System and method for dose guidance and iterative estimation of final irradiation dose in radiotherapy systems

Abstract

The present disclosure relates to a method for dose guidance of a radiation therapy system during a radiation therapy session, the method comprising the steps of performing a substantially real-time dose reconstruction to obtain an exposure radiation dose generated by a radiation therapy beam of the radiation therapy system in at least one volume, iteratively estimating a remaining radiation dose based on an observed and / or simulated motion of the at least one volume, iteratively estimating a final exposure radiation dose of the radiation therapy session as a sum of the exposure radiation dose and the estimated remaining radiation dose, and providing dose guidance for a remaining portion of the radiation therapy session based on the estimated final exposure radiation dose. The present disclosure further relates to a method for continuously estimating a final exposure radiation dose during a radiation therapy session and a decision support system for a radiation therapy system.

Description

Technical Field

[0001] The present disclosure relates to a system and method for iteratively estimating a final irradiation radiation dose during a radiotherapy session. The present disclosure further relates to a method of dose guidance for a radiotherapy system during a radiotherapy session. The method can be executed continuously and substantially in real time during a radiotherapy session.

Background Art

[0002] Radiation therapy is a therapy that uses ionizing radiation, generally as part of cancer treatment to control or kill malignant cells. Since radiation therapy has the ability to control cell growth, it is commonly applied to cancerous tumors. In radiation therapy, it is extremely important to accurately irradiate the treatment dose in order to maximize the ratio of tumor dose to normal tissue dose and effectively cure the patient while minimizing side effects. Dose reconstruction is the process of estimating the radiation dose received by an object or individual.

[0003] During radiation therapy, the target object and the surrounding area may move, rotate, or deform for various reasons. As an example, a lung tumor may move with respiration, but may also move for other reasons such as patient movement relative to the radiation beam or internal movement within or between organs. Some movements / rotations / deformations are difficult to predict. In current radiation therapy for moving tumors / organs, a significant amount of healthy tissue is often irradiated to ensure proper treatment of the tumor. Organ movement during treatment can lead to deterioration of the planned dose distribution.

[0004] Including movement / rotation / deformation in radiation therapy poses a major challenge to quality assurance (QA) protocols that include dose reconstruction of the irradiation dose. Although there are algorithms capable of accurate dose reconstruction, these algorithms are too slow to be used in real time due to computational time or cannot account for the movement / rotation / deformation of the target object.

[0005] Image-guided radiotherapy (IGRT) attempts to compensate for interfractional anatomical changes, i.e., changes between radiotherapy sessions, by aligning the anatomical structures of daily setup images with simulated images. Alignment is typically based on the anatomical structures of bone or soft tissue, or the position of implanted reference markers. However, this geometric arrangement is simplified to account for the patient's rigidity.

[0006] Similarly, researchers are studying the concept of dose-guided patient positioning, where the optimal patient position for dose measurement is calculated based on the anatomical structure on any given day. Similar to IGRT, dose-guided efforts have focused on inter-irradiation anatomical changes observed in pre-treatment imaging. [Overview of the project]

[0007] This disclosure addresses the aforementioned issues. According to the first embodiment, a method is provided for dose guidance of a radiotherapy system during a radiotherapy session. This method is The steps include performing dose reconstruction in virtually real time to obtain the radiation dose generated by the radiotherapy beam of a radiotherapy system in at least one volume, A step of iteratively estimating the remaining radiation dose of a radiotherapy session based on the observation and / or simulated movement of at least one volume, The steps include repeatedly estimating the final radiation dose of a radiotherapy session as the sum of the radiation dose from at least one volume and the remaining estimated radiation dose, The procedure includes the step of providing dose guidance for the remainder of the radiotherapy session based on the estimated final radiation dose.

[0008] By estimating the final radiation dose in virtually real time, it becomes possible to make informed decisions about treatment interventions using a clinically relevant metric—absorbed dose—rather than shape.

[0009] A radiotherapy session typically involves a treatment plan to deliver an appropriate dose to one or more target volumes while minimizing the dose to the criticality avoidance dose. The dose guidance method of this disclosure involves a single process that performs substantially real-time dose reconstruction for at least one volume. This process delivers a cumulative dose distribution to at least one volume until the current time is calculated sequentially.

[0010] In the second process, the method iteratively estimates the remaining radiation dose of the radiotherapy session based on the observed and / or simulated motion of at least one volume. This can be done by convolving the previously calculated planned dose with the unirradiated portion and the motion observed so far. As a non-limiting example, it is possible to use the motion of a volume observed within a time frame of the past 40 seconds, for example.

[0011] There are repeatedly estimated cumulative dose and residual dose; the former is calculated in virtually real time, taking into account the actual volume movement, while the latter is an estimate based on the volume movement up to the present time, and is preferably also estimated in virtually real time. Therefore, the final radiation dose of a radiotherapy session can be repeatedly estimated as the sum of the radiation dose and the estimated residual dose.

[0012] In the context of this disclosure, “substantially real-time” means that calculations can be repeatedly performed with sufficient precision and sufficiently high temporal resolution during a radiotherapy session.

[0013] Continuous / iterative estimation of the final radiation dose, taking observed motion into account, significantly improves decision-making support in radiotherapy systems. The final radiation dose estimation can be used to support decision-making and dose guidance for the remainder of a radiotherapy session. This allows for iterative updating of the estimated final dose-volume histogram (DVH) of the region of interest during a radiotherapy session, which can then be analyzed and displayed during ongoing treatment. Real-time updated DVH parameters can be used in optimization strategies to find treatment indications, such as couch correction, leading to improvements in the optimal final dose. Comparing such estimated final doses with and without dosiologically optimal couch correction enables decision-making based on information regarding couch indications during irradiation, i.e., indications within a radiotherapy session.

[0014] Dose guidance during radiotherapy is generally considered to improve the precision and accuracy of therapeutic irradiation by utilizing the estimated final radiation dose and, if applicable, other relevant information such as observed or simulated motion. Examples of dose guidance include providing patient position and / or couch position, or correction and / or beam correction information, which can be used to adjust and improve the remainder of the radiotherapy session. This may include suggesting couch corrections or re-optimization of the treatment plan for the remainder of the radiotherapy session.

[0015] This technology can be provided as a computer program with a graphical user interface. Treatment plans can be obtained through the user interface. The computer program can also obtain feeds of linear accelerator parameters as well as tumor and organ movement. The computer program can provide continuous information regarding the accumulated dose throughout the session, the estimated residual radiation dose for the radiotherapy session, and the estimated final radiation dose, with or without correction.

[0016] This information can also be used to calculate the gating of the radiotherapy beam for the remainder of the radiotherapy session, or to re-optimize the treatment plan for the remainder of the radiotherapy session. The estimated final radiation dose enables informed decision-making regarding treatment interventions based on the most relevant parameter, i.e., the final dose to the patient. This opens the door to new strategies for adapting to motion during irradiation based on the dosogenic state of the treatment being irradiated, and can take into account conflicting dose considerations of up to multiple targets and / or radiosensitivity risk volumes.

[0017] This disclosure further relates to a method for continuously estimating the final radiation dose during a radiotherapy session, wherein this method... The steps include performing real-time dose reconstruction to obtain the radiation dose generated by the radiotherapy beam of a radiotherapy system in at least one volume, A step of iteratively estimating the remaining radiation dose of a radiotherapy session based on the observation and / or simulated movement of at least one volume, The process includes the step of repeatedly estimating the final radiation dose of a radiotherapy session as the sum of the radiation dose and the estimated remaining radiation dose.

[0018] Those skilled in the art will recognize that the features and further embodiments described herein are not limited to one of the methods but are interchangeable. For example, where a section specifies that the motion may include substantially sinusoidal motion and / or non-rotational motion portions such as baseline drift and / or irregular motion portions, this section is applicable to both the method of dose guidance of a radiotherapy system during a radiotherapy session and the method of sequential estimation of the final radiation dose during a radiotherapy session.

[0019] The present disclosure further relates to a computer program having instructions that, when executed by a computing device or a computing system, cause the computing device or the computing system to execute one of the methods, and to a decision support system for a radiotherapy system, the decision support system comprising: an interface for receiving radiotherapy beam parameters and / or parameters related to the observation and / or simulated movement of at least one volume, and a processing unit, the processing unit being configured to: perform real-time dose reconstruction to obtain the irradiation radiation dose generated by the radiotherapy beam of the radiotherapy system within at least one volume; repeatedly estimate the remaining radiation dose of the radiotherapy session based on the observation and / or simulated movement of at least one volume; repeatedly estimate the final irradiation radiation dose of the radiotherapy session as the sum of the irradiation radiation dose and the estimated remaining radiation dose. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] [Figure 1] An example of a workflow according to the method of the present disclosure for continuous estimation of the final irradiation radiation dose is shown. [Figure 2] A flowchart of a method for dose guidance of a radiotherapy system is shown. [Figure 3] A flowchart of a method according to an embodiment of the method of the present disclosure for continuous estimation of the final irradiation radiation dose is shown. [Figure 4] A schematic diagram of an embodiment of a decision support system for a radiotherapy system of the present disclosure is shown. [Figure 5] An example of a schematic diagram of a radiotherapy model of an object for real-time dose reconstruction is shown. [Figure 6] Examples of the final clinical target volume (CTV) D95 repeatedly predicted during simulated treatment fractions for different scenarios are shown. [Figure 7] Shows the comparison of the predicted final CTV ΔD95 with the actual final CTV ΔD95 in various scenarios. [Figure 8] Shows the absolute error of the real-time predicted CTV ΔD95 for many movements. [Figure 9] Shows an example of a user interface for providing dose guidance using the method of the present disclosure for dose guidance of a radiotherapy system during a radiotherapy session.

Mode for Carrying Out the Invention

[0021] This disclosure relates to a method for dose guidance of a radiotherapy system during a radiotherapy session, and a method for sequentially estimating the final radiation dose during a radiotherapy session. Using either of these methods, the final radiation dose of a radiotherapy session can be iteratively estimated in substantially real time. This information can be used as decision support in a radiotherapy session. The method includes the step of performing real-time dose reconstruction to obtain the radiation dose produced by the radiotherapy beam of the radiotherapy system in at least one volume. A computerized version of the real-time dose reconstruction method is used to calculate the dose irradiated to the moving volume. Real-time dose reconstruction can be based on a live stream of position and accelerator parameters. The method further includes the step of iteratively estimating the remaining radiation dose of a radiotherapy session based on the observed or simulated motion of at least one volume. The inventors have found that this can be done by convolving the previously observed motion with the remaining portion of a pre-calculated planned dose that has not yet been irradiated. Thus, the method can also iteratively estimate the final radiation dose of a radiotherapy session as the sum of the irradiated radiation dose and the estimated remaining radiation dose. The estimated final dose provides valuable information that the system or clinician can use to adjust and improve the rest of the radiotherapy system, for example, by couch correction or reoptimizing the treatment plan for the rest of the radiotherapy session. The “final” dose may be the cumulative dose after the current treatment session, but it may also be the cumulative dose from other (previous and / or future) treatment sessions.

[0022] In the context of this disclosure, “volume” can be any appropriate volume of any size. A volume may include, for example, a single target, multiple targets, a single risk volume, multiple risk volumes, or a small specific volume containing one or more specific points of interest.

[0023] Figures 1A–B provide an example of a workflow according to the method (100) of the present disclosure for continuous estimation of final irradiation dose. This drawing, like the rest of the drawings, is illustrative and intended to illustrate some of the features of the dose guidance and continuous estimation method of final irradiation dose of the present disclosure and should not be construed as limiting the invention of the present disclosure. Prior to a radiotherapy session, a planned static dose for a given plan is calculated (101). Thus, a predefined planned dose for a target and / or at least one volume can be calculated. During the session, a synchronous stream of accelerator parameters (103) and the location of at least one volume (102) is sent to a data queue (104). In a continuous loop, the method reconstructs the cumulative dose distribution (105) irradiated to at least one potentially moving volume up to the present time (DDeliveredReconstructed). Thus, the step of performing real-time dose reconstruction may be repeated when new input data is received. The input data may include parameters related to the apparatus for performing the radiotherapy session, such as accelerator parameters and target position parameters, including parameters related to the translational, rotational, and / or deformational motion of one or more risk targets and / or one or more risk objects. Thus, the input data may include a first data stream containing parameters related to the apparatus for performing the radiotherapy session, such as accelerator parameters. The input data may include a second data stream containing parameters related to the irradiated object, such as parameters related to the translational, rotational, and / or deformational motion of one or more risk targets or one or more risk objects. After each dose calculation, the number, movement, and cumulative reconstructed dose of the irradiated monitor units are added to the internal history (107). The real-time dose guidance loop continuously checks the internal history (107) and estimates the remaining partial dose DRemainingEstimated by convolving (108) a portion of the remaining pre-calculated planned dose (106) to be irradiated based on the movement observed up to that point.Therefore, the step of observing the target movement may include movement with a time window preceding the point in time when the remaining radiation dose is estimated (109). As a non-limiting example, it is possible to use, for example, volume movement observed within a time window of the past 40 seconds. The final radiation dose can then be estimated as the sum of the irradiated dose and the estimated remaining dose (109). This can be done with or without couch correction (110;112) or with couch correction (111;113). Other possibilities also exist. For example, it is possible to re-optimize the treatment plan for the remainder of the radiotherapy session. In the example in Figures 1A-B, couch correction is ultimately applied if the minimum dose for 95% of the clinical target volume (CTV D95) increases by a predetermined threshold due to couch correction (114).

[0024] Figure 2 shows an illustrative flowchart (200) of a method for dose guidance of a radiotherapy system. The method includes the steps of: performing real-time dose reconstruction (201) to obtain the irradiation dose generated by the radiotherapy beam of the radiotherapy system in at least one volume; sequentially estimating the remaining radiation dose of a radiotherapy session based on the observed or simulated motion of at least one volume (202); sequentially estimating the final irradiation dose of a radiotherapy session as the sum of the irradiation dose and the estimated remaining radiation dose (203); and providing dose guidance for the remainder of a radiotherapy session based on the estimated final irradiation dose (204).

[0025] The "final dose" is the cumulative dose after an ongoing treatment session and may be added to the doses of other (previous and / or future) treatment sessions. When dose guidance is provided for multiple treatment sessions, the objective in an ongoing session may be to deliver a session dose that matches the cumulative total dose. For example, if a portion of the target received too little dose in a previous session, dose guidance may target a higher dose (than planned) to that portion of the target in the current session. Similarly, if an organ at risk received a dose higher / lower than acceptable in a previous session, dose guidance should allow a lower / higher dose to that organ at risk in the current session.

[0026] Figure 3 shows an exemplary flowchart (300) of a method for sequentially estimating the final radiation dose during a radiotherapy session. The method includes the steps of: performing real-time dose reconstruction (301) to obtain the radiation dose generated by the radiotherapy beam of the radiotherapy system in at least one volume; sequentially estimating the remaining radiation dose of the radiotherapy session based on the observed or simulated motion of at least one volume (302); and sequentially estimating the final radiation dose of the radiotherapy session as the sum of the radiation dose and the estimated remaining radiation dose (303).

[0027] Observation and simulation of movement Motion observation can be carried out in various ways. In non-limiting examples, observed or simulated motion may be obtained from a transponder or marker in at least one volume, and / or from an external surrogate, and / or from imaging such as magnetic resonance imaging, X-ray imaging, or optical imaging. The transponder may be a wired or wireless electromagnetic transponder. Other methods of observing motion include fluorescence imaging of an embedded reference radiopaque marker, observation of an external surrogate, or a combination of the above. The marker may be made of gold or other material suitable for the purpose. Furthermore, the motion does not necessarily have to be observed motion. In one embodiment of the method of continuous estimation of final irradiation dose and dose guidance of this disclosure, the motion is simulated motion.

[0028] The motion may include periodic motion portions, such as substantially sinusoidal motion, and / or baseline drift and / or irregular motion portions. There are several reasons why volume may move. For example, such motion may be caused by respiration. The patient may also move relative to the radiation beam. Other motions during irradiation may also exist. For example, organs or targets may move due to changes in bone position, or the prostate may move due to bladder or rectal filling. The methods disclosed herein for continuously estimating the final radiation dose and dose guidance may be real-time methods that correspond to motions during irradiation and can therefore estimate the final radiation dose using not only the periodic motion portions but also the baseline drift and / or irregular motion portions. In this regard, “baseline drift” may be defined as a gradually changing position averaged over several motion cycles. “Unstable motion portions” may be defined as motions that are neither periodic nor “baseline drift”.

[0029] Continuously estimate the remaining radiation dose. The present disclosure's method for sequential estimation of final irradiation dose and dose guidance includes the step of iteratively estimating the remaining radiation dose of a radiotherapy session based on observed or simulated motion of at least one volume. The estimated “remaining radiation dose” can be considered the remaining radiation dose that would be delivered if the radiotherapy session continued as planned without modification. However, if the final irradiation dose is calculated (as the sum of the irradiation dose and the estimated remaining radiation dose for at least one volume), dose guidance or decision support may include estimating the remaining dose for modified alternative simulated scenarios, such as couch corrections. This is described in more detail below.

[0030] In this step, the remaining plan dose can be convolved with a representation of the expected movement of at least one volume in the remainder, such as a convolution with an estimated probability density function of at least one volume in the remainder of the treatment session. The remaining radiation dose of a radiotherapy session, including the effects of movement, can be estimated as the remaining plan dose without movement, convolved with a probability density function describing the expected movement of at least one volume in the remainder of the treatment session. Such a convolution may involve shifting and summing the remaining plan dose using weights corresponding to the observed or simulated positions of at least one volume up to a given time. Convolution is a mathematical operation on two functions (e.g., f and g) that produce a third function (f*g). This can be understood as the sum of individual superpositions of the two functions, or as a superposition integral. The advantage of convolution over ordinary superposition sets is that it can be performed much faster by using Fourier space. The sum can be expressed for the remainder of the treatment session as a superposition of the remaining plan doses at the positions visited by at least one volume up to a certain point in time. The remaining planned dose can be convolved not only with an expression of the expected movement of at least one volume in the remainder of the treatment session, but also with additional parameters or entities such as the radiobiological response of the tissue.

[0031] As those skilled in the art will understand, convolution with an estimated probability density function is one of several methods for estimating the remaining radiation dose based on the observed and / or simulated motion of at least one volume. When considering observed and / or simulated motion, it can be assumed that the motion continues in the same manner. However, the motion may involve other assumptions, such as drift or other calculation or estimated trend extrapolation. No motion may be expected for the remainder of the treatment session. The probability density function can be based on the observed or simulated motion. It is also possible to simulate the motion for the remainder of the radiotherapy session using the assumed motion of at least one volume. The probability density function may include the distribution of the positions of at least one volume observed up to the present time during the radiotherapy session. The probability density function can describe future motion based on the motion observed or simulated so far. Thus, the probability density function can describe all positions that at least one volume is expected to take, with appropriate weights corresponding to the relative amount of time spent at each position. According to one embodiment, this method and system assumes that the observed motion continues for the remainder of the radiotherapy session. Another method for estimating the remaining radiation dose of a radiotherapy session based on the observed and / or simulated motion of at least one volume is to simulate the motion for the remainder of the radiotherapy session (based on the observed and / or simulated motion up to the present) and perform dose reconstruction based on the simulation.

[0032] The probability density function may include the distribution of the location of at least one volume observed up to the present time during the radiotherapy session. The probability density function may be based on previous radiotherapy or imaging sessions. Furthermore, the probability density function may also take into account previously observed motion patterns for a predefined patient group. This may be relevant, for example, to prostate treatment, as prostate motion patterns are highly similar among patients. The probability density function may include any of the above data individually or in combination.

[0033] Re-optimization, repositioning, gating The dose guidance methods for radiotherapy systems described herein may further include steps and decision support to improve the remainder of radiotherapy sessions. This may include several approaches.

[0034] Figures 1A and 1B illustrate an example of one embodiment of such decision support. It can be seen that, based on the continuously estimated remaining radiation dose (109), it is possible to estimate the final estimated dose in the form of couch movement (112) without intervention and with optimized couch movement (113).

[0035] One embodiment of the dose guidance method of this disclosure includes a step of calculating the repositioning of an object exposed to radiotherapy and / or the repositioning of a radiotherapy beam in order to improve the estimated remaining radiation dose based on the irradiated radiation dose and observed or simulated motion. More specifically, this step may include calculating a correction for the couch on which the patient is placed and / or a correction for the beam irradiating the patient.

[0036] Another way to support treatment decision-making is to calculate the gating on / off control of the radiotherapy beam during a radiotherapy session based on the irradiation and / or predicted radiation dose, as well as observed or simulated motion. The radiation beam can then be gated to ensure that the target(s) are irradiated and critical organs are avoided. In practice, this means the beam can be turned off if the target organ and / or organs at risk are at an unacceptable dose level, and the beam can be turned on if the target organ and / or organs at risk are at an acceptable dose level.

[0037] According to further embodiments, the dose guidance method of the present disclosure includes the step of performing a re-optimization of the treatment plan for the remainder of the radiotherapy session.

[0038] The method may further include steps to re-estimate the remaining radiation dose for the emulated intervention, such as repositioning the object to be exposed to radiotherapy and / or repositioning the beam and / or re-optimizing the treatment plan for the remainder of the radiotherapy session. The method may further include steps to extract an optimized position for the remainder of the radiotherapy session by using an optimization algorithm. In this way, a cost function can be generated that represents the radiation dose loss and / or the excess dose to healthy tissue at at least one target in the volume. By finding the minimum value of such a function, the optimized position can be extracted. Figures 1A-B show the calculation of the final radiation dose with and without optimal repositioning of the object.

[0039] The method may further include steps of emulating the repositioning of the object and / or organ at risk to be exposed to radiotherapy and / or repositioning the beam and / or reestimating the remaining radiation dose for re-optimization of the treatment plan for the remainder of the radiotherapy session.

[0040] Multiple target and risk areas As stated above, the continuous estimation method for final irradiation dose and dose guidance in this disclosure is particularly useful for addressing motion during irradiation.

[0041] The method is also suitable for sequentially estimating the final radiation dose to the target and adjacent organs, such as organs at risk. It may be important to sequentially estimate not only the final radiation dose to any target, but also the final radiation dose to organs at risk. Therefore, the method may further include a step of performing weighted optimization of repositioning the radiotherapy-exposed object and / or beam repositioning, and / or re-optimization of the treatment plan for the remainder of the radiotherapy session, with respect to the final radiation doses to the target and adjacent organs. The method may also include a step of sequentially estimating the final radiation dose to multiple targets. Multiple targets may be weighted. Therefore, the sequential estimation method for final radiation dose and dose guidance of the present disclosure may further include a step of performing weighted optimization of repositioning the radiotherapy-exposed object with respect to the final radiation doses to the target and a second target. The step of performing weighted optimization of repositioning the radiotherapy-exposed object can be applied to any appropriate number of targets.

[0042] Real-time dose reconstruction Performing real-time dose reconstruction means a) The step of selecting one or more points in a space exposed to radiation therapy, b) The step of projecting at least one of the points onto at least one plane that intersects with the central axis of the radiotherapy beam projected from the radiotherapy beam source, c) A step of calculating the dose in a plane (multiple planes are possible) generated by the radiotherapy beam, d) A step of depth scaling the dose in the plane from the projected point to the selected point, thereby obtaining the irradiation dose at the selected point, e) a step which involves repeatedly repeating steps b) to d).

[0043] The steps for performing real-time dose reconstruction are not limited to this particular method. Real-time dose reconstruction methods can be implemented to act on targets considered homogeneous, but they can also be implemented to act on heterogeneous targets.

[0044] Instead of calculating the dose for the entire volume (a 3D grid of voxels), the number of calculations can be significantly limited by selecting one or more points in space (step a of the method). By projecting the selected points onto a plane intersecting the central axis of the beam, interpolation along the beamline during ray tracing is avoided. By taking advantage of the fact that the radiotherapy target can be considered homogeneous during QA, ray tracing to correct for heterogeneity in dose calculation is not required. This assumption of homogeneity is consistent with common QA procedures. As a result, the depth scaling step reduces the computational load for calculating the dose at the selected points. Figure 5 shows an example schematic diagram of a radiotherapy model of an object. In Figure 5, the calculation points with coordinates (x, y, z) are points from which the dose can be calculated according to the method of this disclosure. The idea is that by selecting points in step a) to cover specific points, dose reconstruction values ​​can be obtained and used to provide a sufficiently accurate estimate of dose error while significantly limiting calculations. The point may be selected adjacent to or within an object that is the target of radiotherapy, or another area of ​​interest, such as an organ at risk of receiving too high a dose, but the point can essentially be any point that could be exposed to radiation.

[0045] In step b), the point is projected onto at least one plane that intersects the central axis of the radiotherapy beam. As mentioned above, one advantage associated with projection onto a plane is that ray tracing is not required to calculate the dose. In Figure 5, this corresponds to the point (x,y,z) being projected onto a plane, and the projected point

number

number

[0046] One aspect of the invention of this disclosure relates to projecting a selected point onto one, two, three, or four planes. Limiting the number of planes restricts the computationally intensive convolution of electron scattering kernels, such as the single-pencil-beam 2D ​​scattering kernel by the Storchi algorithm.

[0047] In one embodiment, the plane is an isoplanar. In this context, the isocenter requires the gantry to rotate in a circular motion around the target of radiotherapy. The isocenter is defined as the point at the center of the circle. The isoplanar is defined as a plane perpendicular to the central axis of the radiotherapy beam, and the isocenter is a part of that plane. One advantage of using an isosurface as the plane onto which a selected point is projected is that the distance of the isosurface to the beam source is fixed. In a further embodiment, the plane is a plane perpendicular to the central axis of the radiotherapy beam, and a particular point within the target or risk volume is a part of the plane. In this embodiment, the distance to the beam source is not fixed, but the lateral electron scattering calculated by the convolution of the plane with scattering is more accurate within the target or risk volume.

[0048] Depth scaling is, The steps include: calculating the absolute deep dose along the central axis of the radiotherapy beam as a function of the distance from the beam source; This can be carried out by performing the steps of calculating the dose irradiated to a selected point based on the dose in the plane and the absolute deep dose.

[0049] The central axis of the radiotherapy beam in Figure 5 can serve as an example of a central axis for this purpose.

[0050] By using a function of the absolute deep dose (with respect to the distance z from the beam source) along the central axis of the radiotherapy beam, continuous dose interpolation (ray tracing) along the beamline is avoided, significantly saving computation time. In further embodiments, interpolation is performed at discrete points along the beamline to improve accuracy with a negligible performance penalty.

[0051] Specifically, the absolute deep dose along the central axis of a radiotherapy beam can be calculated based on measured deep dose percentages. In one embodiment, the absolute deep dose is calculated based on a single deep dose percentage measured in water. Alternatively, the absolute deep dose function can be calculated based on a set of deep dose percentages, thereby improving the accuracy of the disclosed method compared to using a single measured deep dose percentage. More specifically, in one embodiment, the absolute deep dose along the central axis of a radiotherapy beam is

number

number

number

[0052] Next, the dose irradiated to the selected point can be calculated based on the absolute depth dose function and the dose calculated in the plane. The absolute depth dose function is usually expressed as a function of time and distance from the beam source. In one embodiment, the dose irradiated to the selected point is:

number

number

number

number

number

number

number

[0053] Decision support system This disclosure further relates to a decision support system for radiotherapy systems, and this system is: An interface for receiving radiotherapy beam parameters and / or parameters related to the observed or simulated motion of at least one volume, and It is equipped with a processing unit, and this processing unit is Perform real-time dose reconstruction to obtain the radiation dose generated by the radiotherapy beam of the radiotherapy system in at least one volume, To continuously estimate the remaining radiation dose of a radiotherapy session based on the observation and / or simulated movement of at least one volume, The system is configured to continuously estimate the final radiation dose of a radiotherapy session as the sum of the radiation dose and the estimated remaining radiation dose.

[0054] As those skilled in the art will understand, the processing unit of the system may be configured to perform a method for dose guidance of a radiotherapy system according to any embodiment described herein, and a method for sequentially estimating the final radiation dose during a radiotherapy session.

[0055] The decision support system may be an integrated part of any radiotherapy system or an add-on. The radiotherapy system may include a beam source configured to rotate in a circular motion around a target object, and / or a beam source configured to move around a target object in a non-coplanar space, and / or a treatment couch moved relative to the beam source.

[0056] Figure 4 shows a schematic diagram of one embodiment of a decision support system (400) of the present disclosure for a radiotherapy system, comprising a processing unit (401) and an interface (402) for receiving radiotherapy beam parameters and / or parameters relating to the observed or simulated motion of at least one volume. The decision support system may further comprise interfaces for controlling the patient's position and / or controlling the beam(s) and / or gating of the radiotherapy system.

[0057] Examples and Simulations A non-limiting example of the method of continuous estimation of final irradiation dose and dose guidance described herein has been evaluated in simulations.

[0058] We simulated the implementation of real-time dose-guided couch adaptation in 15 patients who had previously received stereotactic radiotherapy for the liver, using motion monitoring during irradiation with an implantable electromagnetic transponder (Calypso Soft Tissue Beacon Transponder from Varian Medical Systems). The planned target volume (PTV) was generated by expanding the CTV with an axial margin of 5 mm and a craniocaudal (CC) margin of 7 mm (n=12) or 10 mm (n=3). The prescribed mean CTV doses were 48 Gy for 4 patients with primary liver tumors, and 45 Gy (n=1), 56.25 Gy (n=9), or 61.8 Gy (n=1) for 11 patients with one (n=9), two (n=1), or three (n=1) metastases. 7-field intensity modulation or 3D conformal planning covered at least 99% of the CTV at 95% of the prescribed dose and the PTV at 67% of the prescribed dose. The treatment involved irradiation in three separate sessions using a TrueBeam linear accelerator (Varian) with Calypso-guided respiratory gating.

[0059] The following tumor motion trajectories were simulated: (1) Sine wave motion in the CC direction with a 4-second period, peak amplitude of 20 mm, and mean position offset by 4 mm and 8 mm in the cranial direction. Optimization of couch shift during simulation with sinusoidal motion was limited to the craniocaudal direction. (2) Internal tumor motion measured by Calypso.

[0060] Using a simulation framework, synchronized accelerator parameters and tumor location streams were broadcast as Unified Data Protocol (UDP) messages at a frequency of 21 Hz. A continuous dose reconstruction loop aggregated the UDP messages into the average accelerator state and tumor location since the last dose calculation, and this state was used for dose increment calculations.

[0061] During the simulation, the calculation times for dose calculation and dose guidance were recorded along with numerous different DVH and motion data for research purposes to analyze various dose guidance methods offline. Calculation time was quantified by calculating the average for each fraction, then the median and range. Calculation points for the same patient were selected differently based on whether the applied motion was one-dimensional (sine wave) or three-dimensional; therefore, calculation time was divided into Calypso-based motion (630 fractions) and sine-based motion (210 fractions) (Calypso).

[0062] The lower curve in Figure 6A shows an example of continuously predicted final CTV D95 over simulated treatment fractions without dose-guided couch correction. Figure 7 compares the predicted final CTV ΔD95 after each treatment field with the actual final decrease (ΔD95) in all treatment simulations. The accuracy of real-time predicted CTV ΔD95 generally improved as the treatment fraction progressed, with less emphasis on the estimated remaining dose and more emphasis on the reconstructed dose delivered so far (Figure 8). For sinusoidal motion (Figures 7A-B), the maximum error of predicted CTV ΔD95 was less than 2 percentage points. The median error after the first field was approximately 0.2 percentage points and decreased as the treatment progressed. For drift-free Calypso motion, the maximum predicted error was 6.3 percentage points, and the maximum median error after the second field was 0.4 percentage points (Figure 7C). In the case of Calypso motion accompanied by drift motion, the errors were larger, with a maximum prediction error of 12.6 percentage points and a median error of 1.3 percentage points after the first field, and these errors steadily decreased as more fields were delivered (Figure 7D).

[0063] Figure 6 (the top two curves) shows a simulation of dose guidance using inter-field couch correction, triggered when the estimated final CTV D95 improves by at least 4 percentage points, and tumor movement measured by Calypso. The top curve shows the continuously estimated final CTV D95 with optimal couch shift. The bottom curve shows the estimated final CTV D95 without couch shift. In this simulation, the continuously optimized D95 ​​(top) was more than 4 percentage points higher than the estimated D95 ​​without treatment change (bottom), so a dose-guided couch correction was triggered once after the second field (95 seconds after treatment initiation). The second curve from the top after couch correction shows the estimated final dose without further couch correction. Since this curve is always less than 4 percentage points lower than the curve above, no further couch correction was triggered in this simulation of dose guidance. For comparison, this figure also shows D95 of an alternative simulation using geometric guide couch correction, which resets the mean geometric error after the second field (third from the top). CTV DVH for various scenarios is shown immediately after couch correction (Figure 6B) and at the end of treatment (Figure 6C).

[0064] Figure 9 shows an example of a user interface for providing dose guidance using the method of this disclosure for dose guidance of a radiotherapy system during a radiotherapy session. At the bottom of the interface, the user can select the time for which the dose difference is displayed in the upper left view. In the upper left view, the user can see the dose difference from the pre-calculated planned dose. The dose difference is displayed as a heatmap of volume or cross-section of volume.

[0065] reference The following references are incorporated herein by reference in their entirety. International application PCT / DK2015 / 050251, “Method for real-time dose reconstruction during radio therapy”. Muurholm CG,Ravkilde T,Skouboe S,Worm E,Hansen R,Hoyer M,Keall PJ,Poulsen PR.“Real-time dose-guidance in radiotherapy:Proof of principle.” Radiother Oncol.2021 Nov;164:175-182.doi:10.1016 / j.radonc.2021.09.024.Epub 2021 Sep 28.PMID:34597738.

Claims

1. A decision support system for radiotherapy systems, - An interface for receiving radiotherapy beam parameters and / or parameters related to the observation and / or simulated motion of at least one volume, and - Equipped with a processing unit, the processing unit is - To perform dose reconstruction in virtually real time to obtain the radiation dose generated by the radiotherapy beam of the radiotherapy system in at least one volume, - Repeatedly estimating the remaining radiation dose of the radiotherapy session based on the observation and / or simulated movement of at least one volume, - The final radiation dose of the radiotherapy session is repeatedly estimated as the sum of the radiation dose and the estimated remaining radiation dose. A decision support system for the radiotherapy system, configured to re-estimate the remaining radiation dose for emulated interventions, such as repositioning objects and / or organs at risk to be exposed to the radiotherapy, and / or repositioning the beam, and / or re-optimizing the treatment plan for the remainder of the radiotherapy session.

2. A decision support system for a radiotherapy system according to claim 1, wherein the repeated estimation of the final radiation dose is performed substantially in real time.

3. A decision support system for a radiotherapy system according to claim 1 or 2, wherein the observed and / or simulated motion is obtained from a transponder and / or an external surrogate and / or imaging such as magnetic resonance imaging, X-ray imaging or optical imaging.

4. A decision support system for a radiotherapy system according to any one of claims 1 to 3, wherein the motion substantially includes a periodic motion portion such as sinusoidal motion, and / or baseline drift, and / or an irregular motion portion.

5. The decision support system for a radiotherapy system according to claim 4, wherein the repeated estimation of the final irradiation dose is based on the periodic motion portion and / or the baseline drift and / or the unstable motion portion.

6. A decision support system for a radiotherapy system according to any one of claims 1 to 5, wherein the processing unit is further configured to calculate the repositioning of an object exposed to radiotherapy and / or the repositioning of the radiotherapy beam in order to improve the estimated remaining radiation dose based on the irradiation radiation dose and the observed and / or simulated motion.

7. The decision support system for a radiotherapy system according to claim 6, wherein the repositioning of the object includes calculating a correction for the patient's couch.

8. A decision support system for a radiotherapy system according to any one of claims 1 to 7, wherein the processing unit is further configured to calculate the gating of the radiotherapy beam during the radiotherapy session based on the irradiation radiation dose and the observed and / or simulated motion.

9. A decision support system for a radiotherapy system according to any one of claims 1 to 8, wherein the processing unit is further configured to perform re-optimization of the treatment plan for the remaining portion of the radiotherapy session.

10. A decision support system for a radiotherapy system according to any one of claims 1 to 9, wherein the processing unit is further configured to re-estimate the remaining radiation dose for emulated repositioning of an object to be exposed to radiotherapy.

11. A decision support system for a radiotherapy system according to any one of claims 1 to 10, wherein the optimized position for the remainder of the radiotherapy session is extracted by an optimization algorithm.

12. The decision support system for a radiotherapy system according to claim 11, wherein the optimization algorithm minimizes a cost function representing dose loss and / or excessive dose to healthy tissue in at least one target within the volume.

13. A decision support system for a radiotherapy system according to any one of claims 6 to 12, wherein the processing unit is configured to repeatedly provide estimated final irradiation radiation doses with and without rearranging the object and / or the beam.

14. A decision support system for a radiotherapy system according to any one of claims 1 to 13, wherein the substantially real-time dose reconstruction described above is repeated when new input data is received.

15. A decision support system for a radiotherapy system according to claim 14, wherein the new input data includes parameters relating to the irradiated object or the apparatus for performing the radiotherapy session, such as parameters relating to the translational, rotational, and / or deformational motion of one or more targets or risk objects, and / or parameters relating to a stream of accelerator parameters and target position parameters.

16. A decision support system for a radiotherapy system according to any one of claims 1 to 15, wherein the observation of the target's movement includes movement in a time window preceding the time at which the remaining radiation dose is estimated.

17. A decision support system for a radiotherapy system according to any one of claims 1 to 16, wherein the radiotherapy session is a predetermined radiotherapy session associated with a predetermined planned dose to a target.

18. The decision support system for a radiotherapy system according to any one of claims 1 to 17, wherein the processing unit is configured to repeatedly estimate the final radiation dose to adjacent organs, such as organs at risk.

19. A decision support system for a radiotherapy system according to claim 18, wherein the processing unit is further configured to perform weighted optimization of interventions, such as repositioning of the object exposed to the radiotherapy and / or repositioning of the beam and / or reoptimization of the treatment plan for the remainder of the radiotherapy session, with respect to the final radiation dose to the target and / or adjacent organs(s).

20. The decision support system for a radiotherapy system according to any one of claims 1 to 19, wherein the processing unit is configured to repeatedly estimate the final radiation dose to multiple targets.

21. A decision support system for a radiotherapy system according to claim 20, wherein the processing unit is further configured to perform weighted optimization of interventions, such as repositioning an object exposed to the radiotherapy and / or repositioning the beam and / or reoptimizing the treatment plan for the remainder of the radiotherapy session, with respect to the final radiation dose of the target and the second target.

22. The decision support system for a radiotherapy system according to claim 1, wherein the processing unit is further configured to provide estimated final radiation doses with and without the intervention.

23. The step of performing substantially real-time dose reconstruction is a) The step of selecting one or more points in the space to be exposed to radiotherapy, b) The step of projecting at least one of the points onto at least one plane that intersects with the central axis of the radiotherapy beam projected from the radiotherapy beam source, c) A step of calculating the dose in the plane(s) generated by the radiotherapy beam, d) A step of depth scaling the dose in the plane from the projected point to the selected point, and obtaining the irradiation dose at the selected point, e) A step that repeats steps b) to d), A decision support system for a radiotherapy system according to any one of claims 1 to 22, including the above.

24. The decision support system for a radiotherapy system according to claim 23, wherein at least one of the planes intersecting the central axis of the radiotherapy beam is a plane perpendicular to the central axis of the radiotherapy beam.

25. A decision support system for a radiotherapy system according to claim 23 or 24, wherein the depth scaling of at least two selected points is based on doses calculated in one, two, three, or four planes.

26. A decision support system for a radiotherapy system according to any one of claims 23 to 25, wherein the processing unit is further configured to accumulate the calculated dose at the selected point.

27. A decision support system for a radiotherapy system according to any one of claims 23 to 26, wherein the beam source rotates in a circular manner around the target object, and / or the beam source moves in a non-coplanar manner around the target object, and / or the treatment couch moves relative to the beam source.

28. The decision support system for a radiotherapy system according to claim 27, wherein the plane is an isoplanar.

29. The projection point on the plane [Math 1] A decision support system for a radiotherapy system according to any one of claims 23 to 28, wherein the distance between the rotation axes of the sources SAD is the distance between the beam source and the plane, and the distance between the surfaces of the sources SSD is the distance between the source and the target object.

30. The aforementioned depth scaling, The steps include: calculating the absolute deep dose of the radiotherapy beam along its central axis as a function of the distance from the beam source; A step of calculating the dose irradiated to the selected point based on the dose in the plane and the absolute depth dose, A decision support system for a radiotherapy system according to any one of claims 23 to 29, which is implemented by performing the following:

31. A decision support system for a radiotherapy system according to claim 30, wherein the absolute deep dose on the central axis of the radiotherapy beam is calculated as a function of time.

32. A decision support system for a radiotherapy system according to claim 30 or 31, wherein the absolute deep dose on the central axis of the radiotherapy beam is calculated based on the measured deep dose percentage.

33. The absolute depth dose of the radiotherapy beam along the central axis [Math 2] It is expressed as, in the formula, [Math 3] This is Mayneold's F coefficient, [Math 4] The decision support system for a radiotherapy system according to any one of claims 30 to 32, wherein is an inverse square law correction coefficient.

34. A decision support system for a radiotherapy system according to any one of claims 23 to 33, wherein the dose irradiated to the selected point is calculated based on the dose in the plane and the deep dose based on the distance from the beam source.

35. The dose irradiated to the selected point is [Math 5] It is calculated as (optional), and in the formula, [Math 6] This is the dose within a plane, [Number 7] This is the absolute deep dose on the central axis of the radiotherapy beam, [Number 8] This is the dose rate, [Number 9] This is a density correction factor, [Number 10] A decision support system for a radiotherapy system according to any one of claims 23 to 34, wherein (optional parameter) is a gantry-dependent attenuation correction coefficient.

36. It is a computer program, When executed by a computing device or computing system, the computer has an instruction to cause the computing device or computing system to perform a method for dose guidance of a radiotherapy system during a radiotherapy session. The aforementioned method, The steps include performing substantially real-time dose reconstruction to obtain the radiation dose generated by the radiotherapy beam of the radiotherapy system in at least one volume, A step of repeatedly estimating the remaining radiation dose of the radiotherapy session based on the observation and / or simulated movement of at least one volume, The steps include repeatedly estimating the final radiation dose of the radiotherapy session as the sum of the radiation dose and the estimated remaining radiation dose, A step of providing dose guidance for the remainder of the radiotherapy session based on the estimated final radiation dose, The computer program includes the step of reestimating the remaining radiation dose for an emulated intervention, such as repositioning an object and / or organ at risk to be exposed to the radiotherapy, and / or repositioning the beam, and / or reoptimizing the treatment plan for the remainder of the radiotherapy session.

Images