Registration and treatment parameter adjustment method and system

Abstract

The embodiment of the invention provides a registration and treatment parameter adjustment method and system, and the method comprises the steps: obtaining a first image of a patient in response to the meeting of a first preset condition, and enabling the first image to comprise a treatment target region; registering the first image and a second image, wherein the second image is an image corresponding to the treatment plan of the patient; based on the registration result, treatment parameters are adjusted, and the treatment parameters are used for executing FLASH proton treatment.

Description

Technical Field

[0001] The present invention relates to the field of particle radiation therapy technology, and in particular to a registration and treatment parameter adjustment method and system. Background Art

[0002] During radiotherapy, the movement of the patient and the physiological movement and changes of the patient (such as positioning error, tumor changes, spontaneous breathing movement, small intestinal peristalsis, bladder filling, changes in pleural effusion, etc.) will cause changes in the treatment target space, thereby affecting the accuracy of the radiation in radiotherapy and causing unnecessary radiation to the patient. In addition, the current conventional radiotherapy method has a low dose and a long treatment time, and the physiological movement and changes of the patient may have an adverse effect on radiotherapy.

[0003] Therefore, there is an urgent need to propose a registration and treatment parameter adjustment method and system to further improve the treatment efficiency and accuracy of radiotherapy. Summary of the invention

[0004] One or more embodiments of the present specification provide a registration and treatment parameter adjustment method, comprising: in response to satisfying a first preset condition, acquiring a first image of a patient, the first image including a treatment target area; registering the first image with a second image, the second image being an image corresponding to the treatment plan of the patient; and adjusting treatment parameters based on the registration result, the treatment parameters being used to perform FLASH proton therapy.

[0005] In some embodiments, the space for acquiring the first image is a treatment room, and the space for acquiring the second image is a preparation room and a control room, and the treatment room, the preparation room, and the control room are independent spaces.

[0006] In some embodiments, the first preset condition includes that the patient's respiratory state is in a stable state, and the respiratory state is determined by: emitting a radiation beam to the treatment target area, obtaining the radiation value after the radiation beam passes through the treatment target area; and determining the respiratory state based on the radiation value.

[0007] In some embodiments, based on the registration result, the treatment parameters are adjusted, including: in response to the registration result being successful, determining whether the time safety margin of the current respiratory cycle meets the safety threshold, the time safety margin being determined based on the respiratory stabilization duration, the registration adjustment duration and the treatment duration of the current respiratory cycle; if satisfied, adjusting the treatment parameters based on the registration result; if not satisfied, waiting for the next respiratory cycle to obtain a new first image; in response to the registration result being a failure, determining whether the remaining duration of the current respiratory cycle meets the re-registration conditions; if satisfied, obtaining a new first image, and registering the new first image with the second image; if not satisfied, waiting for the next respiratory cycle to obtain a new first image.

[0008] In some embodiments, the first preset condition also includes that the predicted safety margin of this respiratory cycle satisfies a safety threshold, and the method for determining the predicted safety margin includes: determining the predicted safety margin of this respiratory cycle through a prediction model based on the patient's historical respiratory status information, the second image and historical registration data.

[0009] In some embodiments, the first preset condition also includes that the registration success prediction rate of this respiratory cycle meets the probability threshold, and the output of the prediction model also includes the registration success prediction rate.

[0010] In some embodiments, the method further includes: in response to failure of registration of the first image and the second image, predicting again whether the current respiratory cycle meets the first preset condition; if so, acquiring a new first image; if not, waiting for the next respiratory cycle to be re-predicted.

[0011] One of the embodiments of the present specification provides a registration and treatment parameter adjustment system, including: an acquisition module, in response to satisfying a first preset condition, for acquiring a first image of a patient, the first image including a treatment target area; a registration module, for registering the first image with a second image, the second image being an image corresponding to the patient's treatment plan; a treatment module, for adjusting treatment parameters based on the registration result, the treatment parameters being used to perform FLASH proton therapy.

[0012] In some embodiments, the first preset condition includes that the patient's respiratory state is in a stable state, and the system also includes a determination module for: emitting a radiation beam to the treatment target area, obtaining the radiation value after the radiation beam passes through the treatment target area; and determining the respiratory state based on the radiation value.

[0013] In some embodiments, the treatment module is further used to: in response to a successful registration result, determine whether the time safety margin of this respiratory cycle meets the safety threshold, the time safety margin is determined based on the respiratory stabilization time, the registration adjustment time and the treatment time of this respiratory cycle; if satisfied, adjust the treatment parameters based on the registration result; if not satisfied, wait for the next respiratory cycle and obtain a new first image; in response to a failed registration result, determine whether the remaining duration of this respiratory cycle meets the re-registration conditions; if satisfied, obtain a new first image, and align the new first image and the second image; if not satisfied, wait for the next respiratory cycle and obtain a new first image.

[0014] In some embodiments, the first preset condition also includes that the predicted safety margin of the current respiratory cycle meets a safety threshold, and the system also includes a prediction module for determining the predicted safety margin of the current respiratory cycle through a prediction model based on the patient's historical respiratory state information, the second image and historical registration data.

[0015] One or more embodiments of the present specification provide a registration and treatment parameter adjustment device, including a processor, wherein the processor is used to execute a registration and treatment parameter adjustment method.

[0016] One or more embodiments of the present specification provide a computer-readable storage medium, wherein the storage medium stores computer instructions. When a computer reads the computer instructions in the storage medium, the computer executes the registration and treatment parameter adjustment method. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 is a schematic diagram of an application scenario of a registration and treatment parameter adjustment system according to some embodiments of this specification;

[0018] Figure 2 is a schematic diagram of another application scenario of the registration and treatment parameter adjustment system shown in some embodiments of this specification;

[0019] Figure 3 is an exemplary module diagram of a registration and treatment parameter adjustment system according to some embodiments of this specification;

[0020] Figure 4 is an exemplary flow chart of a registration and treatment parameter adjustment method according to some embodiments of the present specification;

[0021] Figure 5 is an exemplary flow chart of adjusting treatment parameters based on registration results according to some embodiments of this specification;

[0022] Figure 6 is an exemplary schematic diagram of a prediction model according to some embodiments of the present specification. DETAILED DESCRIPTION

[0023] The following is a brief introduction to the drawings required for describing the embodiments. The drawings do not represent all implementation methods.

[0024] The "system", "device", "unit" and / or "module" used herein are a method for distinguishing different components, elements, parts, portions or assemblies at different levels. If other words can achieve the same purpose, the words can be replaced by other expressions.

[0025] As shown in this specification and claims, unless the context clearly indicates an exception, the words "a", "an", "an" and / or "the" do not refer to the singular and may also include the plural. Generally speaking, the terms "comprise" and "include" only indicate the inclusion of the steps and elements that have been clearly identified, and these steps and elements do not constitute an exclusive list. The method or device may also include other steps or elements.

[0026] When the operations performed in the embodiments of this specification are described in steps, unless otherwise specified, the order of the steps is interchangeable, steps may be omitted, and other steps may be included in the operation process.

[0027] Figure 1 It is a schematic diagram of an application scenario of the registration and treatment parameter adjustment system shown in some embodiments of this specification.

[0028] In some embodiments, the application scenario 100 of the registration and treatment parameter adjustment system may include a scanning device 110 , a processing device 120 , a storage device 130 , a terminal device 140 , a network 150 , and a treatment device 160 .

[0029] The scanning device 110 can scan the patient to obtain the scan data of the patient. In some embodiments, the scanning device 110 scans the treatment target area of ​​the patient to obtain an image containing the treatment target area.

[0030] In some embodiments, the scanning device 110 includes a permanent magnet magnetic resonance imaging device, a normal conductive magnetic resonance imaging device, a superconducting magnetic resonance imaging device, a nuclear magnetic resonance imaging device, an electron paramagnetic resonance imaging device, an electron spin resonance imaging device, and the like, and any combination thereof. In some embodiments, the scanning device 110 includes a single-modality scanner and / or a multi-modality scanner. For example, a single-modality scanner includes a nuclear magnetic resonance imaging (MRI) scanner. A multi-modality scanner includes an X-ray imaging-nuclear magnetic resonance imaging (X-ray-MRI) scanner, a single photon emission computed tomography-nuclear magnetic resonance imaging (SPECT-MRI) scanner, a digital subtraction angiography-nuclear magnetic resonance imaging (DSA-MRI) scanner, and the like, or any combination thereof. In some embodiments, the scanning device 110 includes an X-ray plain film, orthogonal X-ray imaging, a cone beam CT, and a static CT. The relevant description of the above-mentioned scanning device is for illustrative purposes only and is not intended to limit the scope of this specification.

[0031] The processing device 120 can process data and / or information obtained from other components (such as the scanning device 110, etc.) of the application scenario 100. For example, the processing device 120 obtains a treatment target area image of the patient from the scanning device 110, analyzes and processes it, and generates a treatment plan.

[0032] In some embodiments, the processing device 120 is a single server or a server group. The server group can be centralized or distributed. In some embodiments, the processing device 120 is local or remote. For example, the processing device 120 accesses information and / or data from the scanning device 110 via the network 150. For another example, the processing device 120 is directly connected to the scanning device 110 to access information and / or data. In some embodiments, the processing device 120 is implemented on a cloud platform. For example, the cloud platform includes a private cloud, a public cloud, a hybrid cloud, a community cloud, a distributed cloud, a cloud-to-cloud cloud, a multi-cloud, etc. or any combination thereof.

[0033] The storage device 130 may store data, instructions, and / or any other information. In some embodiments, the storage device 130 stores data acquired from the scanning device 110, the processing device 120, the terminal device 140, and the treatment device 160. In some embodiments, the storage device 130 stores data and / or instructions that the processing device 120 uses to execute or use to complete the exemplary methods described in this specification.

[0034] In some embodiments, storage device 130 includes one or more storage components, each of which is a separate device or part of another device (e.g., processing device 120). In some embodiments, storage device 130 includes random access memory (RAM), read-only memory (ROM), mass storage, removable memory, volatile read-write memory, etc. or any combination thereof. In some embodiments, storage device 130 is implemented on a cloud platform.

[0035] The terminal device 140 can communicate and / or connect with the scanning device 110, the processing device 120, and the storage device 130. In some embodiments, the interaction with the user is realized through the terminal device 140. For example, the medical staff uses the terminal device 140 to delineate the target area of ​​the patient's scanned image. For another example, the medical staff uses the terminal device 140 to adjust or confirm the treatment plan generated by the processing device 120.

[0036] In some embodiments, the terminal device 140 includes a mobile device, a tablet computer, a laptop computer, etc. or any combination thereof. In some embodiments, all or part of the functions of the terminal device 140 are integrated into the processing device 120 .

[0037] The network 150 may include any suitable network capable of facilitating information and / or data exchange. In some embodiments, at least one component of the application scenario 100 (e.g., the scanning device 110, the processing device 120, etc.) exchanges information and / or data with at least one other component via the network 150. For example, the processing device 120 acquires a treatment target area image of a patient from the scanning device 110 via the network 150.

[0038] Treatment device 160 refers to a device capable of performing FLASH proton therapy. FLASH proton therapy is an emerging treatment technology that uses protons to provide ultra-high dose rate radiation to treat tumors located deep in the body. In some embodiments, the radiation dose rate of FLASH proton therapy is 10Gy / s-1000Gy / s, and the single radiation dose is 0.1-80Gy.

[0039] Figure 2 It is a schematic diagram of another application scenario of the registration and treatment parameter adjustment system shown in some embodiments of this specification.

[0040] In some embodiments, the application scenario 200 of the registration and treatment parameter adjustment system may include a control room 210 , a preparation room 220 , a waiting room 230 , and a treatment room 240 .

[0041] The control room 210 is a space for centrally processing data and / or information obtained from other spaces in the application scene 200 (e.g., the preparation room 220, etc.). In some embodiments, the control room 210 includes a processing device 211, a storage device 212, and a terminal device 213. The processing device 211 is the same as the processing device 120, the storage device 212 is the same as the storage device 130, and the terminal device 213 is the same as the terminal device 140. For more information, see Figure 1 .

[0042] In some embodiments, the application scenario 200 includes only one control room 210, and the control room 210 includes only one processing device 211 for centrally processing data and / or information. In some embodiments, the control room 210 includes multiple terminal devices 213, and each medical staff is equipped with at least one terminal device 213. In some embodiments, the control room 210 does not include the terminal device 213, and the terminal devices 213 are dispersed in the preparation room 220, the waiting room 230, and the treatment room 240.

[0043] Preparation room 220 refers to a space for obtaining scan data of a patient. In some embodiments, preparation room 220 includes scanning device 221. Scanning device 221 is the same as scanning device 110. For more information, see Figure 1 .

[0044] In some embodiments, the application scenario 200 includes a plurality of preparation rooms 220 .

[0045] In some embodiments, the preparation room 220 is used to obtain scan data of the patient 222. In some embodiments, the scanning device 221 in the preparation room 220 first scans and images the patient 222 on the moving bed 223 to obtain scan data.

[0046] Scan data refers to data used to generate a treatment plan. In some embodiments, the scan data includes an image of a treatment target area, a location area of ​​the treatment target area and a location area of ​​a critical organ, and a relative position relationship between a patient and a mobile bed.

[0047] In some embodiments, control room 210 is used to generate a second image of patient 222 .

[0048] The second image is the image corresponding to the patient's treatment plan. The treatment plan refers to the radiation plan formulated for the treatment target area, including the location area of ​​the treatment target area, the location area of ​​the critical organ, and the dose, direction, shape, and radiation duration of the beam.

[0049] In some embodiments, the preparation room 220 transmits the acquired scan data to the control room 210, and the processing device 211 in the control room 210 processes the scan data to generate a treatment plan for the patient and a corresponding second image. In some embodiments, the processing device 211 directly processes the scan data and generates a treatment plan and a corresponding second image, and the medical staff manually adjusts or confirms the treatment plan and the corresponding second image through the terminal 213.

[0050] The waiting room 230 refers to a space where patients wait for radiation therapy after the treatment plan is completed.

[0051] In some embodiments, the application scenario 200 includes a plurality of waiting rooms 230 .

[0052] In some embodiments, the patient 222 is positioned immediately after completing the scan in the preparation room 220. In some embodiments, the patient 222 is positioned after being transferred to the waiting room 230. In some embodiments, the medical staff manually positions the patient 222. In some embodiments, a mechanical arm (not shown) is provided on the mobile bed 223, and the processor 211 issues a command to the mechanical arm, and the mechanical arm completes the positioning of the patient 222.

[0053] Treatment room 240 refers to a space in which FLASH proton therapy is performed on a patient. In some embodiments, treatment room 240 includes treatment device 241. Treatment device 241 is the same as treatment device 160. For more information, see Figure 1 .

[0054] In some embodiments, the application scenario 200 includes multiple treatment rooms 240 .

[0055] In some embodiments, the patient 222 is transferred between different spaces by a mobile bed 223. In some embodiments, different spaces are connected by rails, and the mobile bed 223 is moved by the rails to achieve the transfer of the patient. In some embodiments, the mobile bed 223 is equipped with wheels, and the mobile bed 223 is moved automatically or manually by medical staff to achieve the transfer of the patient.

[0056] In some embodiments, before radiotherapy, the patient 222 is transferred from the mobile bed 223 to the treatment bed 242 carried by the robot arm. In some embodiments, the mobile bed 223 includes various transport devices that can realize patient transfer.

[0057] In some embodiments, a scanning device (not shown) is disposed on the treatment bed 242. The scanning device disposed on the treatment bed 242 is the same as the scanning device 221 and the scanning device 110. For more information, see Figure 1 In some embodiments, before and during radiotherapy, a scanning device disposed on the treatment bed 242 acquires images of the patient in real time to monitor the location of the treatment target area and perform real-time registration. For more information on real-time registration, see Figure 4-Figure 6 .

[0058] In some embodiments, the robotic arm adjusts the treatment couch 242 based on instructions from the processing device 211 to ensure that the treatment target area is aligned with the treatment plan in real time.

[0059] In some embodiments, treatment room 240 is used to acquire a first image of the patient.

[0060] The first image is an image for registration acquired in real time before and during radiotherapy for the patient. The first image includes the treatment target area of ​​the patient.

[0061] In some embodiments, in response to satisfying the first preset condition, a scanning device disposed on the treatment bed 242 acquires a first image of the patient. For more information about the first preset condition, see Figure 4-Figure 6 .

[0062] In some embodiments of this specification, the entire treatment process is divided into different spaces, and non-treatment preparations (such as developing treatment plans, positioning, etc.) can be divided outside the treatment room, reducing the patient's stay time in the treatment room, thereby improving the use efficiency of large and expensive medical equipment (such as FLASH proton therapy equipment) and reducing medical costs. In addition, the data generated by multiple scanning devices and treatment equipment are uniformly processed by the processor in the control room, so that equipment with stronger processing capabilities can be purchased at the same cost. At the same time, since the computer processing time occupied by each patient is limited, centralized processing can improve the utilization rate of processing equipment.

[0063] It should be noted that the application scenarios are provided for illustrative purposes only and are not intended to limit the scope of this specification. For those of ordinary skill in the art, various modifications or variations can be made based on the description of this specification. For example, the application scenarios can be implemented on other devices to achieve similar or different functions. However, the changes and modifications will not deviate from the scope of this specification.

[0064] Figure 3 is an exemplary module diagram of a registration and treatment parameter adjustment system according to some embodiments of this specification. Figure 3 As shown, in some embodiments, the registration and treatment parameter adjustment system 300 includes an acquisition module 310 , a registration module 320 , and a treatment module 330 .

[0065] The acquisition module 310 is used to acquire a first image of the patient in response to satisfying the first preset condition, wherein the first image includes the treatment target area. For more information about the first image, please refer to Figure 2 ; For more information on the first precondition, see Figure 4-Figure 6 .

[0066] The registration module 320 is used to register the first image with the second image, where the second image is an image corresponding to the treatment plan of the patient. For more information about the second image and the treatment plan, please refer to Figure 2 ; For more information on registration, see Figure 4 .

[0067] In some embodiments, the space for acquiring the first image is a treatment room, the space for acquiring the patient's scan data is a preparation room, and the space for generating the second image is a control room. The treatment room, preparation room, and control room are independent spaces. For more information about treatment rooms, preparation rooms, and control rooms, please refer to Figure 2 .

[0068] The treatment module 330 is used to adjust the treatment parameters based on the registration results, and the treatment parameters are used to perform FLASH proton therapy. For more information about the registration results and treatment parameters, please refer to Figure 4 .

[0069] In some embodiments, the treatment module 330 is also used to: in response to the registration result being successful, determine whether the time safety margin of this respiratory cycle meets the safety threshold, and the time safety margin is determined based on the respiratory stabilization time, registration adjustment time and treatment time of this respiratory cycle; if satisfied, adjust the treatment parameters based on the registration result; if not satisfied, wait for the next respiratory cycle and obtain a new first image; in response to the registration result being a failure, determine whether the remaining time of this respiratory cycle meets the re-registration condition; if satisfied, obtain a new first image and align the new first image with the second image; if not satisfied, wait for the next respiratory cycle and obtain a new first image. For more information about time safety margin, safety threshold, respiratory stabilization time, registration adjustment time, treatment time, and re-registration conditions, please see Figure 5 .

[0070] In some embodiments, the registration and treatment parameter adjustment system 300 further includes a determination module 340 and a prediction module 350 .

[0071] In some embodiments, the first preset condition includes that the patient's respiratory state is in a stable state. The determination module 340 is used to determine the respiratory state, including: emitting a radiation beam to the treatment target area, obtaining the radiation value after the radiation beam passes through the treatment target area; and determining the respiratory state based on the radiation value. For more information about the respiratory state and stable state, please refer to Figure 4 .

[0072] In some embodiments, the first preset condition also includes that the predicted safety margin of the current respiratory cycle meets the safety threshold, and the prediction module 350 is used to predict the safety margin, including: based on the patient's historical respiratory state information, the second image and the historical registration data, through the prediction model, determining the predicted safety margin of the current respiratory cycle. For more information about the predicted safety margin, historical respiratory state information, historical registration data, and prediction model, please refer to Figure 6 .

[0073] In some embodiments, the first preset condition also includes that the registration success prediction rate of this respiratory cycle meets the probability threshold, and the output of the prediction model also includes the registration success prediction rate. For more information about the registration success prediction rate and probability threshold, please refer to Figure 6 .

[0074] In some embodiments, the prediction module 350 is further used to: in response to the failure of the registration of the first image and the second image, predict again whether the current respiratory cycle meets the first preset condition; if so, obtain a new first image; if not, wait for the next respiratory cycle to re-predict. For more information about re-prediction, please refer to Figure 6 .

[0075] It should be noted that the above description of the system and its modules is only for convenience of description and does not limit the present specification to the scope of the embodiments. It is understandable that, after understanding the principle of the system, those skilled in the art may arbitrarily combine the modules or form a subsystem to connect with other modules without deviating from the principle.

[0076] Figure 4 FIG. 1 is an exemplary flow chart of the registration and treatment parameter adjustment method according to some embodiments of the present specification. Figure 4 As shown, process 400 includes the following steps:

[0077] Step 410 , in response to satisfying a first preset condition, acquiring a first image of the patient, wherein the first image includes a treatment target area.

[0078] For more information about the first image, see Figure 2.

[0079] The first preset condition refers to a preset condition for determining whether to acquire the first image. For example, the first preset condition includes detecting that the patient has been transferred to the treatment bed.

[0080] In some embodiments, the first preset condition includes that the patient's respiratory state is in a stable state, and the determination module 340 determines the respiratory state in the following manner: emitting a radiation beam to the treatment target area, obtaining the radiation value after the radiation beam passes through the treatment target area; and determining the respiratory state based on the radiation value.

[0081] Respiratory status refers to the patient's breathing condition over a specific period of time, including characteristics such as respiratory rate, rhythm, depth, and sound.

[0082] The steady state means that the patient's breathing state remains relatively stable within a specific period of time without obvious changes. For example, the patient's inhalation or exhalation rate approaches 0, and the chest size hardly changes.

[0083] In some embodiments, the scanning device on the treatment bed 242 includes a transmitting source and a receiver. In some embodiments, the radiation beam emitted to the treatment target area is one or two beams, that is, not all radiation sources are required to emit radiation beams.

[0084] In some embodiments, the radiation value after the radiation beam passes through the treatment target area is obtained by the receiver. Since part of the energy of the radiation beam will be absorbed by the biological tissue after passing through the treatment target area, the radiation value obtained by the receiver is different from the radiation value at the emission source.

[0085] In some embodiments of the present specification, only one or two beams of radiation are emitted when determining the respiratory state, which can both achieve the determination of the respiratory state and avoid unnecessary radiation to the patient as much as possible.

[0086] Because the distribution of biological tissues in the thorax is different under different respiratory states, and the absorption of radiation beam doses is also different, the distribution of materials in the thorax can be solved based on the acquired radiation values, and then the size of the thorax can be determined. The size of the thorax reflects the amount of gas in the lungs, and thus the respiratory state can be determined.

[0087] In some embodiments, the determination module 340 determines the respiratory state based on the ray value by various feasible data analysis methods.

[0088] In some embodiments, the processor 120 constructs a "radiation value-respiratory state" comparison table based on historical data, stores it in the storage device 130, and updates it in real time or regularly. The determination module 340 determines the respiratory state based on the radiation value through the comparison table. In some embodiments, the comparison table is further subdivided based on the patient's basic information (such as gender, age, etc.) and the patient's condition information (such as disease type, etc.), for example, constructing a "basic information-condition information-radiation value-respiratory state" comparison table.

[0089] In some embodiments, the determination module 340 determines the respiratory state based on the ray value through a respiratory state determination model, and the respiratory state determination model is a machine learning model.

[0090] In some embodiments, the input of the respiratory state determination model includes ray values ​​and the output includes the respiratory state.

[0091] In some embodiments, the input of the respiratory state determination model also includes basic information of the patient and information about the patient's condition. For more information about the basic information of the patient and the patient's condition information, please refer to Figure 6 .

[0092] In some embodiments, the respiratory state determination model is trained separately based on historical data. In some embodiments, the respiratory state determination model is obtained based on the first training sample and the first label training. In some embodiments, the first training sample includes basic information, disease information, and radiation values ​​of multiple sample patients, and the first label is the respiratory state corresponding to the radiation value of the sample patient when the radiation value is obtained. The first training sample can be obtained based on historical data, and the first label can be determined by manual or automatic labeling. The above description is only an example and not a limitation, and the label of the training data can be obtained in various ways.

[0093] During training, the first training sample is input into the initial respiratory state determination model, a loss function is constructed based on the output of the initial respiratory state determination model and the first label, and the parameters of the initial respiratory state determination model are updated by the loss function until the trained initial respiratory state determination model meets the preset conditions, and the trained respiratory state determination model is obtained, wherein the preset conditions may be that the loss function is less than a threshold, converges, or the training cycle reaches a threshold, etc.

[0094] In some embodiments of the present specification, the respiratory state of the patient is determined by a machine learning model, which effectively improves the efficiency and accuracy of determining the respiratory state and avoids affecting the subsequent registration and treatment process. In addition, the input of the model further includes the basic information and condition information of the patient, which can minimize the error in the determination result due to differences between different individuals, so that the determined respiratory state is more consistent with the actual state of the currently detected patient.

[0095] In some embodiments, the determination module 340 obtains at least two radiation values ​​within a preset time period; in response to the difference between the two radiation values ​​satisfying a preset threshold, it is determined that the respiratory state is in a stable state.

[0096] The preset duration refers to the preset duration for obtaining the radiation value. For example, the preset time is 0.1s. In some embodiments, since the duration of a breathing cycle is short and the duration of breathing stability in a breathing cycle is even shorter, the preset duration needs to be as small as possible and needs to be able to complete two irradiations.

[0097] The preset threshold refers to the minimum value of the preset ray value difference. In some embodiments, the preset threshold approaches 0, which indicates that the respiratory state is in a stable state and the chest size has almost no change.

[0098] In some embodiments of the present specification, the respiratory state is judged according to the difference between two radiation values ​​detected within a preset time period, which can quickly determine whether the patient's respiratory state is in a stable state, thereby improving detection efficiency.

[0099] Step 420, registering the first image and the second image, the second image being an image corresponding to the treatment plan for the patient.

[0100] For more information about the second image, see Figure 2 .

[0101] In some embodiments, the registration module 320 registers the first image and the second image using various feasible image registration methods, such as grayscale-based image registration methods, transform domain-based image registration methods, image feature-based image registration methods, and the like.

[0102] Step 430 , based on the registration result, adjust the treatment parameters, the treatment parameters are used to perform FLASH proton therapy.

[0103] Registration results include success and failure.

[0104] In some embodiments, if the error between the first image and the second image after registration is less than an error threshold, the registration is considered successful, otherwise it is considered a failure.

[0105] Treatment parameters refer to parameters related to the status of the treatment device, such as the location, opening size, dose, and radiation duration of the radiation source of the FLASH proton therapy device.

[0106] In some embodiments, the treatment module 330 adjusts treatment parameters accordingly based on the registration results.

[0107] In some embodiments, in response to the registration result being successful, the treatment module 330 directly adjusts the treatment parameters based on the first image, the second image, and the treatment plan. For example, the treatment module 330 adjusts the position and opening size of the radiation source of the FLASH proton therapy device based on the first image and the second image to match the treatment target area. For another example, the treatment module 330 adjusts the dose and radiation duration of the radiation source of the FLASH proton therapy device based on the treatment plan.

[0108] In some embodiments, in response to the registration result being a failure, the registration is performed again until it succeeds, and the treatment parameters are directly adjusted based on the successfully registered first image, the second image, and the treatment plan. For more information on how to adjust the treatment parameters based on the registration result, see Figure 5 .

[0109] In some embodiments of the present specification, FLASH proton therapy is introduced to effectively improve the speed and efficiency of proton therapy by using protons to provide ultra-high dose rate radiation. At the same time, real-time alignment is performed before and during treatment to ensure the treatment accuracy of proton therapy and avoid unnecessary radiation to patients.

[0110] The embodiments in this specification are only for illustration and description, and do not limit the scope of application of this specification. For those skilled in the art, various modifications and changes that can be made under the guidance of this specification are still within the scope of this specification.

[0111] Figure 5 FIG. 1 is an exemplary flow chart of adjusting treatment parameters based on the registration results according to some embodiments of this specification. Figure 5 As shown, process 500 includes the following steps:

[0112] Step 510, determining whether the registration is successful.

[0113] In some embodiments, the registration module 320 directly determines the registration result based on the first image and the second image. For more information about registration and registration results, see Figure 4 .

[0114] Step 520, in response to the registration result being successful, determining whether the time safety margin of this respiratory cycle meets a safety threshold, the time safety margin being determined based on the respiratory stabilization duration, the registration adjustment duration, and the treatment duration of this respiratory cycle.

[0115] The time safety margin refers to the remaining time after successfully completing radiotherapy in a breathing cycle.

[0116] The safety threshold refers to the minimum value of the time safety margin in one respiratory cycle to ensure that the radiotherapy can be successfully completed. In some embodiments, the safety threshold is determined based on prior experience or historical data.

[0117] Respiratory stability time refers to the total time that the respiratory state is in a stable state during a respiratory cycle. For more information about respiratory state and stable state, please refer to Figure 4 .

[0118] In some embodiments, a preset number of respiratory cycles of the patient are pre-monitored by a respiratory monitoring device, and the respiratory stabilization duration of the current respiratory cycle is determined based on the respiratory stabilization duration of each respiratory cycle. For example, the average of the respiratory stabilization durations of each respiratory cycle is used as the respiratory stabilization duration of the current respiratory cycle.

[0119] In some embodiments, the prediction module 350 predicts the respiratory stability duration of the current respiratory cycle through a prediction model.

[0120] The prediction model refers to a model used to predict the stable duration of a respiratory cycle. The input of the prediction model may include historical respiratory status information, basic information of the patient, and information about the patient's condition, and the output may include the stable duration of breathing in this respiratory cycle.

[0121] In some embodiments, the prediction model is trained separately based on historical data. In some embodiments, the prediction model is obtained based on the second training sample and the second label training. In some embodiments, the second training sample includes basic information, disease information, and historical respiratory state information of multiple sample patients, and the second label is the respiratory stability duration of any respiratory cycle in the historical respiratory cycle of the sample patient. In some embodiments, the historical respiratory state information of the sample patient is the respiratory state information before the respiratory cycle corresponding to the second label. For example, the second label is the respiratory stability duration of the 15th respiratory cycle in the historical respiratory cycle of the sample patient, and the historical respiratory state information of the sample patient is the respiratory state information of the first 14 respiratory cycles. The second training sample can be obtained based on historical data, and the second label can be determined by manual annotation or automatic annotation. The above description is only an example, not a limitation, and the labels of the training data can be obtained in various ways.

[0122] For more information about the prediction model and its training method, historical respiratory status, basic patient information, and patient condition information, please see Figure 6 .

[0123] The breathing stability duration of this breathing cycle may also be determined by other methods.

[0124] In some embodiments of the present specification, a machine learning model is trained based on a large amount of historical data, and the machine learning model is used to predict the respiratory stabilization duration of the current respiratory cycle, which can improve the treatment efficiency while ensuring the accuracy of the prediction results.

[0125] Registration adjustment time refers to the total time it takes to complete image acquisition, image registration, and treatment parameter adjustment in one respiratory cycle. Registration adjustment time includes registration time and adjustment time. Registration time refers to the total time it takes to complete image acquisition, image registration, and output the registration result. Adjustment time refers to the total time it takes to complete treatment parameter adjustment based on a successful registration result.

[0126] In some embodiments, the processing device 120 obtains the registration duration of this respiratory cycle in real time. For example, the processing device 120 calculates the total time from issuing an instruction to obtain the first image to outputting the registration result in real time, and determines it as the registration duration of this respiratory cycle.

[0127] In some embodiments, the adjustment duration of this breathing cycle is determined based on historical data. For example, a preset number of historical adjustment durations are obtained, and a weighted average is calculated, and the calculation result is used as the adjustment duration of this breathing cycle. In some embodiments, the weight is related to the time interval between the historical data and the current time. The longer the time interval, the smaller the weight, and vice versa.

[0128] In some embodiments of the present specification, the weight for determining the adjustment duration is related to the time interval between the historical data and the current time, taking into account the age and degree of wear of the equipment. The closer the time is to the current time, the greater the weight, so that the final determined alignment adjustment duration can be more in line with the actual situation.

[0129] Treatment duration refers to the total duration of radiation therapy during one respiratory cycle. In some embodiments, treatment duration is determined based on a treatment plan. For more information about treatment plans, see Figure 2 .

[0130] In some embodiments, the time safety margin is determined based on the breathing stabilization duration, the registration adjustment duration, and the treatment duration of the current breathing cycle. For example, the time safety margin is obtained by subtracting the registration adjustment duration and the treatment duration from the breathing stabilization duration.

[0131] Step 530: If satisfied, adjust treatment parameters based on the registration result.

[0132] In some embodiments, if the registration result is successful and the time safety margin of the current respiratory cycle meets the safety threshold, the treatment module 330 directly adjusts the treatment parameters based on the first image, the second image and the treatment plan.

[0133] Step 540: If not, wait for the next breathing cycle to acquire a new first image.

[0134] In some embodiments, if the registration result is successful, but the time safety margin of the current breathing cycle does not meet the safety threshold, wait for the next breathing cycle, obtain a new first image, and re-execute step 510.

[0135] The respiratory cycle is the time from the start of one inhalation to the start of the next inhalation. The next respiratory cycle is the time from the start of the next inhalation to the start of the next inhalation.

[0136] The new first image refers to the reacquired first image. For more information about the first image, see Figure 2 .

[0137] Step 550: In response to the registration result being a failure, it is determined whether the remaining duration of the current respiratory cycle satisfies a re-registration condition.

[0138] The remaining time refers to the total time from when the registration result is determined to be a failure to the end of this respiratory cycle.

[0139] In some embodiments, the remaining time is determined based on the breathing stabilization time and the registration time. For example, the breathing stabilization time of this cycle is subtracted from the registration time to obtain the remaining time.

[0140] The re-registration condition means that the remaining time is greater than or equal to the sum of the safety threshold and the registration adjustment time.

[0141] In some embodiments, one respiratory cycle includes at least one registration adjustment. In some embodiments, the re-registration condition refers to the remaining time being greater than or equal to the sum of the safety threshold and the at least one registration adjustment time.

[0142] In some embodiments, the next registration adjustment duration of this respiratory cycle is determined based on historical data. For example, a preset number of historical registration adjustment durations are obtained, and a weighted average is calculated, and the calculation result is used as the registration adjustment duration of this respiratory cycle. In some embodiments, the weight is related to the time interval between the historical data and the current time. The longer the time interval, the smaller the weight, and vice versa.

[0143] In some embodiments of the present specification, it is determined whether the remaining time satisfies the re-registration condition. If so, the first image is acquired again and registered. Multiple registration adjustments can be made in one breathing cycle, thereby making full use of the time of this breathing cycle and improving treatment efficiency.

[0144] Step 560: If satisfied, obtain a new first image, and register the new first image with the second image.

[0145] In some embodiments, if the registration result is a failure, but the remaining duration of the current respiratory cycle satisfies the re-registration condition, a new first image is acquired, and the new first image and the second image are registered, and step 510 is re-executed.

[0146] The new registration method of the first image and the second image is related to the registration method of the first image and the second image. For more information, see Figure 4 .

[0147] Step 540: If not, wait for the next breathing cycle to acquire a new first image.

[0148] In some embodiments, if the registration result is a failure and the remaining duration of the current respiratory cycle does not satisfy the re-registration condition, step 540 is executed to wait for the next respiratory cycle, obtain a new first image, and then re-execute step 510.

[0149] In some embodiments of this specification, when determining whether to perform FLASH proton therapy, in addition to determining whether the registration is successful, the time safety margin is further considered to ensure that the radiotherapy can be successfully completed in this breathing cycle and avoid accidents during the treatment process (e.g., the patient starts the next breathing cycle during the treatment process, causing the treatment target area to shift, resulting in unnecessary radiation exposure to the non-treatment target area). In addition, when the registration result is a failure, it is further considered whether the current breathing cycle can be re-registered, making full use of the time of this breathing cycle and effectively improving the treatment efficiency.

[0150] Figure 6 is an exemplary schematic diagram of a prediction model according to some embodiments of the present specification.

[0151] In some embodiments, the first preset condition also includes that the predicted safety margin of the current respiratory cycle satisfies a safety threshold, and the prediction module 350 determines the predicted safety margin in the following manner: based on the patient's historical respiratory status information, the second image and historical registration data, through a prediction model, determining the predicted safety margin of the current respiratory cycle.

[0152] The predicted safety margin refers to the predicted value of the time safety margin of this breathing cycle predicted in advance.

[0153] For more information about the second image, see Figure 2 For more information about time safety margin and safety threshold, see Figure 5 .

[0154] The patient's historical respiratory status information refers to the patient's respiratory status information obtained in the historical time, including the respiratory status information of the few respiratory cycles closest to the current respiratory cycle. The patient's historical respiratory status information includes the total duration of each historical respiratory cycle, the duration of stable breathing, the respiratory rate, etc.

[0155] Historical registration data refers to the relevant information data in the historical registration process. Historical registration data includes the registration time, adjustment time, historical first image, historical second image, historical registration success probability, etc. of each historical registration.

[0156] In some embodiments, the prediction module 350 obtains the historical respiratory state information and the historical registration data from the storage device 130 , and the storage device 130 updates the historical respiratory state information and the historical registration data in real time or periodically.

[0157] Prediction model refers to the model used to determine the prediction safety margin.

[0158] In some embodiments, the prediction model may be various feasible neural network models, such as convolutional neural networks (CNN), deep neural networks (DNN), etc., or a combination thereof.

[0159] like Figure 6 As shown, the input of the prediction model 620 includes the patient's historical respiratory state information 611, the second image 612 and the historical registration data 613, and the output includes the predicted safety margin 631 of the current respiratory cycle.

[0160] In some embodiments, the input of the prediction model 620 also includes the patient's basic information 614 and the patient's condition information 615 .

[0161] The patient's basic information refers to the information reflecting the patient's basic personal situation. The patient's basic information includes gender, age, height, weight, etc.

[0162] The patient's medical condition information refers to information related to the patient's medical condition, including the type of disease, historical medical conditions, historical treatment conditions, family medical history, and current physiological indicators of various aspects of the body.

[0163] In some embodiments of the present specification, the predicted safety margin of the current respiratory cycle is predicted by the prediction model. If the prediction result does not meet the safety threshold, there is no need to obtain the first image for registration, which effectively avoids unnecessary radiation to the patient. In addition, the input of the prediction model further includes the basic information and condition information of the patient, which can minimize the situation where the determination result has errors due to differences between different individuals, so that the predicted safety margin is more in line with the actual condition of the current detected patient.

[0164] In some embodiments, the first preset condition also includes that the registration success prediction rate of the current respiratory cycle meets a probability threshold, and the output of the prediction model 620 also includes the registration success prediction rate 632.

[0165] The registration success prediction rate refers to the predicted value of the success probability of this registration result.

[0166] The probability threshold refers to a preset minimum value for obtaining a registration success prediction rate of the first image. In some embodiments, the probability threshold is determined based on prior experience or historical data.

[0167] In some embodiments of the present specification, the registration success prediction rate of this respiratory cycle is predicted by a prediction model. If the prediction result does not meet the probability threshold, there is no need to obtain the first image for registration, thereby effectively avoiding unnecessary radiation to the patient.

[0168] In some embodiments, the prediction model is trained separately based on historical data. In some embodiments, the prediction model is obtained based on the second training sample and the second label training. In some embodiments, the second training sample includes historical respiratory state information, basic information, disease information, a second image and sample historical registration data of multiple sample patients, and the second label is the time safety margin and registration result (success or failure) of any respiratory cycle in the historical respiratory cycle of the sample patient. In some embodiments, the historical respiratory state information of the sample patient is the respiratory state information before the respiratory cycle corresponding to the second label. For example, the second label is the respiratory stability duration of the 15th respiratory cycle in the historical respiratory cycle of the sample patient, and the historical respiratory state information of the sample patient is the respiratory state information of the first 14 respiratory cycles. The second training sample can be obtained based on historical data, and the second label can be determined by manual annotation or automatic annotation. The above description is only an example, not a limitation, and the label of the training data can be obtained in various ways.

[0169] During training, the second training sample is input into the initial prediction model, a loss function is constructed based on the output of the initial prediction model and the second label, and the parameters of the initial prediction model are updated through the loss function until the trained initial prediction model meets the preset conditions, and the trained prediction model is obtained, wherein the preset conditions may be that the loss function is less than a threshold, converges, or the training cycle reaches a threshold, etc.

[0170] In some embodiments, the output of the prediction model 620 also includes start instructions and wait instructions.

[0171] The start instruction refers to an instruction to start executing the FLASH proton therapy by adjusting the treatment parameters based on the registration result. In some embodiments, in response to the predicted safety margin meeting the safety threshold and the registration success prediction rate meeting the probability threshold, the prediction model 620 outputs the start instruction.

[0172] The waiting instruction refers to an instruction to wait for the next respiratory cycle to acquire a new first image or to wait for re-registration. In some embodiments, in response to the predicted safety margin not meeting the safety threshold, or the registration success prediction rate not meeting the probability threshold, the prediction model 620 outputs a waiting instruction.

[0173] In some embodiments, when the prediction model 620 outputs a start instruction or a wait instruction, the second label of the training prediction model 620 is whether any respiratory cycle in the historical respiratory cycle of the sample patient starts to execute the instruction of adjusting the treatment parameters based on the registration result and executing the FLASH proton therapy. If it is started, the second label is 1, and if it is not started, the second label is 0.

[0174] In some embodiments of the present specification, a start instruction or a wait instruction is directly outputted through a prediction model, so that the prediction result is more intuitive.

[0175] In some embodiments, in response to the failure of the registration of the first image and the second image, it is predicted again whether the current respiratory cycle meets the first preset condition; if so, a new first image is acquired; if not, waiting for the next respiratory cycle to be re-predicted. Figure 6 The same as above.

[0176] In some embodiments of the present specification, if the prediction result is incorrect (for example, the registration success prediction rate meets the probability threshold, but the actual registration fails), it can be predicted again in time to make full use of the time of this breathing cycle, improve treatment efficiency, and avoid unnecessary radiation to the patient as much as possible.

[0177] Furthermore, certain features, structures or characteristics in one or more embodiments of this specification may be appropriately combined.

[0178] In some embodiments, numbers describing the number of components and attributes are used. It should be understood that such numbers used in the description of the embodiments are modified by the modifiers "about", "approximately" or "substantially" in some examples. Unless otherwise specified, "about", "approximately" or "substantially" indicate that the numbers are allowed to vary by ±20%. Accordingly, in some embodiments, the numerical parameters used in the specification and claims are approximate values, which may change according to the required features of individual embodiments. In some embodiments, the numerical parameters should take into account the specified significant digits and adopt the general method of retaining digits. Although the numerical domains and parameters used to confirm the breadth of their range in some embodiments of this specification are approximate values, in specific embodiments, the setting of such numerical values ​​is as accurate as possible within the feasible range.

[0179] If there is any inconsistency or conflict between the descriptions, definitions, and / or usage of terms in the materials cited in this specification and the contents described in this specification, the descriptions, definitions, and / or usage of terms in this specification shall prevail.

Claims

1. A registration and treatment parameter adjustment method, characterized in that: include: In response to satisfying a first preset condition, acquiring a first image of the patient, wherein the first image includes a treatment target area; Registering the first image with a second image, wherein the second image is an image corresponding to the treatment plan of the patient; Based on the registration results, treatment parameters are adjusted, which are used to perform FLASH proton therapy.

2. The method according to claim 1, characterized in that The space for acquiring the first image is a treatment room, the space for acquiring the scan data of the patient is a preparation room, and the space for generating the second image is a control room. The treatment room, the preparation room, and the control room are independent spaces.

3. The method according to claim 1, characterized in that The first preset condition includes that the patient's respiratory state is in a stable state, and the method for determining the respiratory state includes: emitting a radiation beam to the treatment target area, and obtaining a radiation value of the radiation beam after the radiation beam passes through the treatment target area; Based on the radiation value, the respiratory state is determined.

4. The method according to claim 3, characterized in that The step of adjusting treatment parameters based on the registration result includes: In response to the registration result being successful, Determine whether the time safety margin of this respiratory cycle meets the safety threshold, wherein the time safety margin is determined based on the respiratory stabilization duration, the registration adjustment duration, and the treatment duration of this respiratory cycle; If satisfied, adjusting the treatment parameters based on the registration result; If not, wait for the next breathing cycle to obtain a new first image; In response to the registration result being a failure, Determine whether the remaining time of this respiratory cycle meets the re-registration condition; If satisfied, acquiring a new first image, and registering the new first image with the second image; If not satisfied, wait for the next breathing cycle to obtain a new first image.

5. The method according to claim 3, characterized in that The first preset condition also includes that the predicted safety margin of the current breathing cycle meets the safety threshold, and the method for determining the predicted safety margin includes: Based on the patient's historical respiratory state information, the second image and historical registration data, the predicted safety margin of this respiratory cycle is determined through a prediction model.

6. The method according to claim 5, characterized in that The first preset condition also includes that the registration success prediction rate of this respiratory cycle meets a probability threshold, and the output of the prediction model also includes the registration success prediction rate.

7. The method according to claim 6, characterized in that The method further comprises: In response to a failure in registration of the first image and the second image, predicting again whether the current respiratory cycle meets a first preset condition; If satisfied, obtain a new first image; If not satisfied, wait for the next breathing cycle to re-predict.

8. A registration and treatment parameter adjustment system, characterized in that: include: an acquisition module, in response to satisfying a first preset condition, for acquiring a first image of the patient, wherein the first image includes a treatment target area; a registration module, configured to register the first image with a second image, wherein the second image is an image corresponding to a treatment plan for the patient; The treatment module is used to adjust treatment parameters based on the registration result, wherein the treatment parameters are used to perform FLASH proton therapy.

9. The system according to claim 8, characterized in that The first preset condition includes that the patient's respiratory state is in a stable state, and the system further includes a determination module for: emitting a radiation beam to the treatment target area, and obtaining a radiation value of the radiation beam after the radiation beam passes through the treatment target area; Based on the radiation value, the respiratory state is determined.

10. The system according to claim 9, characterized in that The treatment module is further configured to: In response to the registration result being successful, Determine whether the time safety margin of this respiratory cycle meets the safety threshold, wherein the time safety margin is determined based on the respiratory stabilization duration, the registration adjustment duration, and the treatment duration of this respiratory cycle; If satisfied, adjusting the treatment parameters based on the registration result; If not, wait for the next breathing cycle to obtain a new first image; In response to the registration result being a failure, Determine whether the remaining time of this respiratory cycle meets the re-registration condition; If satisfied, acquiring a new first image, and registering the new first image with the second image; If not satisfied, wait for the next breathing cycle to obtain a new first image.

11. The system according to claim 9, characterized in that The first preset condition also includes that the predicted safety margin of the current breathing cycle meets the safety threshold, and the system also includes a prediction module for: Based on the patient's historical respiratory state information, the second image and historical registration data, the predicted safety margin of this respiratory cycle is determined through a prediction model.

12. A real-time registered high-dose rate proton therapy device, comprising a processor, wherein the processor is used to execute the registration and treatment parameter adjustment method according to any one of claims 1 to 7.

13. A computer-readable storage medium, wherein the storage medium stores computer instructions. When a computer reads the computer instructions in the storage medium, the computer executes the registration and treatment parameter adjustment method according to any one of claims 1 to 7.

Images