Adaptive radiotherapy plan determination method, adaptive radiotherapy system and storage medium

Abstract

The invention relates to a self-adaptive radiotherapy plan determination method, a self-adaptive radiotherapy system and a storage medium. The method comprises the following steps: acquiring a current region-of-interest segmentation result and current image features corresponding to a current image of an object; according to the segmentation result of the current region of interest and the current image features, candidate radiotherapy plans are selected from a plan library of the object, and the plan library comprises historical images of the object and corresponding historical radiotherapy plans; calculating a calculated dose distribution corresponding to the current image based on parameters in the candidate radiotherapy plan; determining the applicability of the candidate radiotherapy plan for the current image according to the calculated dose distribution, the segmentation result of the current region of interest and the features of the current image; and the target radiotherapy plan for the current image is determined based on the applicability, so that the generation efficiency and effectiveness of the adaptive radiotherapy plan are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of radiotherapy planning, and in particular to a method for determining an adaptive radiotherapy plan, an adaptive radiotherapy system and a storage medium. BACKGROUND

[0002] With research confirming that online adaptive radiotherapy technology can significantly improve the dose benefit of patients receiving radiotherapy, more and more medical institutions hope to introduce and practice online adaptive radiotherapy. However, the traditional online adaptive radiotherapy process requires technicians, physicists and doctors to participate synchronously on site, and because of its long time-consuming characteristics, it brings great pressure to the allocation of hospital human resources.

[0003] In view of the problem of low efficiency of adaptive radiotherapy plan generation in the related art, no effective solution has been proposed so far. SUMMARY

[0004] The present application provides a method for determining an adaptive radiotherapy plan, an adaptive radiotherapy system and a storage medium to solve the problem of low efficiency of adaptive radiotherapy plan generation in the related art.

[0005] In a first aspect, the present application provides a method for determining an adaptive radiotherapy plan, comprising:

[0006] obtaining a current region of interest segmentation result and a current image feature corresponding to a current image of a subject;

[0007] selecting a candidate radiotherapy plan from a plan library of the subject according to the current region of interest segmentation result and the current image feature, the plan library comprising historical images and corresponding historical radiotherapy plans of the subject;

[0008] calculating a calculated dose distribution corresponding to the current image based on parameters in the candidate radiotherapy plan;

[0009] determining the applicability of the candidate radiotherapy plan to the current image according to the calculated dose distribution, the current region of interest segmentation result and the current image feature; and

[0010] determining a target radiotherapy plan for the current image based on the applicability.

[0011] In one embodiment, determining the applicability of the candidate radiotherapy plan to the current image according to the calculated dose distribution, the current region of interest segmentation result and the current image feature; and determining a target radiotherapy plan for the current image based on the applicability, comprises:

[0012] determining an optimization probability that needs to be optimized for the candidate radiotherapy plan according to the calculated dose distribution, the current region of interest segmentation result and the current image feature;

[0013] recommending the candidate radiotherapy plan or the optimized historical radiotherapy plan as an optimal plan of the adaptive plan workflow according to the optimization probability.

[0014] In one of the embodiments, determining an optimization probability that needs to be optimized for the candidate radiotherapy plan according to the calculated dose distribution, the current region of interest segmentation result and the current image feature comprises:

[0015] setting a first event, a second event and a third event, wherein the first event represents directly taking the candidate radiotherapy plan as the target radiotherapy plan, the second event represents being able to determine the target radiotherapy plan through a simplified adaptive radiotherapy plan generation algorithm, and the third event represents needing to determine the target radiotherapy plan through a complete adaptive radiotherapy plan generation algorithm;

[0016] determining a classification probability corresponding to each type of event according to the calculated dose distribution, the current region of interest segmentation result and the current image feature; and

[0017] determining an optimization probability that needs to be optimized for the candidate radiotherapy plan according to the classification probability corresponding to each type of event.

[0018] In one of the embodiments, determining a classification probability corresponding to each type of event according to the calculated dose distribution, the current region of interest segmentation result and the current image feature comprises:

[0019] determining a classification probability corresponding to each type of event by outputting three nodes through a SOFTMAX activation function based on a deep learning model according to the current region of interest segmentation result, deformation registration information between the current image and a historical image corresponding to the candidate radiotherapy plan and the calculated dose distribution.

[0020] In one of the embodiments, determining an optimization probability that needs to be optimized for the candidate radiotherapy plan according to the classification probability corresponding to each type of event comprises:

[0021] if the output event of the deep learning model falls into the classification probability of each type of event and P(e1|x) is the highest, it is considered that the optimization probability belongs to a low probability;

[0022] if the output event of the deep learning model falls into the classification probability of each type of event and P(e2|x) is the highest, it is considered that the optimization probability belongs to a medium probability;

[0023] If the output event of the deep learning model falls into the classification probability of each type of event, P(e3|x) is the highest, it is considered that the optimization probability belongs to a high probability;

[0024] Wherein, e1 represents a first event, e2 represents a second event, e3 represents a third event, P(e1|x), P(e2|x), P(e3|x) correspond to the classification probability of each type of event, P(e1|x)+P(e2|x)+P(e3|x)=1 as a constraint condition.

[0025] In one of the embodiments, according to the optimization probability, the candidate radiotherapy plan or the optimized historical radiotherapy plan is recommended as the optimal plan of the adaptive plan workflow, including:

[0026] When the optimization probability does not exceed the preset probability threshold, the candidate radiotherapy plan is recommended as the optimal plan of the adaptive plan workflow;

[0027] When the optimization probability exceeds the preset probability threshold, the optimized historical radiotherapy plan is recommended as the optimal plan of the adaptive plan workflow.

[0028] In one of the embodiments, the historical radiotherapy plan is optimized, including:

[0029] When the optimization probability is within a first preset probability range, the historical radiotherapy plan determined based on the initial positioning image is taken as a reference plan;

[0030] The beam sub-field information in the reference plan is copied as the initial solution of the optimal plan;

[0031] Based on the target parameters for optimizing the reference plan, shape and jump number based optimization solution is performed on each beam sub-field in the reference plan, and the optimal plan is determined according to the solution result;

[0032] Or, when the optimization probability is within a second preset probability range, the historical radiotherapy plan determined based on the initial positioning image is taken as a reference plan;

[0033] Based on the target parameters for optimizing the reference plan, inverse rule optimization radiotherapy fluence is performed;

[0034] According to the optimized radiotherapy fluence, the shape and jump number of each beam sub-field in the reference plan are cut, and the optimal plan is determined based on the cutting result.

[0035] In one of the embodiments, according to the current region of interest segmentation result and the current image feature, a candidate radiotherapy plan is selected from the plan library of the object, including:

[0036] The current image, the current region of interest segmentation result and the current image feature are taken as inputs to traverse the plan library, and a historical radiotherapy plan corresponding to a historical image most similar to the current image in terms of the segmentation result and the image feature is determined in the plan library as the candidate radiotherapy plan.

[0037] In a second aspect, an adaptive radiotherapy system is provided in the embodiments, which comprises:

[0038] An image analysis module configured to determine a current region of interest segmentation result and a current image feature corresponding to a current image of a subject;

[0039] A candidate plan selection module configured to select a candidate radiotherapy plan from a plan library of the subject according to the current region of interest segmentation result and the current image feature, the plan library comprising historical images of the subject and corresponding historical radiotherapy plans;

[0040] A dose calculation module configured to calculate a calculated dose distribution corresponding to the current image based on parameters in the candidate radiotherapy plan; and

[0041] A plan evaluation module configured to determine an applicability of the candidate radiotherapy plan to the current image according to the calculated dose distribution, the current region of interest segmentation result and the current image feature;

[0042] A target plan selection module configured to determine a target radiotherapy plan for the current image based on the applicability.

[0043] In a third aspect, a computer readable storage medium is provided in the embodiments, which stores a computer program, and the computer program is executed by a processor to implement the steps of the method of the first aspect.

[0044] The adaptive radiotherapy plan determination method, the adaptive radiotherapy system and the storage medium have the following advantages. The current region of interest segmentation result and the current image feature are determined based on the current image, the candidate radiotherapy plan is selected from the plan library of the subject according to the current region of interest segmentation result and the current image feature, the calculated dose distribution corresponding to the current image is calculated based on the parameters in the candidate radiotherapy plan, the applicability of the candidate radiotherapy plan to the current image is determined according to the calculated dose distribution, the current region of interest segmentation result and the current image feature, and the target radiotherapy plan for the current image is determined based on the applicability, i.e., only the candidate radiotherapy plan or the optimized historical radiotherapy plan based on the applicability needs to be selected as the optimal plan of the adaptive plan workflow, and the treatment scheme does not need to be frequently re-evaluated and optimized, so that the generation efficiency and effectiveness of the adaptive radiotherapy plan are improved.

[0045] The details of one or more embodiments of the application are set forth in the accompanying drawings and the description below. Other features, objects, and advantages of the application will be apparent from the description and drawings, and from the claims. BRIEF DESCRIPTION OF DRAWINGS

[0046] The accompanying drawings are included to provide a further understanding of the application, and are incorporated in and constitute a part of this application, illustrate embodiments of the present application and serve to explain the present application, and do not limit the present application. In the drawings:

[0047] Figure 1 A hardware structure diagram of an electronic system in an embodiment;

[0048] Figure 2 A flow chart of an adaptive radiotherapy plan determination method in an embodiment;

[0049] Figure 3 A flow chart of an optimization probability calculation method in an embodiment;

[0050] Figure 4 A flow chart of an adaptive radiotherapy plan determination method in another embodiment;

[0051] Figure 5 A structure diagram of an adaptive radiotherapy system in an embodiment;

[0052] Figure 6 An internal structure diagram of a computer device in an embodiment. DETAILED DESCRIPTION

[0053] In order to more clearly understand the objects, technical solutions and advantages of the present application, the present application is described and explained in detail below in conjunction with the accompanying drawings and embodiments.

[0054] Unless otherwise defined, technical terms or scientific terms used in the present application shall have the same meaning as those commonly understood by a person of ordinary skill in the art to which the present application belongs. The terms "one", "a", "an", "the", "these", and similar terms in the present application do not indicate quantity of limitation, and they can be singular or plural. The terms "include", "contain", "have", and any variants thereof in the present application are intended to cover non-exclusive inclusion; for example, a process, method, and system, product or device containing a series of steps or modules (units) are not limited to the listed steps or modules (units), but can include steps or modules (units) not listed, or can include other steps or modules (units) inherent to the process, method, product or device. The terms "connect", "connected", "couple" and similar terms in the present application are not limited to physical or mechanical connection, but can include electrical connection, whether direct or indirect. The term "multiple" in the present application refers to two or more. The term "and / or" describes the association between the associated objects, which means that there can be three relationships, for example, "A and / or B" can mean that A exists alone, A and B exist together, and B exists alone. Generally, the character " / " represents an "or" relationship between the associated objects. The terms "first", "second", "third" and the like in the present application are only used to distinguish similar objects, and do not represent a specific order of the objects.

[0055] The method embodiments provided in the present embodiment can be executed in an electronic system (for example, arranged in or outside a medical device). Figure 1 is a hardware structure diagram of an electronic system of an embodiment of the present application. As shown in Figure 1 , the electronic system can include one or more (only one is shown in Figure 1 ) processors 101 and a memory 102 for storing data, wherein the processor 101 can include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA. The above-mentioned electronic system can also include a transmission device 103 for communication function and an input / output device 104. Those skilled in the art can understand that the structure shown in Figure 1 is only schematic, which does not limit the structure of the above-mentioned electronic system. For example, the electronic system can include more or less components than those shown in Figure 1 , or have a different configuration from that shown in Figure 1 .

[0056] The memory 102 can be used to store computer programs, such as software programs of application software and modules, such as the computer program corresponding to the adaptive radiotherapy plan determination method in the embodiment. The processor 101 executes various functional applications and data processing, i.e., implements the method described above, by running the computer program stored in the memory 102. The memory 102 can include a high-speed random access memory, and can further include a non-volatile memory, such as one or more magnetic storage devices, flash memories, or other non-volatile solid-state memories. In some examples, the memory 102 can further include memories disposed remotely with respect to the processor 101, which can be connected to the electronic system through a network. Examples of the network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0057] The transmission device 103 is configured to receive or send data via a network. The network includes a wireless network provided by a communication provider of the electronic system. In an example, the transmission device 103 includes a network interface controller (NIC) that can be connected to other network devices through a base station to communicate with the Internet. In an example, the transmission device 103 can be a radio frequency (RF) module configured to communicate with the Internet in a wireless manner.

[0058] In conventional radiotherapy, a treatment plan is usually developed based on one or several pre-treatment imaging examinations, and it is assumed that the anatomical structure of the subject remains unchanged during the entire treatment period. However, in actual situations, due to factors such as tumor shrinkage, organ movement (such as breathing and intestinal peristalsis), and physiological function (such as the filling degree of the bladder and stomach), the positions of the target region and the surrounding normal tissue can change, which can result in the actual irradiation region not completely matching the initial plan.

[0059] To address this problem, the related art proposes an adaptive radiotherapy (ART) technique, which allows adjusting the radiotherapy plan according to the real-time or time-varying physical conditions of the subject during the radiotherapy process, such as changes in tumor size and position, movement or filling state changes of normal organs. ART requires frequent re-evaluation and optimization of the treatment plan to improve treatment accuracy and reduce damage to surrounding normal tissue. However, the adaptive radiotherapy plan generation efficiency in the related art is low. To address this problem, in an embodiment, as shown in Figure 2 , an adaptive radiotherapy plan determination method is provided. Taking the electronic system in Figure 1 as an example, the method includes the following steps:

[0060] In step S201, a current region of interest segmentation result and a current image feature corresponding to a current image of an object are obtained.

[0061] The object can be a real human body or a human body model. The current image includes anatomical structure and / or functional and metabolic molecular information obtained by image scanning of the object. For example, the object can be scanned using medical imaging technology such as computed tomography (CT), magnetic resonance imaging (MRI), or positron emission computed tomography (PET), to obtain an anatomical structure image and / or functional and metabolic molecular information.

[0062] The region of interest (ROI) of particular interest in radiation delivery (such as a tumor region, a dangerous region around the tumor, etc.) can be identified and marked by threshold segmentation, region growing, edge detection, etc. on the current image. Further, image features are extracted by one or more of image segmentation, feature extraction, dimension reduction processing, feature selection, model establishment, etc. on the current image. The image features include, but are not limited to, one or more of morphological features, first-order statistical features, second-order advanced texture features, semantic features, non-semantic features, and agnostic features.

[0063] The morphological features are used to describe the geometric characteristics of the region of interest (ROI), such as volume, surface area, diameter, etc.

[0064] The first-order statistical features are used to reflect the symmetry, uniformity, and local intensity distribution change of the voxels, including median, mean, minimum, maximum, standard deviation, skewness, kurtosis, etc.

[0065] The second-order and high-order texture features reflect the spatial arrangement relationship between image voxels, such as gray level co-occurrence matrix (GLCM), gray level run-length matrix (GLRLM), gray level size zone matrix (GLSZM), and neighborhood gray tone difference matrix (NGTDM), etc.

[0066] The semantic features are features such as burr sign, indentation sign, scales, central necrotic area, etc. These features can be recorded by radiologists in image descriptions.

[0067] The non-semantic features are features extracted from images by mathematical methods, such as shape features, first-order, second-order, and high-order features, etc.

[0068] The agnostic features mainly refer to quantitative analysis of tumor heterogeneity using texture features and histograms.

[0069] Step S202, selecting a candidate radiotherapy plan from a plan library of the subject according to the current region of interest segmentation result and the current image feature, the plan library including historical images of the subject and corresponding historical radiotherapy plans.

[0070] The historical radiotherapy plan refers to a medical image sequence of each previous treatment of the subject.

[0071] In this step, the current image, the current region of interest segmentation result and the current image feature can be taken as inputs to traverse the plan library, and the historical radiotherapy plan corresponding to the historical image most similar to the current image in terms of segmentation result and image feature is determined in the plan library as the candidate radiotherapy plan. For example, the processor or the electronic system automatically compares the features of the current image with the historical images of the subject and the corresponding historical radiotherapy plans stored in the plan library, and selects the most relevant historical radiotherapy plan as the candidate radiotherapy plan based on a similarity measure (such as the matching degree of anatomical structure, the position change of the region of interest, etc.). The segmentation result refers to the generation of different regions by distinguishing different structures or tissues in the current image according to their characteristics (such as brightness, color, texture, etc.). Or further, different regions are contoured again. In this embodiment, unless otherwise stated, contouring and segmentation can be used interchangeably.

[0072] In this step, the candidate radiotherapy plan can also be selected from the plan library of the subject by the doctor according to experience or personal judgment.

[0073] Step S203, calculating a calculated dose distribution corresponding to the current image based on the parameters in the candidate radiotherapy plan.

[0074] The parameters include but are not limited to the number of beams, the size of beams, the angle of irradiation, the dose of beams, the number of machine jumps, the shape and / or size of the field, the time length, etc. Specifically, for each candidate radiotherapy plan, the dose distribution in the three-dimensional space, especially in the region of interest and the surrounding normal tissue, is calculated by setting the number of beams, the size of beams, the angle of irradiation, the dose of beams, the number of machine jumps, the shape and / or size of the field, the time length, etc.

[0075] Step S204, determining the applicability of the candidate radiotherapy plan to the current image according to the calculated dose distribution, the current region of interest segmentation result and the current image feature.

[0076] To evaluate the quality of the candidate radiotherapy plans, i.e., whether the dose distribution of each candidate plan meets the treatment goals, i.e., whether it effectively covers the tumor region while maximally sparing the surrounding healthy tissue, it is necessary to determine the suitability of the candidate radiotherapy plans for the current image. The suitability can include an indicator or metric that evaluates whether the candidate radiotherapy plan is suitable for the current image, such as a probability of needing to re-optimize the candidate radiotherapy plan based on the current image, a degree of matching between the candidate radiotherapy plan and the current image or the corresponding current state of the patient, etc. As an example, the suitability can be determined by any one or a combination of the following ways 1, 2.

[0077] Way 1 of determining the suitability: based on the calculated dose distribution, the current segmentation result of the region of interest, and the current image features, an optimization probability of needing to optimize the candidate radiotherapy plan is automatically determined (e.g., by an electronic system), and the suitability is determined based on the optimization probability. For example, when the optimization probability is small, it means that the candidate radiotherapy plan can be adopted without optimization, and its suitability is high. Conversely, when the optimization probability is large, it means that the candidate radiotherapy plan needs to be optimized before being adopted, and its suitability is low.

[0078] Way 2 of determining the suitability: a doctor evaluates the suitability of the candidate radiotherapy plan with respect to the current image based on the dose distribution, the current segmentation result of the region of interest, and the current image features. In particular, the doctor uses his professional experience to judge the degree of matching between the candidate radiotherapy plan and the current state of the subject. For example, a threshold of the degree of matching (such as 50%, 60%, or at least 90%) can be set as a criterion for recommending the radiotherapy plan. Further, even if the overall degree of matching is high, if it is found that there is a mismatch in a key region, especially when the irradiation region of the plan involves important organs such as the brain, the doctor can not recommend the candidate radiotherapy plan for safety considerations.

[0079] Step S205: determining the target radiotherapy plan for the current image based on the suitability.

[0080] Based on the suitability, it is determined whether the candidate radiotherapy plan can be used as the optimal plan of the adaptive plan workflow. If the suitability of the candidate radiotherapy plan for the current image is high, it means that the candidate radiotherapy plan meets the clinical requirements, and the candidate radiotherapy plan can be directly recommended as the optimal plan of the adaptive plan workflow. Conversely, if the suitability of the candidate radiotherapy plan for the current image is low, it means that the candidate radiotherapy plan does not meet the clinical requirements, and the optimized historical radiotherapy plan is recommended as the optimal plan of the adaptive plan workflow, so that the recommended plan meets the clinical requirements. It can be understood that in the case where the candidate radiotherapy plan meets the clinical requirements, the candidate radiotherapy plan is preferred to be recommended, and the plan does not need to be modified, which can reduce the time of online adaptive plan optimization. It should be noted that the target radiotherapy plan and the optimal plan can be used interchangeably.

[0081] Steps S201 to S205 above involve selecting candidate radiotherapy plans from the object's planning library based on the segmentation results of the current region of interest (ROI) and the current image features determined from the current image. Based on the parameters in the candidate radiotherapy plans, the calculated dose distribution corresponding to the current image is calculated. Based on the calculated dose distribution, the current ROI segmentation results, and the current image features, the applicability of the candidate radiotherapy plans to the current image is determined. Based on the applicability, the target radiotherapy plan for the current image is determined. In other words, it is only necessary to select the candidate radiotherapy plan or the optimized historical radiotherapy plan as the optimal plan for the adaptive planning workflow based on the applicability, without the need for frequent re-evaluation and optimization of the treatment plan, thus improving the efficiency and effectiveness of adaptive radiotherapy plan generation.

[0082] In addition, the treatment plan library can be a PotD (Plan of the Day) library. This PotD library can include medical images and corresponding radiotherapy plans for each treatment session. Therefore, by determining candidate radiotherapy plans based on this PotD library, it is possible to reduce the dose received outside the target area while meeting the required dose within the target area.

[0083] In one embodiment, the applicability of a candidate radiotherapy plan to the current image is determined based on the calculated dose distribution, the current region of interest segmentation result, and the current image features; and a target radiotherapy plan for the current image is determined based on the applicability, including: determining the optimization probability that the candidate radiotherapy plan needs to be optimized based on the calculated dose distribution, the current region of interest segmentation result, and the current image features; and recommending the candidate radiotherapy plan or the optimized historical radiotherapy plan as the optimal plan for the adaptive planning workflow based on the optimization probability.

[0084] Figure 3 The method for calculating the optimization probability is given, such as... Figure 3 As shown, based on the calculated dose distribution, the current region of interest segmentation results, and the current image features, the optimization probability for the candidate radiotherapy plan is determined, including the following steps:

[0085] Step S301: Set the first event, the second event, and the third event. The first event represents directly using the candidate radiotherapy plan as the target radiotherapy plan. The second event represents that the target radiotherapy plan can be determined by a simplified adaptive radiotherapy plan generation algorithm. The third event represents that the target radiotherapy plan needs to be determined by a complete adaptive radiotherapy plan generation algorithm.

[0086] In this step, three types of events can be set based on the segmentation results between historical radiotherapy plans and current images in the planning library, the deformation registration matrix, and the planning dosimetry parameters of the current images.

[0087] Step S302, according to the calculated dose distribution, the current region of interest segmentation result and the current image feature, the classification probability corresponding to each type of event is determined.

[0088] In this step, according to the current region of interest segmentation result, the deformation registration information between the current image and the historical image corresponding to the candidate radiotherapy plan, and the calculated dose distribution, three nodes are output by the SOFTMAX activation function based on the deep learning model, and the classification probability corresponding to each type of event is determined. The SOFTMAX activation function is used to convert the original score (also known as logits) into a value representing a probability distribution, so that the probability value of each category is between 0 and 1, and the sum of the probabilities of all categories is equal to 1.

[0089] Step S303, according to the classification probability corresponding to each type of event, the optimization probability of the candidate radiotherapy plan that needs to be optimized is determined.

[0090] If the output event of the deep learning model falls into the classification probability of each type of event, P(e1|x) is the highest, it is considered that the optimization probability belongs to the low probability; if the output event of the deep learning model falls into the classification probability of each type of event, P(e2|x) is the highest, it is considered that the optimization probability belongs to the medium probability; if the output event of the deep learning model falls into the classification probability of each type of event, P(e3|x) is the highest, it is considered that the optimization probability belongs to the high probability. Wherein, e1 represents the first event, e2 represents the second event, e3 represents the third event, P(e1|x), P(e2|x), P(e3|x) correspond to the classification probability of each type of event, P(e1|x)+P(e2|x)+P(e3|x) = 1 as a constraint condition.

[0091] In this embodiment, when the first event occurs, it represents that the candidate radiotherapy plan meets the clinical requirements, that is, the probability of re-optimization is low.

[0092] When the second event occurs, it represents that directly using the candidate radiotherapy plan as the target radiotherapy plan does not meet the clinical requirements, but the changes of the target region and the important OAR (Organ at Risk; OAR) of the historical radiotherapy plan and the current image are not large, so the target radiotherapy plan determined by the simplified adaptive radiotherapy plan generation algorithm meets the clinical requirements, and the probability of re-optimization is medium.

[0093] When the third event occurs, it represents that directly using the candidate radiotherapy plan as the target radiotherapy plan does not meet the clinical requirements, and the changes of the target region and the important OAR of the historical radiotherapy plan and the current image are large, so the target radiotherapy plan determined by the complete adaptive radiotherapy plan generation algorithm meets the clinical requirements, and the probability of re-optimization is high.

[0094] In one embodiment, the candidate radiotherapy plan or the optimized historical radiotherapy plan is recommended as the optimal plan of the adaptive plan workflow according to the optimization probability, including:

[0095] When the optimization probability does not exceed the preset probability threshold, the candidate radiotherapy plan is recommended as the optimal plan of the adaptive plan workflow; when the optimization probability exceeds the preset probability threshold, the optimized historical radiotherapy plan is recommended as the optimal plan of the adaptive plan workflow.

[0096] It can be understood that when the optimization probability does not exceed the preset probability threshold, the probability of re-optimization is low, which represents that the candidate radiotherapy plan meets the clinical requirements. When the optimization probability exceeds the preset probability threshold, the probability of re-optimization is medium or high, which represents that directly using the candidate radiotherapy plan as the target radiotherapy plan does not meet the clinical requirements, and the historical radiotherapy plan needs to be optimized to meet the clinical requirements.

[0097] In one embodiment, the historical radiotherapy plan is optimized. The optimization can be performed by one of the following schemes 1 and 2 or a combination thereof:

[0098] Optimization scheme 1: when the optimization probability is within a first preset probability range, the historical radiotherapy plan determined based on the initial positioning image is taken as a reference plan; the beam sub-field information in the reference plan is copied as an initial solution of the optimal plan; based on the target parameters used for optimizing the reference plan, shape and jump-based optimization solution is performed on each beam sub-field in the reference plan, and the optimal plan is determined according to the solution result. In this scheme, when the optimization probability is within the first preset probability range, it represents that the ROI deformation degree (Dice) of the current image is low, and the simplified adaptive algorithm historical radiotherapy plan can meet the clinical requirements. The optimization time of the simplified adaptive algorithm is short, which can save the plan optimization time.

[0099] Optimization scheme 2: when the optimization probability is within a second preset probability range, the historical radiotherapy plan determined based on the initial positioning image is taken as a reference plan; based on the target parameters used for optimizing the reference plan, the radiotherapy fluence is optimized according to the inverse rule; according to the optimized radiotherapy fluence, the shape and jump of each beam sub-field in the reference plan are cut, and the optimal plan is determined based on the cutting result. In this scheme, when the optimization probability is within the second preset probability range, it represents that the ROI deformation degree (Dice) of the current image is high, and the complete adaptive algorithm needs to be used to optimize the historical radiotherapy plan to meet the clinical requirements. The optimization time of the complete adaptive algorithm is long, and the modification range of the historical radiotherapy plan is large.

[0100] Traditional adaptive radiotherapy plan determination methods mainly select appropriate adaptive plan logic according to the changes of anatomical structures on the current image and the original image, to a certain extent, optimize the selection speed of online adaptive plan. However, for clinical users, it is still necessary to review the accuracy of segmentation and the effectiveness of the plan on site, and the convenience is still insufficient. Related technologies lack a smooth and intelligent adaptive plan workflow, which cannot relieve the pressure of hospitals to carry out adaptive radiotherapy.

[0101] To solve the above problems, in one embodiment, Figure 4 A flowchart of an adaptive radiotherapy plan determination method is provided. As shown in Figure 4 The electronic system obtains a current image obtained by image scanning of an object, and a processing result obtained by processing the current image, including the following steps:

[0102] Step S401, through IGRT (Image-Guided Radiation Therapy), a current image obtained by medical image scanning of an object is obtained.

[0103] Step S402, pre-processing the current image to obtain a pre-processing result. That is, determining the current region of interest segmentation result and the current image feature based on the current image.

[0104] Step S403, send the notification of needing to perform adaptive plan workflow and the pre-processing result to the target user end. The target user can include doctors, physicists, technicians, nurses and other medical staff, and the target user end can be a mobile phone, a computer, a tablet computer, a smart watch and the like.

[0105] Step S404, receive the delineation information determined by the target user end based on the pre-processing result, and update the pre-processing result based on the delineation information to obtain a final processing result.

[0106] Step S405, according to the final processing result, select a candidate radiotherapy plan from the plan library of the object. The plan library includes historical images and corresponding historical radiotherapy plans of the object, and the historical radiotherapy plan can be a radiotherapy plan generated based on the first positioning CT after diagnosis of the object.

[0107] Step S406, based on the parameters in the candidate radiotherapy plan, calculate the calculated dose distribution corresponding to the current image.

[0108] Step S407, according to the calculated dose distribution, the current region of interest segmentation result and the current image feature, determine the optimization probability of the candidate radiotherapy plan that needs to be optimized; if the optimization probability belongs to the low probability, execute step S408; if the optimization probability belongs to the medium probability, execute step S409; if the optimization probability belongs to the high probability, execute step S410;

[0109] Step S408, the recommended candidate radiotherapy plan is the optimal plan of the adaptive plan workflow.

[0110] Step S409, determining the target radiotherapy plan based on the simplified adaptive radiotherapy plan generation algorithm. Specifically, the historical radiotherapy plan determined based on the initial positioning image is taken as the reference plan; the beam sub-field information in the reference plan is copied as the initial solution of the optimal plan; based on the target parameters for optimizing the reference plan, shape and number of jumps based optimization solution is performed on each beam sub-field in the reference plan, and the optimal plan is determined according to the solution result.

[0111] Step S410, determining the target radiotherapy plan based on the complete adaptive radiotherapy plan generation algorithm. Specifically, the historical radiotherapy plan determined based on the initial positioning image is taken as the reference plan; based on the target parameters for optimizing the reference plan, the radiotherapy fluence is optimized according to the inverse rule; according to the optimized radiotherapy fluence, the shape and number of jumps of each beam sub-field in the reference plan are cut, and the optimal plan is determined based on the cutting result.

[0112] Step S411, receiving the evaluation result returned by the target user end, which can include evaluation information of whether the dose of the optimal plan meets the clinical requirements, and / or evaluation information of whether the parameters of the optimal plan meet the clinical specifications.

[0113] Step S412, determining whether to approve the optimal plan to pass according to the evaluation result.

[0114] In this embodiment, after entering the adaptive plan workflow, doctors, physicists and technicians can operate and communicate online, for example, the doctor end and the physicist end can enter the process through a remote way without going to the treatment room on site to participate, which can reduce unnecessary time. According to the automatic delineation information in the pre-processing result, the doctor and the physicist can modify the target volume and the organ at risk simultaneously, improve the modification efficiency of the target volume and the OAR in the adaptive process, shorten the use time of ART, and confirm entering the next step after modification. Further, after the optimal plan is recommended, the electronic system can also receive the evaluation result returned by the target user end, which includes evaluation information of whether the dose of the optimal plan meets the clinical requirements, and evaluation information of whether the parameters of the optimal plan meet the clinical specifications; whether to approve the optimal plan to pass is determined according to the evaluation result. In this embodiment, the doctor can online evaluate whether the dose of the optimal plan meets the clinical requirements, and the physicist can online evaluate whether the parameters of the optimal plan meet the clinical specifications. If both meet, it is confirmed to enter the next step so as to approve the optimal plan to pass, which can be used for the treatment of the object.

[0115] In one embodiment, Figure 5A structural diagram of an adaptive radiotherapy system is provided, which includes an image analysis module 501, a candidate plan selection module 502, a dose calculation module 503, a plan evaluation module 504 and a target plan selection module 505.

[0116] The image analysis module 501 is configured to determine a current region of interest segmentation result and a current image feature corresponding to a current image of a subject; the candidate plan selection module 502 is configured to select a candidate radiotherapy plan from a plan library of the subject according to the current region of interest segmentation result and the current image feature, the plan library including historical images of the subject and corresponding historical radiotherapy plans; the dose calculation module 503 is configured to calculate a calculated dose distribution corresponding to the current image based on parameters in the candidate radiotherapy plan; the plan evaluation module 504 is configured to determine an applicability of the candidate radiotherapy plan to the current image according to the calculated dose distribution, the current region of interest segmentation result and the current image feature; and the target plan selection module 505 is configured to determine a target radiotherapy plan for the current image based on the applicability.

[0117] The above-mentioned modules in the adaptive radiotherapy system can be implemented by software, hardware or a combination thereof in whole or in part. The above-mentioned modules can be embedded in or independent of a processor in a computer device in hardware form, or stored in a memory in the computer device in software form, so as to be called and executed by the processor to perform the operations corresponding to the above-mentioned modules.

[0118] In an embodiment, a computer readable storage medium is provided, which stores a computer program, and the computer program is executed by a processor to implement the steps of any of the above-mentioned adaptive radiotherapy plan determination methods.

[0119] For example, a current region of interest segmentation result and a current image feature corresponding to a current image of a subject are acquired;

[0120] A candidate radiotherapy plan is selected from a plan library of the subject according to the current region of interest segmentation result and the current image feature, the plan library including historical images of the subject and corresponding historical radiotherapy plans;

[0121] A calculated dose distribution corresponding to the current image is calculated based on parameters in the candidate radiotherapy plan;

[0122] An applicability of the candidate radiotherapy plan to the current image is determined according to the calculated dose distribution, the current region of interest segmentation result and the current image feature; and a target radiotherapy plan for the current image is determined based on the applicability.

[0123] In an embodiment, a computer device is provided, which can be a terminal, and an internal structural diagram of the computer device can be as shown in Figure 6As shown in the figure. The computer device includes a processor, a memory, a communication interface, a display unit and an input device connected by a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium, an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operating system and the computer program in the non-volatile storage medium to run. The communication interface of the computer device is used for wired or wireless communication with external terminals. Wireless mode can be achieved through WIFI, mobile cellular network, NFC (near field communication) or other technologies. The computer program is executed by the processor to implement an adaptive radiotherapy plan determination method. The display unit of the computer device can be a liquid crystal display unit or an electronic ink display unit. The input device of the computer device can be a touch layer overlaid on the display unit, or a key, trackball or touchpad provided on the computer device shell, or an external keyboard, touchpad or mouse, etc.

[0124] Those skilled in the art can understand that, Figure 6 The skilled in the art can understand that,

[0125] It should be understood that the specific embodiments described above are only used to explain the related application, but not to limit it. According to the embodiments provided in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0126] Obviously, the drawings are only some examples or embodiments of the present application, and those skilled in the art can also apply the present application to other similar situations according to the drawings without creative labor. In addition, it can be understood that, although the work done in the development process may be complex and long, for those skilled in the art, some design, manufacture or production changes according to the technical content disclosed in the present application are only routine technical means, and should not be regarded as insufficient disclosure of the present application.

[0127] The word "embodiment" in this application refers to the specific features, structures or characteristics described in conjunction with the embodiments can be included in at least one embodiment of the application. The phrase appears in various places in the specification does not necessarily mean the same embodiment, nor does it mean independence or alternative to other embodiments. It is clear or implicitly understood by those skilled in the art that the embodiments described in this application can be combined without conflict with other embodiments.

[0128] It should be noted that the user information (including but not limited to user equipment information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or authorized by all parties. The acquisition, storage, use, processing, etc. of data in the embodiments of the application comply with the relevant provisions of national laws and regulations.

[0129] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer readable storage medium, and when executed, can include the processes of the above-mentioned embodiment methods. Any reference to memory, database or other medium used in the embodiments provided by the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical storage, high-density embedded non-volatile memory, resistive memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration but not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc. The database involved in the embodiments provided by the present application can include at least one of a relational database and a non-relational database. The non-relational database can include a distributed database based on a block chain, etc., without being limited thereto. The processor involved in the embodiments provided by the present application can be a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., without being limited thereto.

[0130] The above-mentioned embodiments only express several implementation manners of the present application, and the description is relatively specific and detailed, but it should not be understood as a limitation on the scope of patent protection. It should be noted that for those skilled in the art, without departing from the concept of the present application, a number of modifications and improvements can be made, which are all within the scope of protection of the present application. Therefore, the scope of protection of the present application should be subject to the appended claims.

Claims

1. An adaptive radiotherapy planning method, characterized in that, include: Obtain the current region of interest segmentation result and current image features corresponding to the current image of the object; Based on the current region of interest segmentation results and current image features, candidate radiotherapy plans are selected from the planning library of the object, which includes historical images of the object and corresponding historical radiotherapy plans; Based on the parameters in the candidate radiotherapy plan, calculate the calculated dose distribution corresponding to the current image; Based on the calculated dose distribution, the current region of interest segmentation results, and the current image features, the applicability of the candidate radiotherapy plan to the current image is determined; as well as Based on the applicability, a target radiotherapy plan is determined for the current image.

2. The adaptive radiotherapy planning method according to claim 1, characterized in that, Based on the calculated dose distribution, the current region of interest segmentation results, and the current image features, the applicability of the candidate radiotherapy plan to the current image is determined; And determining a target radiotherapy plan for the current image based on the applicability, including: Based on the calculated dose distribution, the current region of interest segmentation results, and the current image features, the optimization probability that needs to be optimized for the candidate radiotherapy plan is determined; Based on the optimization probability, the candidate radiotherapy plan or the optimized historical radiotherapy plan is recommended as the optimal plan for the adaptive planning workflow.

3. The adaptive radiotherapy planning method according to claim 2, characterized in that, Based on the calculated dose distribution, the current region of interest segmentation results, and the current image features, the optimization probability for optimizing the candidate radiotherapy plan is determined, including: The first event, the second event, and the third event are defined, wherein the first event represents directly using the candidate radiotherapy plan as the target radiotherapy plan, the second event represents that the target radiotherapy plan can be determined by a simplified adaptive radiotherapy plan generation algorithm, and the third event represents that the target radiotherapy plan needs to be determined by a complete adaptive radiotherapy plan generation algorithm. Based on the calculated dose distribution, the current region of interest segmentation results, and the current image features, the classification probabilities for each type of event are determined; and Based on the classification probabilities of the corresponding events, the optimization probability that needs to be optimized for the candidate radiotherapy plan is determined.

4. The adaptive radiotherapy planning method according to claim 3, characterized in that, Based on the calculated dose distribution, the current region of interest segmentation result, and the current image features, the classification probability of each type of event is determined, including: Based on the current region of interest segmentation result, the deformation registration information between the current image and the historical images corresponding to the candidate radiotherapy plan, and the calculated dose distribution, a deep learning model is used to output three nodes through the SOFTMAX activation function to determine the classification probability of each type of event.

5. The adaptive radiotherapy planning method according to claim 4, characterized in that, Based on the classification probabilities of various events, the optimization probabilities that require optimization of the candidate radiotherapy plan are determined, including: If the output event of the deep learning model falls into the classification probability of each type of event and P(e1|x) is the highest, then the optimized probability is considered to be a low probability. If the output event of the deep learning model falls into the classification probability of each type of event and P(e2|x) is the highest, then the optimized probability is considered to be of medium probability. If the output event of the deep learning model falls into the classification probability of each type of event and P(e3|x) is the highest, then the optimized probability is considered to be of high probability. Where e1 represents the first event, e2 represents the second event, e3 represents the third event, P(e1|x), P(e2|x), and P(e3|x) correspond to the classification probabilities of each type of event, and P(e1|x)+P(e2|x)+P(e3|x)=1 serves as a constraint condition.

6. The adaptive radiotherapy planning method according to claim 2, characterized in that, Based on the optimization probability, the candidate radiotherapy plan or the optimized historical radiotherapy plan is recommended as the optimal plan for the adaptive planning workflow, including: When the optimization probability does not exceed a preset probability threshold, the candidate radiotherapy plan is recommended as the optimal plan for the adaptive planning workflow; When the optimization probability exceeds the preset probability threshold, the optimized historical radiotherapy plan is recommended as the optimal plan for the adaptive planning workflow.

7. The adaptive radiotherapy planning method according to claim 6, characterized in that, Optimize the aforementioned historical radiotherapy plan, including: When the optimization probability is within the first preset probability range, the historical radiotherapy plan determined based on the initial positioning image will be used as a reference plan; The beam subfield information in the reference plan is copied as the initial solution of the optimal plan; Based on the target parameters used to optimize the reference plan, an optimization solution based on shape and hop count is performed for each beam subfield in the reference plan, and the optimal plan is determined based on the solution results; Alternatively, when the optimization probability is within the second preset probability range, the historical radiotherapy plan determined based on the initial positioning image is used as a reference plan; Based on the target parameters used to optimize the reference plan, the radiotherapy flux is optimized according to the inverse rule; Based on the optimized radiotherapy flux, the shape and number of hops of each subfield in the reference plan are cut, and the optimal plan is determined based on the cutting results.

8. The adaptive radiotherapy planning method according to claim 1, characterized in that, Based on the current region of interest segmentation results and current image features, candidate radiotherapy plans are selected from the planning library for the object, including: Using the current image, the current region of interest segmentation result, and the current image features as input, the radiotherapy plan library is traversed to determine the historical radiotherapy plan corresponding to the historical image that is most similar to the current image in terms of segmentation result and image features, and this plan is used as the candidate radiotherapy plan.

9. An adaptive radiotherapy system, comprising: The image analysis module is configured to determine the current region of interest segmentation result and current image features corresponding to the current image of the object; The candidate plan selection module is configured to select candidate radiotherapy plans from the plan library of the object based on the current region of interest segmentation results and current image features. The plan library includes historical images of the object and corresponding historical radiotherapy plans. The dose calculation module is configured to calculate the calculated dose distribution corresponding to the current image based on parameters in the candidate radiotherapy plan; as well as The planning evaluation module is configured to determine the suitability of the candidate radiotherapy plan for the current image based on the calculated dose distribution, the current region of interest segmentation results, and the current image features. The target plan selection module is configured to determine a target radiotherapy plan for the current image based on the applicability.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 8.

Images