. Sensitivity differences of different treatment parts on various mechanical parameter deviations are considered through a self-adaptive tolerance standard, full-chain analysis from the mechanical parameter deviations to dose influences is achieved, and the accuracy and effectiveness of mechanical parameter verification are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of parameter verification, and more specifically, to a method and system for verifying mechanical parameters of a medical electron linear accelerator. Background Art

[0002] With the rapid development of radiotherapy technology, linear accelerators have become the core equipment for modern tumor radiotherapy. Modern radiotherapy techniques such as intensity-modulated radiotherapy and volumetric-modulated arc therapy can achieve highly conformal dose distributions, improving the dose coverage of the target area while reducing the dose to surrounding normal tissues. However, the treatment effects of these advanced techniques highly depend on the precise execution of the mechanical parameters of the linear accelerator. Minor deviations in the mechanical parameters may lead to significant differences between the actual dose distribution and the plan, affecting the treatment quality.

[0003] Existing mechanical parameter verification methods have limitations in many aspects. First, the verification results usually only use whether the parameters are within the allowable thresholds as the evaluation criteria, lacking a quantitative analysis of how these deviations affect the actual dose distribution. Second, when abnormal dose distributions are found, it is difficult to trace back to the specific sources of mechanical parameter deviations. This disconnection between mechanical parameters and dose effects makes it difficult for operators to intuitively understand the actual significance of parameter deviations, which may lead to overreactions or improper neglect of the deviations, affecting the accuracy and effectiveness of the verification work. In addition, existing mechanical parameter verification methods mostly adopt fixed tolerance standards and fail to consider the sensitivity differences of different treatment sites to deviations in various mechanical parameters. For example, head and neck treatments usually have higher requirements for the accuracy of the multi-leaf collimator position, while pelvic treatments may be more sensitive to deviations in the gantry angle. Using a unified tolerance standard may result in overly wide tolerances for some key parameters and overly strict tolerances for non-critical parameters, unable to specifically guarantee the treatment quality. Summary of the Invention

[0004] In order to overcome the problem that the prior art cannot specifically guarantee the treatment quality, the present invention proposes a method and system for verifying mechanical parameters of a medical electron linear accelerator to solve the above problems.

[0005] The present invention provides the following technical solutions:

[0006] A method for verifying mechanical parameters of a medical electron linear accelerator, comprising:

[0007] Obtaining a log file of a single use of the linear accelerator, extracting the actually executed mechanical parameters from the log file, and forming an actual execution sequence set according to the time stamp sequence in the log file;

[0008] Obtaining a plan file corresponding to the single use, extracting the expected mechanical parameters and their timing distributions of each control point from the plan file, and forming an expected sequence set;

[0009] Obtain historical usage records and generate an adaptive tolerance standard based on the historical usage records;

[0010] Compare each parameter sequence in the actual execution sequence set with the corresponding parameter sequence in the expected sequence set, and apply the adaptive tolerance standard to determine whether each mechanical parameter is qualified;

[0011] When at least one mechanical parameter is unqualified, calculate the actual execution dose distribution based on the actual execution sequence set;

[0012] Obtain the original planned dose distribution from the plan file, compare the difference between the actual execution dose distribution and the original planned dose distribution, and generate a dose deviation assessment report.

[0013] Preferably, the mechanical parameters include gantry angle, collimator angle, tungsten gate position, multi-leaf collimator position, and monitor units.

[0014] Preferably, the generating an adaptive tolerance standard based on the historical usage records includes:

[0015] Read the mechanical parameter log file and the corresponding dose measurement data in the historical usage records; screen out the treatment records with mechanical parameter deviations and completed dose verification;

[0016] Classify the mechanical parameter deviation values and the dose deviation values according to the treatment site, extract the deviation sample set of each mechanical parameter and the corresponding dose deviation sample set, and perform standardization processing;

[0017] For each mechanical parameter, calculate the dose change amount caused by its unit deviation;

[0018] Adopt the multiple linear regression method to calculate the partial derivative values of each mechanical parameter on the dose distribution;

[0019] Organize the partial derivative values of each treatment site and each mechanical parameter into a matrix form to obtain a parameter sensitivity matrix;

[0020] Generate an adaptive tolerance table according to the preset reference tolerance value and the parameter sensitivity matrix, and use the adaptive tolerance table as the adaptive tolerance standard.

[0021] Preferably, the applying the adaptive tolerance standard to determine whether each mechanical parameter is qualified includes:

[0022] For each mechanical parameter, perform the following judgment steps:

[0023] Obtain the actual execution sequence and the expected sequence of the mechanical parameter, and judge the overall consistency between the actual execution sequence and the expected sequence;

[0024] If the overall consistency judgment result is inconsistent, it is determined that the mechanical parameter is unqualified;

[0025] If the overall consistency judgment result is consistent, time-align the actual execution sequence with the expected sequence;

[0026] Calculate the differences at each time point of the aligned sequences to obtain the deviation sequence of the mechanical parameter;

[0027] Query the tolerance threshold corresponding to the mechanical parameter from the adaptive tolerance table;

[0028] Statistically calculate the proportion of the number of points exceeding the tolerance threshold to the total number of points;

[0029] When the proportion exceeds the preset allowable proportion of the mechanical parameter, it is determined that the mechanical parameter sequence is unqualified; otherwise, it is determined that the mechanical parameter is qualified.

[0030] Preferably, the judgment of the overall consistency between the actual execution sequence and the expected sequence includes:

[0031] Calculate the difference between the extreme value of the actual execution sequence and the extreme value of the expected sequence;

[0032] Calculate the difference between the mean value of the actual execution sequence and the mean value of the expected sequence;

[0033] Calculate the difference between the standard deviation of the actual execution sequence and the standard deviation of the expected sequence;

[0034] Judge whether the differences between the extreme values, the differences between the mean values, and the differences between the standard deviations are respectively within the corresponding preset thresholds. If any of the differences exceeds the corresponding preset threshold, the judgment result is inconsistent; otherwise, the judgment result is consistent.

[0035] Preferably, the time-alignment of the actual execution sequence with the expected sequence includes:

[0036] Set the first point of the actual execution sequence and the first point of the expected sequence as the time reference origins of their respective sequences;

[0037] Based on each point in the expected sequence, search for the point corresponding in time in the actual execution sequence;

[0038] If there is a point in the actual execution sequence that exactly corresponds in time to a point in the expected sequence, directly extract the parameter value of that point;

[0039] If there is no point in the actual execution sequence that exactly corresponds in time to a point in the expected sequence, establish an insertion point, determine the reference point of the insertion point, calculate the parameter value of the insertion point based on the parameter value of the reference point, and insert the insertion point into the actual execution sequence at the corresponding time;

[0040] Delete the points in the actual execution sequence that do not correspond to the time in the expected sequence to complete the time alignment between the actual execution sequence and the expected sequence.

[0041] Preferably, the reference points for determining the insertion points include:

[0042] Find the time point in the actual execution sequence with the time value closest to the time of this point in the expected sequence as the nearest reference point;

[0043] Find the point before the nearest reference point in the actual execution sequence as the previous reference point;

[0044] Find the point after the nearest reference point in the actual execution sequence as the next reference point;

[0045] Determine the nearest reference point, the previous reference point, and the next reference point as the reference points for the insertion point;

[0046] The calculation of the parameter value of the insertion point based on the parameter values of the reference points includes:

[0047] Obtain the parameter values of the previous reference point, the nearest reference point, and the next reference point;

[0048] According to the parameter values of the three reference points, calculate the parameter value corresponding to the time of this point in the expected sequence by weighted summation;

[0049] Among them, the sum of the weight coefficients of the three reference points is 1, and it is set based on the time relationship between the three reference points and the insertion point.

[0050] Preferably, the calculation of the actual execution dose distribution based on the actual execution sequence set includes:

[0051] Obtain the patient anatomical structure information from the plan file, including the contour data of the planned target volume, clinical target volume, total tumor volume, and each organ at risk;

[0052] Based on the time stamps in the actual execution sequence set, discretize the continuous mechanical parameter sequence into a finite number of irradiation state points;

[0053] For each discrete irradiation state point, construct the corresponding irradiation geometric configuration according to the mechanical parameters;

[0054] Adopt the convolution / superposition algorithm to calculate the energy deposition distribution of each irradiation state point;

[0055] Weight the dose contribution according to the time proportion of each irradiation state point and accumulate to obtain the complete three-dimensional dose distribution matrix;

[0056] The three-dimensional dose distribution matrix includes the planned target volume, clinical target volume, total tumor volume, and dose values, isodose curves, dose volume histograms, and dose statistical parameters of each voxel point in each organ at risk;

[0057] The dose statistical parameters include the maximum dose, minimum dose, average dose, median dose, D95, D98, D2, homogeneity index, and conformity index of the target volume, as well as the maximum dose, average dose, and specific volume dose index of the organ at risk.

[0058] Preferably, comparing the difference between the actually executed dose distribution and the original planned dose distribution to generate a dose deviation assessment report includes:

[0059] Comparing the dose values of each voxel point in the actually executed dose distribution and the original planned dose distribution, and calculating the absolute value and relative percentage of the voxel point dose difference;

[0060] Comparing the isodose curves of the actually executed dose distribution and the original planned dose distribution to determine the isodose curve displacement;

[0061] Comparing the dose volume histograms of the actually executed dose distribution and the original planned dose distribution, and calculating the dose coverage deviation of the planned target volume, clinical target volume, total tumor volume, and each organ at risk;

[0062] Comparing the dose statistical parameters of the actually executed dose distribution and the original planned dose distribution, and calculating the deviation values of each parameter;

[0063] According to the preset tolerance standard, determining the acceptability of each deviation, and generating a deviation assessment report including the above deviation quantification indexes and acceptability assessment conclusions.

[0064] The present invention also provides a mechanical parameter verification system for a medical electron linear accelerator, which is used to implement a mechanical parameter verification method for a medical electron linear accelerator, including:

[0065] A log parsing module, which is used to obtain the log file of a single use of the linear accelerator, extract the actually executed mechanical parameters from the log file, and form an actually executed sequence set according to the time stamp sequence in the log file;

[0066] A plan parsing module, which is used to obtain the plan file corresponding to a single use, extract the expected mechanical parameters and their timing distributions of each control point from the plan file, and form an expected sequence set;

[0067] A tolerance generation module, which is used to obtain the historical usage records and generate an adaptive tolerance standard according to the historical usage records;

[0068] A parameter comparison module, which is used to compare each parameter sequence in the actual execution sequence set with the corresponding parameter sequence in the expected sequence set, and apply an adaptive tolerance standard to judge whether each mechanical parameter is qualified;

[0069] A dose reconstruction module, which is used to calculate the actual executed dose distribution based on the actual execution sequence set when at least one mechanical parameter is unqualified;

[0070] An evaluation report module, which is used to obtain the original planned dose distribution from the plan file, compare the difference between the actual executed dose distribution and the original planned dose distribution, and generate a dose deviation evaluation report.

[0071] The present invention provides a method and system for verifying mechanical parameters of a medical electron linear accelerator, having the following beneficial effects:

[0072] By analyzing the correlation between mechanical parameter deviations and dose deviations in historical usage records, an adaptive tolerance standard for different treatment sites is generated. This adaptive tolerance standard based on actual data fully considers the sensitivity differences of different treatment sites to mechanical parameter deviations, enabling the tolerance requirements to ensure treatment quality while avoiding unnecessary strict restrictions.

[0073] By converting the unqualified mechanical parameters into the actually executed irradiation geometric configuration, using the convolution / superposition algorithm to calculate the actual executed dose distribution, and then performing multi-dimensional comparison with the original planned dose distribution, the operator can intuitively understand the actual impact of mechanical parameter deviations on the dose distribution. This method breaks through the limitation in traditional verification methods where the mechanical parameters are disconnected from the dose impact, making the verification result no longer limited to a simple judgment of whether the parameters are within the allowable threshold, but providing a quantitative analysis of the impact of parameter deviations on the dose distribution.

[0074] By generating a detailed dose deviation evaluation report, including various quantitative indicators such as voxel point dose difference, isodose curve displacement, dose volume histogram deviation, and dose statistical parameter changes, the operator can comprehensively understand the clinical significance of mechanical parameter deviations. This evaluation method based on dose impact avoids overreaction or improper neglect of mechanical parameter deviations, improving the effectiveness of the verification work and the rationality of decision-making. Brief Description of the Drawings

[0075] Figure 1 It is a schematic flow chart of a method for verifying mechanical parameters of a medical electron linear accelerator according to the present invention;

[0076] Figure 2 It is a schematic module diagram of a system for verifying mechanical parameters of a medical electron linear accelerator according to the present invention. Detailed Embodiment

[0077] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0078] Embodiment 1

[0079] Please refer to Figure 1 , in this embodiment, a method for verifying mechanical parameters of a medical electron linear accelerator includes:

[0080] S1. Obtain the log file of a single use of the linear accelerator, extract the actually executed mechanical parameters from the log file, and form an actual execution sequence set according to the time stamp sequence in the log file;

[0081] The mechanical parameters include gantry angle, collimator angle, tungsten gate position, multi-leaf collimator position, and monitor units.

[0082] In this embodiment, first, export the treatment log file from the linear accelerator control system. This file records the real-time status and motion parameters of each mechanical component during the treatment process. Next, parse the log file and extract the key mechanical parameter data. The specifically extracted parameters include gantry angle, collimator angle, tungsten gate position, multi-leaf collimator position, and monitor units. The gantry angle records the angle value of the gantry rotation during the treatment process; the collimator angle records the angle value of the collimator rotation; the tungsten gate position records the position value of the tungsten gate; the multi-leaf collimator position records the position values of each leaf; the monitor units record the irradiation dose unit value.

[0083] At the same time, extract the time stamp information corresponding to each parameter recording point. The time stamp records the exact time of each control point. According to the extracted time stamp sequence and the corresponding mechanical parameter values, form the actual execution sequences of each mechanical parameter. Each mechanical parameter corresponds to a sequence, and all the sequences together form the actual execution sequence set, which reflects the actual motion trajectories of each mechanical component during use and provides basic data for subsequent comparative analysis with the planned parameters.

[0084] S2. Obtain the plan file corresponding to the single use, extract the expected mechanical parameters and their timing distributions of each control point from the plan file, and form an expected sequence set;

[0085] In this embodiment, first, obtain a plan file corresponding to a single use of a linear accelerator. This plan file is usually generated by a radiotherapy planning system (TPS) and contains detailed information on the treatment plan and the expected mechanical parameter settings for each control point. Next, parse the plan file and extract the expected mechanical parameter data for each control point. The specifically extracted parameters include gantry angle, collimator angle, tungsten gate position, multi-leaf collimator position, and monitor units. The gantry angle specifies the angle value by which the gantry should rotate during the treatment; the collimator angle specifies the angle value by which the collimator should rotate; the tungsten gate position specifies the expected position value of the tungsten gate; the multi-leaf collimator position specifies the expected position values of each leaf; and the monitor units specify the dose unit values to be irradiated at each control point.

[0086] At the same time, extract the timing distribution information corresponding to each control point. The timing distribution records the expected time points or time intervals of each control point during the treatment. According to the extracted timing distribution and the corresponding expected mechanical parameter values, form the expected sequences for each mechanical parameter respectively. Each mechanical parameter corresponds to an expected sequence, and all the expected sequences together form an expected sequence set, which reflects the expected movement trajectories of the mechanical components in the plan and provides reference data for the comparative analysis with the actual execution sequence set.

[0087] S3. Obtain the historical usage records and generate an adaptive tolerance standard according to the historical usage records;

[0088] Generating an adaptive tolerance standard according to the historical usage records includes:

[0089] Read the mechanical parameter log file and the corresponding dose measurement data in the historical usage records; screen out the treatment records with mechanical parameter deviations and completed dose verification;

[0090] Classify the mechanical parameter deviation values and the dose deviation values according to the treatment sites, extract the deviation sample sets of each mechanical parameter and the corresponding dose deviation sample sets, and perform standardization processing;

[0091] For each mechanical parameter, calculate the dose change amount caused by its unit deviation;

[0092] Adopt the multiple linear regression method to calculate the partial derivative values of each mechanical parameter on the dose distribution;

[0093] Organize the partial derivative values of each treatment site and each mechanical parameter into a matrix form to obtain a parameter sensitivity matrix;

[0094] Generate an adaptive tolerance table according to the preset reference tolerance value and the parameter sensitivity matrix, and use the adaptive tolerance table as the adaptive tolerance standard.

[0095] In this embodiment, first, the mechanical parameter log file and the corresponding dose measurement data in the historical usage records are read. The historical usage records contain the usage conditions of the linear accelerator over a past period of time, and each record includes the mechanical parameter log file, the plan file, and the corresponding dose measurement results.

[0096] Next, the treatment records with mechanical parameter deviations and completed dose verification are screened. Through screening, a set of representative samples are obtained. These samples contain both mechanical parameter deviation information and the corresponding dose deviation information, and can be used to establish the correlation between the two.

[0097] Then, the screened records are classified according to the treatment site (such as head and neck, chest, pelvis, etc.). The purpose of classification is to consider the sensitivity differences of different sites to mechanical parameter deviations.

[0098] For each classification, the deviation sample set of each mechanical parameter (gantry angle, collimator angle, tungsten gate position, multi-leaf collimator position, and monitor units) and the corresponding dose deviation sample set are extracted. For example, for the head and neck treatment classification, there may be 50 records, from which the set of gantry angle deviation values and the corresponding set of dose deviation values are extracted. The dose deviation values can include various forms, such as the D95 deviation of the target area, the maximum dose deviation, the decrease value of the gamma analysis passing rate, etc.

[0099] To make different parameters and dose indicators comparable, the extracted sample sets are standardized. The standardization process makes the mean of the processed sample set 0 and the standard deviation 1. For example, for the gantry angle deviation set, its mean and standard deviation are calculated, and then each deviation value is standardized. The same method is also applied to the dose deviation sample set.

[0100] After the standardization process is completed, for each mechanical parameter, the dose change amount caused by its unit deviation is calculated. This step aims to quantify the sensitivity relationship between mechanical parameter deviations and dose deviations. The calculation method is to perform a regression analysis on the standardized mechanical parameter deviations and the corresponding dose deviation samples to obtain the linear relationship coefficient between the two.

[0101] Then, the multiple linear regression method is adopted to calculate the partial derivative values of each mechanical parameter on the dose distribution. Multiple linear regression considers the situation of simultaneous deviations of multiple mechanical parameters and can more accurately reflect the interactive effects between parameters. Through regression analysis, the partial derivative values are obtained, indicating the dose change amount caused by the unit deviation of each mechanical parameter.

[0102] For the classification of each treatment site, repeat the above regression analysis process to obtain the corresponding partial derivative values. Organize the partial derivative values of each treatment site and each mechanical parameter into a matrix form to obtain the parameter sensitivity matrix. For example, each row of the matrix represents a treatment site, and each column represents a mechanical parameter (in order: gantry angle, collimator angle, tungsten gate position, multi-leaf collimator position, and monitor units). Each element in the matrix represents the dose sensitivity coefficient of the corresponding parameter for the corresponding site.

[0103] Finally, generate an adaptive tolerance table based on the preset reference tolerance value and the parameter sensitivity matrix. The preset reference tolerance value is the basic tolerance requirement set based on industry standards or institutional experience, such as gantry angle ±1 degree, collimator angle ±1 degree, tungsten gate position ±2 mm, multi-leaf collimator position ±2 mm, and monitor units ±2 MU. The adaptive tolerance table is obtained by adjusting the reference tolerance value, and the adjustment is based on the sensitivity coefficients of each parameter for each site. The higher the sensitivity coefficient, the greater the impact of the parameter on the dose, and the more stringent the tolerance requirement should be; the lower the sensitivity coefficient, the smaller the impact of the parameter on the dose, and a relatively loose tolerance requirement can be adopted.

[0104] The generated adaptive tolerance table contains the adaptive tolerance values of the five mechanical parameters (gantry angle, collimator angle, tungsten gate position, multi-leaf collimator position, and monitor units) corresponding to each treatment site.

[0105] The adaptive tolerance table generated in this way will be used as the adaptive tolerance standard for subsequent judgment of the eligibility of mechanical parameters. When the system needs to judge whether the mechanical parameters of a certain treatment are eligible, first identify its treatment site, then look up the corresponding tolerance values in the adaptive tolerance table, and use these tolerance values to replace the fixed preset threshold for judgment.

[0106] In this way, the system can provide customized tolerance standards for different treatment sites based on historical data and actual dose effects, which not only ensures the dose accuracy requirements but also avoids unnecessary strict restrictions, improving the scientificity and rationality of mechanical parameter verification.

[0107] S4. Compare each parameter sequence in the actual execution sequence set with the corresponding parameter sequence in the expected sequence set, and apply the adaptive tolerance standard to judge whether each mechanical parameter is eligible;

[0108] Applying the adaptive tolerance standard to judge whether each mechanical parameter is eligible includes:

[0109] For each type of mechanical parameter, perform the following judgment steps:

[0110] Obtain the actual execution sequence and the expected sequence of this mechanical parameter, and judge the overall consistency between the actual execution sequence and the expected sequence;

[0111] If the overall consistency judgment result is inconsistent, it is determined that the mechanical parameter is unqualified;

[0112] If the overall consistency judgment result is consistent, time-align the actual execution sequence with the expected sequence;

[0113] Calculate the differences at each time point of the aligned sequences to obtain the deviation sequence of the mechanical parameter;

[0114] Obtain the preset deviation tolerance threshold corresponding to the mechanical parameter, and count the proportion of the number of points exceeding the preset deviation tolerance threshold to the total number of points;

[0115] When the proportion exceeds the preset allowable proportion of the mechanical parameter, it is determined that the mechanical parameter sequence is unqualified, otherwise it is determined that the mechanical parameter is qualified.

[0116] Judging the overall consistency between the actual execution sequence and the expected sequence includes:

[0117] Calculate the difference between the extreme value of the actual execution sequence and the extreme value of the expected sequence;

[0118] Calculate the difference between the mean value of the actual execution sequence and the mean value of the expected sequence;

[0119] Calculate the difference between the standard deviation of the actual execution sequence and the standard deviation of the expected sequence;

[0120] Judge whether the differences between the extreme values, the differences between the mean values, and the differences between the standard deviations are respectively within the corresponding preset thresholds. If any of the differences exceeds the corresponding preset threshold, the judgment result is inconsistent; otherwise, the judgment result is consistent.

[0121] Time-aligning the actual execution sequence with the expected sequence includes:

[0122] Set the first point of the actual execution sequence and the first point of the expected sequence as the time reference origins of their respective sequences;

[0123] Based on each point in the expected sequence, find the point corresponding in time in the actual execution sequence;

[0124] If there is a point in the actual execution sequence whose time exactly corresponds to a point in the expected sequence, directly extract the parameter value of that point;

[0125] If there is no point in the actual execution sequence whose time exactly corresponds to a point in the expected sequence, establish an insertion point, determine the reference point of the insertion point, calculate the parameter value of the insertion point based on the parameter value of the reference point, and insert the insertion point into the actual execution sequence at the corresponding time;

[0126] Delete the points in the actual execution sequence that do not correspond in time to the points in the expected sequence to complete the time alignment between the actual execution sequence and the expected sequence.

[0127] The reference points for determining the insertion point include:

[0128] Find the time point in the actual execution sequence with the time value closest to the time at this point in the expected sequence as the nearest reference point;

[0129] Find the point before the nearest reference point in the actual execution sequence as the previous reference point;

[0130] Find the point after the nearest reference point in the actual execution sequence as the next reference point;

[0131] Determine the nearest reference point, the previous reference point, and the next reference point as the reference points for the insertion point;

[0132] Calculating the parameter value of the insertion point based on the parameter values of the reference points includes:

[0133] Obtain the parameter values of the previous reference point, the nearest reference point, and the next reference point;

[0134] According to the parameter values of the three reference points, calculate the parameter value corresponding to the time at this point in the expected sequence through weighted summation;

[0135] Among them, the sum of the weight coefficients of the three reference points is 1, and it is set based on the time relationship between the three reference points and the insertion point.

[0136] In this embodiment, first, the obtained actual execution sequence set and the expected sequence set are compared and analyzed to determine whether each mechanical parameter is qualified. The comparative analysis is carried out separately for each mechanical parameter such as the gantry angle, collimator angle, tungsten gate position, multi-leaf collimator position, and monitor units.

[0137] For each mechanical parameter, first obtain its actual execution sequence and the expected sequence, and then judge the overall consistency between the two sequences. The overall consistency judgment is achieved by comparing the statistical characteristics of the sequences. Specifically, calculate the difference between the range value (maximum value minus minimum value) of the actual execution sequence and the range value of the expected sequence, calculate the difference between the mean value of the actual execution sequence and the mean value of the expected sequence, and calculate the difference between the standard deviation of the actual execution sequence and the standard deviation of the expected sequence. For example, if the range of the actual execution sequence of the gantry angle is 182 degrees and the range of the expected sequence is 180 degrees, then the range difference is 2 degrees; if the mean value of the actual execution sequence is 90 degrees and the mean value of the expected sequence is 90.5 degrees, then the mean difference is 0.5 degrees.

[0138] Next, determine whether these differences are within their respective preset thresholds. For example, for the rack angle, the range difference threshold can be set to 5 degrees, the mean difference threshold can be set to 2 degrees, and the standard deviation difference threshold can be set to 3 degrees. If all three differences are within the thresholds, the overall consistency result is determined to be consistent; if any difference exceeds the threshold, it is determined to be inconsistent. If the determination result is inconsistent, the mechanical parameter is directly determined to be unqualified.

[0139] If the overall consistency is determined to be consistent, the next step is to perform time alignment between the actual execution sequence and the expected sequence. Time alignment first sets the first points of the two sequences as the time reference origins respectively. Then, based on each point in the expected sequence, find the point with the corresponding time in the actual execution sequence. For example, if the time of a certain point in the expected sequence is 10 seconds, find the point with a time of 10 seconds in the actual execution sequence.

[0140] If there is a point in the actual execution sequence with exactly the same time as a certain point in the expected sequence, directly extract the parameter value of this point for subsequent comparison. If there is no exactly corresponding point, an interpolation point needs to be established. When establishing the interpolation point, first find the point in the actual execution sequence with the time value closest to the time of this point in the expected sequence as the nearest reference point, and then find the previous point and the next point of this reference point as the previous reference point and the next reference point respectively. For example, if the time of a certain point in the expected sequence is 15.5 seconds, and there are two points of 15.3 seconds and 15.8 seconds in the actual execution sequence, the point of 15.3 seconds is the nearest reference point, its previous point (such as 15.0 seconds) is the previous reference point, and its next point (such as 15.8 seconds) is the next reference point.

[0141] Then, based on the parameter values of these three reference points, calculate the parameter value of the interpolation point through weighted summation. The sum of the weight coefficients is 1 and is set based on the time relationship between the three reference points and the interpolation point. For example, the reciprocal of the time distance can be used as the weight basis, and the reference point closer in time has a greater weight. After the calculation is completed, insert the interpolation point into the actual execution sequence according to the corresponding time, and delete the points in the actual execution sequence that do not correspond to the time in the expected sequence to complete the time alignment.

[0142] After the time alignment is completed, calculate the differences at each time point of the aligned sequence to obtain the deviation sequence of the mechanical parameters. For example, if the actual rack angle at a certain time point after alignment is 91.2 degrees and the expected rack angle is 90.0 degrees, the deviation at this time point is 1.2 degrees.

[0143] Next, identify the current treatment site (such as the head and neck, chest, or pelvis, etc.), and then look up the corresponding tolerance value from the adaptive tolerance table. For example, the deviation tolerance threshold for the gantry angle is 1 degree. Calculate the proportion of the number of points exceeding the preset deviation tolerance threshold in the deviation sequence to the total number of points. For example, if there are 100 points in the deviation sequence and 12 of them exceed 1 degree, the exceeding proportion is 12%.

[0144] Finally, compare this proportion with the preset allowable proportion of the mechanical parameter (such as 10%). If the exceeding proportion is greater than the preset allowable proportion, it is determined that the mechanical parameter is unqualified; otherwise, it is determined to be qualified. In this way, the qualification of all mechanical parameters is judged, providing a basis for subsequent possible dose deviation assessment.

[0145] S5. When at least one mechanical parameter is unqualified, calculate the actual executed dose distribution based on the actual execution sequence set;

[0146] Calculating the actual executed dose distribution based on the actual execution sequence set includes:

[0147] Obtain the patient anatomical structure information from the planning file, including the planned target volume, clinical target volume, total tumor volume, and contour data of each organ at risk;

[0148] Based on the timestamps in the actual execution sequence set, discretize the continuous mechanical parameter sequence into a finite number of irradiation status points;

[0149] For each discrete irradiation status point, construct the corresponding irradiation geometric configuration according to the mechanical parameters;

[0150] Adopt the convolution / superposition algorithm to calculate the energy deposition distribution of each irradiation status point;

[0151] Weight the dose contributions according to the time proportion of each irradiation status point and accumulate to obtain a complete three-dimensional dose distribution matrix;

[0152] The three-dimensional dose distribution matrix includes dose curves, dose volume histograms, and dose statistical parameters such as dose values of each voxel point in the planned target volume, clinical target volume, total tumor volume, and each organ at risk;

[0153] The dose statistical parameters include the maximum dose, minimum dose, average dose, median dose, D95, D98, D2, homogeneity index, and conformity index of the target volume, as well as the maximum dose, average dose, and specific volume dose index of the organ at risk.

[0154] In this embodiment, when the judgment result shows that at least one mechanical parameter is unqualified, it is necessary to calculate the actual executed dose distribution based on the actual execution sequence set to evaluate the actual impact of the mechanical parameter deviation on the treatment dose.

[0155] First, extract patient anatomical structure information from the aforementioned acquired planning documents, including the contour data of the planning target volume (PTV), clinical target volume (CTV), gross tumor volume (GTV), and each organ at risk (OARs). These contour data are usually stored in the form of a set of three-dimensional coordinate points, describing the boundaries of each anatomical structure. For example, the anatomical structure information of a patient with a lung tumor may include the GTV (the tumor itself), CTV (the clinical target volume considering microscopic infiltration), PTV (the planning target volume considering setup errors), and the contour data of organs at risk such as the lung, spinal cord, esophagus, and heart.

[0156] Next, based on the timestamps in the actual execution sequence set, discretize the continuous mechanical parameter sequence into a finite number of irradiation state points. Since the data points in the actual execution sequence may be very dense, to improve the calculation efficiency, it is necessary to simplify them into representative irradiation state points. For example, an irradiation state point can be set every 1 degree or every 5 degrees according to the gantry angle, or the irradiation state points can be set according to the change of dose rate.

[0157] For each discrete irradiation state point, construct an irradiation geometric configuration according to the mechanical parameters at the corresponding moment. The irradiation geometric configuration includes the ray source position (determined by the gantry angle), collimator direction (determined by the collimator angle), irradiation field shape (determined by the tungsten gate position and multi-leaf collimator position), and ray intensity (determined by the monitor units). For example, when the gantry angle is 45 degrees and the collimator angle is 0 degrees, the ray source position is at the 45-degree position in the front right of the patient, the ray direction points to the isocenter, and the irradiation field shape is jointly determined by the current tungsten gate and multi-leaf collimator positions.

[0158] After constructing the irradiation geometric configuration, use the convolution / superposition algorithm to calculate the energy deposition distribution of each irradiation state point. The convolution / superposition algorithm is a commonly used dose calculation method that takes into account the attenuation and scattering of rays in tissues. In the calculation process, first determine the primary energy deposition kernel, and then convolve it with the density distribution of each point in the phantom to obtain the energy deposition distribution. For each irradiation state point, a corresponding three-dimensional dose distribution is calculated.

[0159] Then, weight the dose contributions of each irradiation state point according to their time ratios, and accumulate the weighted dose distributions of all irradiation state points to obtain a complete three-dimensional dose distribution matrix. The time ratio can be calculated according to the time interval between irradiation state points. For example, if the times of two adjacent state points are t1 and t2 respectively, the time ratio of this irradiation state point can be (t2 - t1) / total treatment time.

[0160] The finally obtained three-dimensional dose distribution matrix contains the dose values of the planning target volume, clinical target volume, total tumor volume, and each voxel point in each organ at risk. Based on these dose values, isodose curves can be generated to show the spatial characteristics of the dose distribution; dose-volume histograms can be generated to show the volume percentages of each structure receiving different dose levels; and dose statistical parameters can be calculated to quantitatively evaluate the characteristics of the dose distribution.

[0161] Dose statistical parameters include the maximum dose of the target volume (representing the highest dose received by any voxel in the target volume), minimum dose (representing the lowest dose received by any voxel in the target volume), average dose (representing the average value of the doses of all voxels in the target volume), median dose (representing the dose received by 50% of the volume of the target volume), D95 (representing the dose received by 95% of the volume of the target volume), D98 (representing the dose received by 98% of the volume of the target volume), D2 (representing the dose received by 2% of the volume of the target volume), homogeneity index (representing the degree of uniformity of the dose distribution, usually defined as D5 / D95), and conformity index (representing the degree of conformity of the dose distribution to the shape of the target volume). For organs at risk, the maximum dose, average dose, and specific volume dose metrics (such as V20, representing the volume percentage of the organ receiving a dose above 20 Gy) are calculated.

[0162] By calculating the actually delivered dose distribution, the actual impact of mechanical parameter deviations on the treatment effect can be objectively evaluated, providing a data basis for subsequent dose deviation assessment.

[0163] S6. Obtain the original planned dose distribution from the planning file, compare the differences between the actually delivered dose distribution and the original planned dose distribution, and generate a dose deviation assessment report.

[0164] Comparing the differences between the actually delivered dose distribution and the original planned dose distribution, the generated dose deviation assessment report includes:

[0165] Comparing the dose values of each voxel point in the actually delivered dose distribution and the original planned dose distribution, calculating the absolute value and relative percentage of the voxel point dose difference;

[0166] Comparing the isodose curves of the actually delivered dose distribution and the original planned dose distribution to determine the isodose curve displacement;

[0167] Comparing the dose-volume histograms of the actually delivered dose distribution and the original planned dose distribution, calculating the dose coverage deviations of the planning target volume, clinical target volume, total tumor volume, and each organ at risk;

[0168] Comparing the dose statistical parameters of the actually delivered dose distribution and the original planned dose distribution, calculating the deviation values of each parameter;

[0169] Determine the acceptability of each deviation according to the preset allowable criteria, and generate a deviation assessment report including the above deviation quantification indicators and the assessment conclusion of acceptability.

[0170] In this embodiment, after calculating the actual executed dose distribution, it is necessary to compare it with the original planned dose distribution to evaluate the difference between the actual treatment and the plan, and generate a dose deviation assessment report.

[0171] First, extract the original planned dose distribution from the obtained plan document. The original planned dose distribution is the ideal dose distribution calculated based on the plan parameters in the treatment planning system, representing the expected treatment effect.

[0172] Next, compare the dose values of each voxel point in the actual executed dose distribution and the original planned dose distribution, and calculate the absolute value and relative percentage of the voxel point dose difference. For example, if the value of a voxel point in the planned dose distribution is 60 Gy and the value in the actual executed dose distribution is 58.5 Gy, then the absolute value of the dose difference at this point is 1.5 Gy, and the relative percentage is 2.5%. In this way, the distribution of dose differences in the entire three-dimensional space can be obtained, with particular attention to the dose differences within the target area and critical organs at risk.

[0173] Then, compare the isodose curves of the actual executed dose distribution and the original planned dose distribution to determine the isodose curve displacement. The isodose curve is a curve formed by connecting points with the same dose value, which can intuitively reflect the spatial characteristics of the dose distribution. By calculating the displacement distance of the key isodose curves (such as the 95% isodose curve of the prescription dose) on different sections, the spatial offset of the dose distribution can be quantified. For example, if the 50 Gy isodose curve of the actual execution is offset by 3 mm in a certain direction relative to the 50 Gy isodose curve of the plan, then record the displacement of this isodose curve as 3 mm.

[0174] Next, compare the dose volume histograms of the actual executed dose distribution and the original planned dose distribution, and calculate the dose coverage deviation of the planned target volume, clinical target volume, total tumor volume, and each organ at risk. The dose coverage deviation can be calculated by comparing the volume differences of the structures receiving the same dose level. For example, if 95% of the volume of the PTV in the plan receives the prescription dose, while only 92% of the volume receives the prescription dose in the actual execution, then the prescription dose coverage deviation of the PTV is -3%. Similarly, for organs at risk, the coverage deviation of important dose constraint points is also calculated, such as the difference between the actual value and the planned value of the lung V20 (the percentage of the lung tissue volume receiving a dose above 20 Gy).

[0175] Then, compare the dose statistical parameters of the actually executed dose distribution with those of the original planned dose distribution, and calculate the deviation values of each parameter. The dose statistical parameters include the maximum dose, minimum dose, average dose, median dose, D95, D98, D2, homogeneity index, and conformity index of the target area mentioned above, as well as the maximum dose, average dose, and specific volume dose index of the organs at risk. For example, if the D95 of the PTV in the plan is 65 Gy and it is 63.5 Gy during actual execution, the deviation of D95 is -1.5 Gy or -2.3%.

[0176] Finally, according to the preset acceptance criteria, determine the acceptability of each deviation, and generate a deviation assessment report containing the above deviation quantification indicators and the acceptability assessment conclusion. The acceptance criteria are usually formulated based on clinical practice and radiotherapy quality assurance guidelines. For example, it can be stipulated that the deviation of D95 of the PTV should not exceed ±3% of the prescribed dose, and the deviation of the maximum dose of the organs at risk should not exceed ±5%, etc.

[0177] By extracting the actually executed mechanical parameters from the log file of the linear accelerator and making a detailed comparison with the expected parameters in the plan file, the deviations at the mechanical execution level are identified. Different from the traditional method that only stays at the qualification judgment of mechanical parameters, this solution further analyzes the influence of these mechanical parameter deviations on the actual dose distribution. By converting the mechanical parameters into the irradiation geometric configuration, calculating the actually executed dose distribution, and then making a multi-dimensional comparison with the original planned dose distribution, a comprehensive dose deviation assessment report is finally generated. This full-chain analysis method establishes a clear correlation between mechanical parameter deviations and dose distribution, enabling operators to intuitively understand the influence degree of mechanical execution errors on the actual dose distribution, and thus making more reasonable decisions based on dose distribution differences rather than simply on the qualification of mechanical parameters.

[0178] Embodiment 2

[0179] Please refer to Figure 2 , the present invention provides a mechanical parameter verification system for a medical electron linear accelerator, which is used to implement a mechanical parameter verification method for a medical electron linear accelerator, including:

[0180] A log parsing module, which is used to obtain the log file of a single use of the linear accelerator, extract the actually executed mechanical parameters from the log file, and form an actually executed sequence set according to the time stamp sequence in the log file;

[0181] A plan parsing module, which is used to obtain the plan file corresponding to a single use, extract the expected mechanical parameters and their timing distributions of each control point from the plan file, and form an expected sequence set;

[0182] A tolerance generation module, which is used to obtain the historical usage records and generate an adaptive tolerance standard according to the historical usage records;

[0183] A parameter comparison module, configured to compare each parameter sequence in the actual execution sequence set with the corresponding parameter sequence in the expected sequence set, and determine whether each mechanical parameter is qualified by applying an adaptive tolerance standard;

[0184] A dose reconstruction module, configured to calculate an actual execution dose distribution based on the actual execution sequence set when at least one mechanical parameter is unqualified;

[0185] An evaluation report module, configured to obtain an original planned dose distribution from a plan file, compare the difference between the actual execution dose distribution and the original planned dose distribution, and generate a dose deviation evaluation report.

[0186] In several embodiments provided by the present invention, 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 one type, 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 mutual coupling or direct coupling or communication connection may be through some interfaces, and the indirect coupling or communication connection of the devices or units may be in an electrical, mechanical, or other form.

[0187] The above is only a specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of changes or substitutions, which should all be covered by the protection scope of the present invention.

[0188] Finally: The above is only a preferred embodiment of the present invention and is not used to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for verifying mechanical parameters of a medical electron linear accelerator, characterized in that, Including: Obtain the log file of a single use of a linear accelerator, extract the actually executed mechanical parameters from the log file, and form an actual execution sequence set according to the time stamp sequence in the log file; Obtain the plan file corresponding to the single use, extract the expected mechanical parameters and their timing distributions of each control point from the plan file, and form an expected sequence set; Obtain the historical usage records, and generate an adaptive tolerance standard according to the historical usage records; Compare each parameter sequence in the actual execution sequence set with the corresponding parameter sequence in the expected sequence set, and apply the adaptive tolerance standard to judge whether each mechanical parameter is qualified; When at least one mechanical parameter is unqualified, calculate the actually executed dose distribution based on the actual execution sequence set; Obtain the original planned dose distribution from the plan file, compare the difference between the actually executed dose distribution and the original planned dose distribution, and generate a dose deviation evaluation report.

2. The mechanical parameter verification method of a medical electron linear accelerator according to claim 1, wherein The mechanical parameters include gantry angle, collimator angle, tungsten gate position, multi-leaf collimator position, and monitor units.

3. A method for verifying mechanical parameters of a medical electron linear accelerator according to claim 2, characterized in that The generating an adaptive tolerance standard according to the historical usage records includes: Read the mechanical parameter log file and the corresponding dose measurement data in the historical usage records; screen out the treatment records with mechanical parameter deviations and completed dose verification; Classify the mechanical parameter deviation values and the dose deviation values according to the treatment site, extract the deviation sample set of each mechanical parameter and the corresponding dose deviation sample set, and perform standardization processing; For each mechanical parameter, calculate the dose change amount caused by its unit deviation; Adopt the multiple linear regression method to calculate the partial derivative values of each mechanical parameter to the dose distribution; Organize the partial derivative values of each treatment site and each mechanical parameter into a matrix form to obtain a parameter sensitivity matrix; Generate an adaptive tolerance table according to the preset reference tolerance value and the parameter sensitivity matrix, and use the adaptive tolerance table as the adaptive tolerance standard.

4. A method for verifying mechanical parameters of a medical electron linear accelerator according to claim 3, characterized in that, The applying the adaptive tolerance standard to judge whether each mechanical parameter is qualified includes: For each mechanical parameter, perform the following judgment steps: Obtain the actual execution sequence and the expected sequence of this mechanical parameter, and judge the overall consistency between the actual execution sequence and the expected sequence; If the result of the overall consistency judgment is inconsistent, determine that this mechanical parameter is unqualified; If the result of the overall consistency judgment is consistent, align the actual execution sequence and the expected sequence in time; Calculate the difference between the aligned sequences at each time point to obtain the deviation sequence of this mechanical parameter; Query the tolerance threshold corresponding to this mechanical parameter from the adaptive tolerance table; Count the proportion of the number of points exceeding the tolerance threshold to the total number of points; When the proportion exceeds the preset allowable proportion of this mechanical parameter, determine that this mechanical parameter sequence is unqualified, otherwise determine that this mechanical parameter is qualified.

5. A method for verifying mechanical parameters of a medical electron linear accelerator according to claim 4, characterized in that, The judging the overall consistency between the actual execution sequence and the expected sequence includes: Calculate the difference between the extreme value of the actual execution sequence and the extreme value of the expected sequence; Calculate the difference between the mean value of the actual execution sequence and the mean value of the expected sequence; Calculate the difference between the standard deviation of the actual execution sequence and the standard deviation of the expected sequence; Determine whether the differences between the extreme value differences, the mean differences, and the standard deviation differences are respectively within the corresponding preset thresholds. If any of the differences exceeds the corresponding preset threshold, the determination result is inconsistent; otherwise, the determination result is consistent.

6. A method for verifying mechanical parameters of a medical electron linear accelerator according to claim 5, characterized in that, The time alignment of the actual execution sequence and the expected sequence includes: Set the first point of the actual execution sequence and the first point of the expected sequence as the time reference origins of their respective sequences; Based on each point in the expected sequence, find the point corresponding in time in the actual execution sequence; If there is a point in the actual execution sequence that exactly corresponds in time to a point in the expected sequence, directly extract the parameter value of that point; If there is no point in the actual execution sequence that exactly corresponds in time to a point in the expected sequence, establish an insertion point, determine the reference point of the insertion point, calculate the parameter value of the insertion point based on the parameter value of the reference point, and insert the insertion point into the actual execution sequence according to the corresponding time; Delete the points in the actual execution sequence that do not correspond in time to the points in the expected sequence to complete the time alignment between the actual execution sequence and the expected sequence.

7. A method for verifying the mechanical parameters of a medical electron linear accelerator according to claim 6, characterized in that The determination of the reference point of the insertion point includes: Find the time point in the actual execution sequence with the time value closest to the time of this point in the expected sequence as the nearest reference point; Find the previous point of the nearest reference point in the actual execution sequence as the previous reference point; Find the next point of the nearest reference point in the actual execution sequence as the next reference point; Determine the nearest reference point, the previous reference point, and the next reference point as the reference points of the insertion point; The calculation of the parameter value of the insertion point based on the parameter value of the reference point includes: Obtain the parameter values of the previous reference point, the nearest reference point, and the next reference point; According to the parameter values of the three reference points, calculate the parameter value corresponding to the time of this point in the expected sequence through weighted summation; Among them, the sum of the weight coefficients of the three reference points is 1, and it is set based on the time relationship between the three reference points and the insertion point.

8. A method for verifying mechanical parameters of a medical electron linear accelerator according to claim 2, characterized in that, The calculation of the actual execution dose distribution based on the actual execution sequence set includes: Obtain the patient anatomical structure information from the plan file, including the planned target volume, clinical target volume, total tumor volume, and contour data of each organ at risk; Based on the time stamps in the actual execution sequence set, discretize the continuous mechanical parameter sequence into a finite number of irradiation state points; For each discrete irradiation state point, construct the corresponding irradiation geometric configuration according to the mechanical parameters; Use the convolution / superposition algorithm to calculate the energy deposition distribution of each irradiation state point; Weight the dose contributions of each irradiation state point according to the time proportion, and accumulate to obtain the complete three-dimensional dose distribution matrix; The three-dimensional dose distribution matrix includes the dose values, isodose curves, dose volume histograms, and dose statistical parameters of each voxel point in the planned target volume, clinical target volume, total tumor volume, and each organ at risk; The dose statistical parameters include the maximum dose, minimum dose, average dose, median dose, D95, D98, D2, uniformity index, and conformity index of the target volume, as well as the maximum dose, average dose, and specific volume dose index of the organ at risk.

9. A method for verifying mechanical parameters of a medical electron linear accelerator according to claim 8, characterized in that, The comparison of the differences between the actual execution dose distribution and the original planned dose distribution to generate a dose deviation assessment report includes: Compare the dose values of each voxel point in the actually delivered dose distribution and the original planned dose distribution, and calculate the absolute value and relative percentage of the dose difference at the voxel points; Compare the isodose curves of the actually delivered dose distribution and the original planned dose distribution to determine the displacement of the isodose curves; Compare the dose volume histograms of the actually delivered dose distribution and the original planned dose distribution, and calculate the dose coverage deviations of the planning target volume, clinical target volume, total tumor volume, and each organ at risk; Compare the dose statistical parameters of the actually delivered dose distribution and the original planned dose distribution, and calculate the deviation values of each parameter; According to the preset tolerance criteria, determine the acceptability of each deviation, and generate a deviation assessment report containing the above deviation quantification indicators and the acceptability assessment conclusion.

10. A mechanical parameter verification system for a medical electron linear accelerator, which is used to implement a mechanical parameter verification method for a medical electron linear accelerator as described in any one of claims 1-9, characterized in that, Including: A log parsing module for obtaining the log file of a single use of a linear accelerator, extracting the actually executed mechanical parameters from the log file, and forming an actually executed sequence set according to the time stamp sequence in the log file; A plan parsing module for obtaining the plan file corresponding to a single use, extracting the expected mechanical parameters and their timing distributions of each control point from the plan file, and forming an expected sequence set; A tolerance generation module for obtaining historical usage records and generating an adaptive tolerance standard according to the historical usage records; A parameter comparison module for comparing each parameter sequence in the actually executed sequence set with the corresponding parameter sequence in the expected sequence set, and judging whether each mechanical parameter is qualified by applying the adaptive tolerance standard; A dose reconstruction module for calculating the actually delivered dose distribution based on the actually executed sequence set when at least one mechanical parameter is unqualified; An evaluation report module for obtaining the original planned dose distribution from the plan file, comparing the difference between the actually delivered dose distribution and the original planned dose distribution, and generating a dose deviation evaluation report.

Images