Recommendation Method, Device, Computer Equipment and Storage Medium for Lung Cancer Treatment Plan

Abstract

This application relates to a method, device, computer device, and storage medium for recommending a lung cancer treatment plan. The method includes: obtaining four-dimensional medical images of a target object, and determining a plurality of functional area images based on the four-dimensional medical images, where the four-dimensional medical images are images including the lungs of the target object, and the plurality of functional area images are images of multiple functional areas of the lungs; determining the target treatment category of each functional area based on the plurality of functional area images, the four-dimensional medical images, and a scoring model; and determining the target treatment plan of the target object based on the target treatment category of each functional area and the four-dimensional medical images. By adopting this solution, a suitable target treatment plan can be recommended for the target object, providing richer reference information for clinicians and helping doctors determine the final treatment plan.

Description

Technical Field

[0001] This application relates to the technical field of medical image processing, and particularly to methods, devices, computer equipment, and storage media for lung cancer treatment plans. Background Art

[0002] Before lung cancer treatment for lung cancer patients, it is necessary to detect the lungs of the patients to understand their lung function in order to determine whether the patients are suitable for treatments such as surgery and radiotherapy. The existing lung detections mainly rely on multi-channel electrophysiological recorders or pulmonary function testers to determine the functional parameters of the lungs, and then doctors, based on clinical experience, judge the tolerance of the patient's lung function to radiotherapy, chemotherapy, or surgery, and then determine the suitable treatment plan for the patient. Therefore, the doctor's clinical experience plays a decisive role in the final clinical decision-making. If the doctor has less clinical experience, it will affect the judgment. Therefore, it is urgent to provide richer reference information to assist doctors in making judgments. Summary of the Invention

[0003] Based on this, in view of the above technical problems, it is necessary to provide a recommended method, device, computer equipment, and storage media for lung cancer treatment plans that can recommend treatment plans for lung cancer to doctors.

[0004] In a first aspect, this application provides a recommended method for a lung cancer treatment plan. The method includes:

[0005] Obtain four-dimensional medical images of a target object, and determine a plurality of functional area images based on the four-dimensional medical images, where the four-dimensional medical images are images including the lungs of the target object, and the plurality of functional area images are images of a plurality of functional areas of the lungs;

[0006] Determine the target treatment category of each functional area based on the plurality of functional area images, the four-dimensional medical images, and a scoring model;

[0007] Determine the target treatment plan of the target object based on the target treatment category of each functional area and the four-dimensional medical images.

[0008] In one of the embodiments, the determining a plurality of functional area images based on the four-dimensional medical images includes:

[0009] Input the four-dimensional medical images into a trained delineation model to obtain images of a plurality of functional areas of the lungs. The trained delineation model is obtained by training the delineation model based on a plurality of training four-dimensional medical images of the lungs and a reference functional area image set of each training four-dimensional medical image until the training is completed.

[0010] In one embodiment, determining the target treatment category of each functional area based on the multiple functional area images, the four-dimensional medical image and the scoring model comprises:

[0011] Based on any functional area image, determining a functional parameter of any functional area;

[0012] Determining a target score for any functional area based on the four-dimensional medical image, the multiple functional area images, the functional parameters of any functional area, and the scoring model;

[0013] According to the target score of any functional area, a target treatment category of any functional area is determined from a plurality of preset treatment categories.

[0014] In one embodiment, determining the target score of any functional area based on the four-dimensional medical image, the multiple functional area images, the functional parameters of any functional area and the scoring model comprises:

[0015] Inputting the four-dimensional medical image, the multiple functional area images, and the functional parameters of any functional area into the scoring model to obtain a first scoring set for any functional area, wherein the first scoring set includes the scores of the multiple treatment categories;

[0016] The highest score in the first score set of any functional area is used as the target score of any functional area.

[0017] In one embodiment, determining the target treatment plan for the target object based on the target treatment category of each functional area and the four-dimensional medical image includes:

[0018] Delineating the four-dimensional medical image to obtain a gross tumor target volume image, a clinical tumor target volume image, and several organ-at-risk images;

[0019] A target treatment plan for the target object is determined based on the target treatment category of each functional area, the gross tumor target volume image, the clinical tumor target volume image, and the plurality of organ-at-risk images.

[0020] In one embodiment, the step of determining the target treatment plan for the target object based on the target treatment category of each functional area, the gross tumor target volume image, the clinical tumor target volume image, and the plurality of organ-at-risk images comprises:

[0021] Input the target treatment category of any functional area, the gross tumor target volume image, the clinical tumor target volume image, and the several critical organ images into a target classification model to obtain a second scoring set for any functional area, where the second scoring set includes: scores of multiple preset plans for the target treatment category;

[0022] Take the preset plan corresponding to the highest score in the second scoring set of any functional area as the treatment plan for any functional area;

[0023] Determine the target treatment plan for the target object according to the treatment plans of each functional area.

[0024] In one embodiment, after determining the target treatment plan for the target object based on the target treatment category of each functional area, the gross tumor target volume image, the clinical tumor target volume image, and the several critical organ images, it further includes:

[0025] If the treatment plan for the target object includes any target radiotherapy plan, determine any target radiotherapy plan according to the any target radiotherapy plan, where the target radiotherapy plan includes: target irradiation ray dose, target rotation angle of the radiotherapy device, and target grating position;

[0026] Based on the any target radiotherapy plan and the four-dimensional medical image, determine the probability of radiation pneumonitis for the any target radiotherapy plan;

[0027] If the probability of radiation pneumonitis for any target radiotherapy plan is greater than a preset pneumonitis threshold, adjust the any target radiotherapy plan for the target object.

[0028] In one embodiment, the adjusting the any target radiotherapy plan for the target object includes:

[0029] For any candidate radiotherapy plan among the multiple preset plans corresponding to radiotherapy, determine the candidate probability of radiation pneumonitis for the any candidate radiotherapy plan according to the any candidate radiotherapy preset plan and the four-dimensional medical image;

[0030] Among all the candidate probabilities of radiation pneumonitis, select any candidate probability of radiation pneumonitis that is less than the preset pneumonitis threshold, and take the candidate radiotherapy plan corresponding to the selected any candidate probability of radiation pneumonitis as the target radiotherapy plan for the target object.

[0031] In a second aspect, the present application further provides a recommendation device for a lung cancer treatment plan. The device includes:

[0032] A functional area image determination module, configured to obtain a four-dimensional medical image of a target object, and determine a plurality of functional area images based on the four-dimensional medical image, where the four-dimensional medical image is an image including the lungs of the target object, and the plurality of functional area images are images of multiple functional areas of the lungs;

[0033] A treatment category determination module, configured to determine a target treatment category for each functional area based on the plurality of functional area images, the four-dimensional medical image, and a scoring model;

[0034] A recommendation module, configured to determine a target treatment plan for the target object based on the target treatment category of each functional area and the four-dimensional medical image.

[0035] In a third aspect, the present application further provides a computer device. The computer device includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0036] Obtain a four-dimensional medical image of a target object, and determine a plurality of functional area images based on the four-dimensional medical image, where the four-dimensional medical image is an image including the lungs of the target object, and the plurality of functional area images are images of multiple functional areas of the lungs;

[0037] Determine a target treatment category for each functional area based on the plurality of functional area images, the four-dimensional medical image, and a scoring model;

[0038] Determine a target treatment plan for the target object based on the target treatment category of each functional area and the four-dimensional medical image.

[0039] In a fourth aspect, the present application further provides a computer-readable storage medium. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the following steps are implemented:

[0040] Obtain a four-dimensional medical image of a target object, and determine a plurality of functional area images based on the four-dimensional medical image, where the four-dimensional medical image is an image including the lungs of the target object, and the plurality of functional area images are images of multiple functional areas of the lungs;

[0041] Determine a target treatment category for each functional area based on the plurality of functional area images, the four-dimensional medical image, and a scoring model;

[0042] Determine a target treatment plan for the target object based on the target treatment category of each functional area and the four-dimensional medical image.

[0043] Fifth aspect, the present application further provides a computer program product. The computer program product includes a computer program which, when executed by a processor, implements the following steps:

[0044] Obtain a four-dimensional medical image of a target object, and determine a plurality of functional area images based on the four-dimensional medical image, wherein the four-dimensional medical image is an image including the lungs of the target object, and the plurality of functional area images are images of a plurality of functional areas of the lungs;

[0045] Determine the target treatment category of each functional area based on the plurality of functional area images, the four-dimensional medical image, and a scoring model;

[0046] Determine the target treatment plan of the target object based on the target treatment category of each functional area and the four-dimensional medical image.

[0047] The above-mentioned method, device, computer device, and storage medium for recommending a lung cancer treatment plan determine the images of each functional area of the lungs according to the four-dimensional medical image of the target object, and score each functional area through deep learning based on the images of each functional area to obtain the target treatment category of each functional area, that is, determine the target treatment category according to the tolerance of the functional area, and recommend a suitable target treatment plan for the target object according to the four-dimensional medical image and the target treatment category of each functional area, which can provide clinicians with richer reference information to help doctors decide the final treatment plan. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 It is a schematic flowchart of a method for recommending a lung cancer treatment plan in an embodiment;

[0049] Figure 2 It is a schematic flowchart of the steps of a method for recommending a lung cancer treatment plan in a specific embodiment;

[0050] Figure 3 It is a structural block diagram of a device for recommending a lung cancer treatment plan in another embodiment;

[0051] Figure 4 It is an internal structure diagram of a computer device in an embodiment. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0052] In order to make the objectives, technical solutions, and advantages of the present application clearer, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0053] In one embodiment, as Figure 1As shown, a method for recommending a lung cancer treatment plan is provided. In this embodiment, the method is exemplified by its application to a terminal. It can be understood that the method can also be applied to a server, or to a system including a terminal and a server, and is implemented through the interaction between the terminal and the server. In this embodiment, the method includes the following steps:

[0054] S101, obtain the four-dimensional medical image of the target object, and determine multiple functional area images based on the four-dimensional medical image.

[0055] Among them, the four-dimensional medical image is based on the three-dimensional medical image plus the fourth-dimensional time vector, and is also called real-time three-dimensional medical image. Through the four-dimensional medical image, the dynamic movement of the organ can be detected. The four-dimensional medical image is obtained by photographing the target object with a four-dimensional medical imaging device. The four-dimensional medical image is an image including the lungs of the target object, and the four-dimensional medical image also includes the organs at risk of the lungs; the multiple functional area images are images of multiple functional areas of the lungs; the multiple functional areas are different regions divided based on the size affected by respiratory movement, and the ventilation efficiencies of the multiple functional areas are different.

[0056] Specifically, a plurality of functional area images of the four-dimensional medical image can be determined through a delineation model; the plurality of functional area images are input into the delineation model, and a plurality of functional area images of the lungs are output through the delineation model.

[0057] S102, determine the target treatment category of each functional area based on the multiple functional area images, the four-dimensional medical image, and a scoring model.

[0058] Among them, the scoring model is used to obtain the score of each treatment category of each functional area and determine the target treatment category of each functional area. Each treatment category includes: radiotherapy, chemotherapy, and surgery, and the target treatment category is radiotherapy, or chemotherapy, or surgery.

[0059] Specifically, determine the functional parameters of each functional area, and the functional parameters are used to reflect the size of the functional area affected by respiratory movement. For each functional area, input multiple four-dimensional medical images, multiple functional area images, and the functional parameters of this functional area into the scoring model to obtain the target treatment category of this functional area.

[0060] S103, determine the target treatment plan of the target object based on the target treatment category of each functional area and the four-dimensional medical image.

[0061] Specifically, based on the four-dimensional medical image, a gross tumor target volume image, a clinical tumor target volume image, and several organ-at-risk images are delineated. Then, based on the gross tumor target volume image, the clinical tumor target volume image, the several organ-at-risk images, and a classification model, among the multiple preset plans corresponding to the target treatment category of any functional area, a treatment plan for any functional area is determined. The classification model is used to obtain the score of each preset plan and take the preset plan with the highest score as the treatment plan. Then, the target treatment plan is determined according to the treatment plan of each functional area.

[0062] In the above method for recommending a lung cancer treatment plan, the images of each functional area of the lungs are determined according to the four-dimensional medical image of the target object. Based on the images of each functional area, each functional area is scored through deep learning to obtain the target treatment category of each functional area, that is, the target treatment category is determined according to the tolerance of the functional area. And according to the four-dimensional medical image and the target treatment category of each functional area, a suitable target treatment plan is recommended for the target object, which can provide richer reference information for clinicians and help doctors decide the final treatment plan.

[0063] In one embodiment, S101 includes:

[0064] S111, input the four-dimensional medical image into the trained delineation model to obtain the images of multiple functional areas of the lungs.

[0065] Among them, the multiple functional areas include: a fast area, a normal area, and a slow area; the images of the multiple functional areas are also four-dimensional images. The images of the multiple functional areas include: a fast area image, a normal area image, and a slow area image.

[0066] The multiple functional areas are divided according to the degree of influence of respiratory motion. The fast area is greatly affected by respiratory motion, the slow area is less affected by respiratory motion, and the normal area is not affected by respiratory motion. That the fast area is greatly affected by respiratory motion means that the displacement change of any point in the fast area caused by respiratory motion is large. That the slow area is less affected by respiratory motion means that the displacement change of any point in the slow area caused by respiratory motion is small. That the normal area is not affected by respiratory motion means that almost no displacement change occurs for any point in the normal area following respiratory motion.

[0067] Among them, the trained delineation model is obtained by training the delineation model based on multiple training four-dimensional medical images of the lungs and the reference functional area image set of each training four-dimensional medical image until the training is completed. The reference functional area image set includes: a reference fast area image that is greatly affected by respiration, a reference slow area image that is less affected by respiration, and a reference normal area image that is not affected by respiration.

[0068] Specifically, the delineation model can be implemented by a decision tree, a random forest, or a deep neural network. During training, the training four-dimensional medical images are input into the delineation model, and the training functional area image set corresponding to the training four-dimensional medical images is output through the delineation model. The first loss function value is calculated according to the reference functional area image set and the training functional area image set corresponding to the training four-dimensional medical images. The model parameters of the delineation model are modified according to the first loss function value, and then one training is completed. Then, repeat the above process of determining the training functional area image set corresponding to the output training medical images until the delineation model converges, and the training is completed to obtain the trained delineation model. Among them, the training functional area image set includes: training fast area images greatly affected by breathing, training slow area images less affected by breathing, and training normal area images not affected by breathing.

[0069] In one embodiment, S102 includes:

[0070] S121, based on any one of the functional area images, determine the functional parameters of the any one of the functional areas.

[0071] Among them, the functional parameters of any one of the functional areas at least include pulmonary ventilation volume.

[0072] Specifically, the any one of the functional area images is a four-dimensional image, that is, any one of the functional area images includes multiple respiratory images with different respiratory phases. The functional parameters are determined according to the volume of any one of the functional areas in the multiple respiratory images. For the multiple respiratory images included in any one of the functional area images, determine the volume of any one of the functional areas in each respiratory image to obtain the maximum volume and the minimum volume, calculate the difference between the maximum volume and the minimum volume to obtain the pulmonary ventilation volume of any one of the functional areas. The tidal volume, expiratory reserve volume, and functional residual capacity can also be calculated according to the pulmonary ventilation volume. Calculating the tidal volume, expiratory reserve volume, and functional residual capacity according to the pulmonary ventilation volume are all existing methods and will not be elaborated here.

[0073] S122, based on the four-dimensional medical image, the multiple functional area images, the functional parameters of any one of the functional areas, and the scoring model, determine the target score of any one of the functional areas.

[0074] Specifically, input the four-dimensional medical image, the multiple functional area images, and the functional parameters of any one of the functional areas into the scoring model to obtain the first score set of any one of the functional areas, and use the highest score in the first score set as the target score of any one of the functional areas.

[0075] In one embodiment, S122 includes:

[0076] S1221, input the four-dimensional medical image, the multiple functional area images, and the functional parameters of any one of the functional areas into the scoring model to obtain the first score set of any one of the functional areas.

[0077] Specifically, input the four-dimensional medical image, the multiple functional area images, and the functional parameters of any one functional area into the scoring model. The scoring model outputs the scores for each treatment category, obtaining a first score set. That is to say, the first score set includes the scores for each treatment category, which is used to reflect the probability that any one functional area is suitable for each treatment category. The sum of the multiple scores in the first score set is 1.

[0078] S1222. Take the highest score in the first score set of any one functional area as the target score of any one functional area.

[0079] Specifically, the highest score is the score with the largest value output by the scoring model. Take the highest score as the target score of any one functional area. For example, the first score set of any one functional area includes: 0.1, 0.3, 0.6, then take 0.6 as the target score.

[0080] S123. Determine the target treatment category of any one functional area from a preset multiple treatment categories according to the target score of any one functional area.

[0081] Specifically, the first score set includes multiple first scores, and the multiple first scores correspond one by one to a preset multiple treatment categories. After determining the target score, take the treatment category corresponding to the target score as the target treatment category.

[0082] For example, the first score set of any one functional area includes: 0.1, 0.3, 0.6, where the treatment category corresponding to 0.1 is: radiotherapy, the treatment category corresponding to 0.3 is chemotherapy, and the treatment category corresponding to 0.6 is surgery; take 0.6 as the target score, and take the treatment category corresponding to 0.6: surgery, as the target treatment category.

[0083] If there are two identical first scores in the first score set and they are higher than another first score, then the treatment categories corresponding to the two identical first scores can be taken as the target treatment categories respectively to obtain a determined combined treatment category. For example: radiotherapy and chemotherapy treatment categories, or preoperative radiotherapy, etc. If there are three identical first scores in the first score set, then determine a combined treatment plan for radiotherapy, chemotherapy, and surgery. For example, preoperative radiotherapy, postoperative chemotherapy, etc.

[0084] The scoring model is obtained by training a first preset model based on a plurality of training four-dimensional medical images, a reference functional area image set of each training four-dimensional medical image, a reference functional parameter set and a reference scoring set of each reference functional area until the training is completed. The reference functional area image set includes a plurality of reference functional area images, the reference functional parameter set includes: a reference functional parameter of each reference functional area image, a reference scoring set of each reference functional area, including a reference score for each reference functional area suitable for chemotherapy, radiotherapy and surgery, and in the reference scoring set of each reference functional area, only one reference score should not be 0, and two reference scores should be 0.

[0085] Specifically, the first preset model can be implemented by a decision tree, a random forest or a deep neural network. During training, the training four-dimensional medical image, the reference functional area image set and the reference functional parameter set of the training four-dimensional medical image are input into the first preset model to obtain the first training score set of any reference functional area, and the highest score in the first training score set of any reference functional area is used as the training target score of any reference functional area. The training score set of the training four-dimensional medical image is obtained according to the training target score of each reference functional area, and the second loss function value is calculated according to the training score set and the reference score set of the training four-dimensional medical image. The parameters of the first preset model are modified according to the second loss function value, and then one training is completed, and the above process of determining the training score set is repeated until the first preset model converges, and the training is completed to obtain the scoring model.

[0086] In one embodiment, S103 includes:

[0087] S311 , outlining the four-dimensional medical image to obtain a gross tumor target volume image, a clinical tumor target volume image, and images of several organs at risk.

[0088] Among them, gross tumor target volume (GTV) images refer to images of tumor location and tumor range that are directly visible or palpable and can be confirmed by diagnostic examination methods; clinical target volume (CTV) images refer to confirmed tumors and potential invaded tissues; organs at risk of lung cancer include: bilateral lungs (either left lung or right lung), esophagus, heart (or pericardium), great blood vessels, spinal cord, trachea (or proximal bronchial tree), chest wall, ribs, skin, stomach and liver.

[0089] Specifically, outlining the four-dimensional medical image to obtain a gross tumor target volume image, a clinical tumor target volume image and several images of organs at risk can be achieved through the existing PV-iRT intelligent radiotherapy auxiliary system.

[0090] S312. Determine the target treatment plan for the target object based on the target treatment category of each functional area, the gross tumor target volume image, the clinical tumor target volume image, and the plurality of organ-at-risk images.

[0091] Specifically, input the target treatment category of each functional area, the gross tumor target volume image, the clinical tumor target volume image, and the plurality of organ-at-risk images into a target classification model to obtain the plan scores for each functional area, determine the treatment plan corresponding to the plan score of each functional area in a preset plan set, and determine the target treatment plan according to the treatment plans of each functional area.

[0092] In one embodiment, S312 includes:

[0093] S3121. Input the target treatment category of any one functional area, the gross tumor target volume image, the clinical tumor target volume image, and the plurality of organ-at-risk images into the target classification model to obtain the second score set for the any one functional area.

[0094] Specifically, the second score set includes: the scores of multiple preset plans for the target treatment category. The second score set includes multiple second scores, and the multiple second scores correspond one-to-one to the multiple preset plans corresponding to the target category. The sum of all the second scores in the second score set is equal to 1. The target classification model is used to determine the second scores of any one functional area in the multiple preset plans corresponding to the target treatment category.

[0095] The multiple preset plans corresponding to radiotherapy include: three-dimensional conformal radiotherapy (3DCRT), intensity-modulated conformal radiotherapy (IMRT), volumetric modulated arc therapy (VMRT), and stereotactic body radiotherapy (SBRT); the multiple preset plans corresponding to chemotherapy include: multiple chemotherapy drugs; the multiple preset plans corresponding to surgery include: local resection, segmentectomy, lobectomy, bronchial sleeve lobectomy, bronchial-pulmonary artery sleeve lobectomy, tracheal carina resection and reconstruction, and pneumonectomy.

[0096] For example, the target treatment category for the slow area is radiotherapy. For radiotherapy, the target treatment category, gross tumor target volume image, clinical tumor target volume image, and several organ-at-risk images are input into the target classification model to obtain the second scoring set for the slow area. Suppose the second scoring set includes: 0.15, 0.1, 0.05, 0.7. Suppose the preset plan corresponding to 0.15 is: three-dimensional conformal radiotherapy (3DCRT), the preset plan corresponding to 0.1 is: intensity-modulated radiotherapy (IMRT), the preset plan corresponding to 0.05 is: volumetric modulated arc therapy (VMRT), and the preset plan corresponding to 0.7 is: stereotactic body radiotherapy (SBRT); it means that the probability that the slow area is suitable for 3DCRT is 0.15, the probability that it is suitable for IMRT is 0.1, the probability that it is suitable for VMRT is 0.05, and the probability that it is suitable for SBRT is 0.7.

[0097] The target classification model is obtained by training a second preset model based on multiple training data sets until the training is completed. Each training data set includes: training treatment category, training gross tumor target volume image, training clinical tumor target volume image, several training organ-at-risk images, and a reference treatment plan.

[0098] Specifically, the second preset model can be implemented by a decision tree, a random forest, or a deep neural network. During training, the training treatment category, training gross tumor target volume image, training clinical tumor target volume image, and several training organ-at-risk images in a training data set are input into the second preset model to obtain the training treatment plan corresponding to this training data set. The parameters of the second preset model are adjusted according to the reference treatment plan and the training treatment plan, and then one training is completed. Repeat the above process of determining the training treatment plan until the second preset model converges, and then the training is completed to obtain the target classification model.

[0099] S313. Use the preset plan corresponding to the highest score in the second scoring set of any functional area as the treatment plan for any functional area.

[0100] S314. Determine the target treatment plan of the target object according to the treatment plan of each functional area.

[0101] Specifically, the target treatment plan of the target object includes: the treatment plan for the slow area, the treatment plan for the fast area, and the treatment plan for the normal area.

[0102] For example, the target treatment category of the fast zone is radiotherapy. Assume that the second score set corresponding to the fast zone includes: 0.15, 0.1, 0.05, 0.7, among which the preset scheme corresponding to 0.7 is: suitable for stereotactic body radiotherapy SBRT; the target treatment category of the normal zone is chemotherapy. Assume that the second score set corresponding to the normal zone includes: 0.2, 0.05, 0.75, among which the preset scheme corresponding to 0.75 is: chemotherapy drug 1; Assume that the target treatment category of the slow zone is radiotherapy. Assume that the second score set corresponding to the slow zone includes: 0.05, 0.05, 0.15, 0.75, among which the preset scheme corresponding to 0.75 is: suitable for stereotactic body radiotherapy SBRT, then the target treatment schemes corresponding to the targets include: fast zone: stereotactic body radiotherapy SBRT, slow zone: stereotactic body radiotherapy SBRT, normal zone: chemotherapy drug 1.

[0103] In order to facilitate understanding of the above-mentioned recommended method for lung cancer treatment, in a specific embodiment, see Figure 2 , the recommended methods of lung cancer treatment include:

[0104] According to the four-dimensional medical image of the target object, a plurality of functional area images of the lungs are obtained by using the trained delineation model, and according to the plurality of functional area images and the four-dimensional medical image, a target score of each is obtained by using the scoring model, and then the target treatment category of each functional area is determined from the preset multiple treatment categories. According to the four-dimensional medical image, a gross tumor target volume image, a clinical tumor target volume image and several images of organs at risk are delineated, and according to the gross tumor target volume image, the clinical tumor target volume image and the images of several organs at risk, a treatment plan for any functional area is determined from the multiple preset plans corresponding to the target treatment category of any functional area by using the target classification model, and then the target radiotherapy plan is obtained.

[0105] In one embodiment, if the treatment category of any functional area is radiotherapy, the probability of radiation pneumonitis caused by the target radiotherapy regimen used in any functional area can be further predicted. If the probability of radiation pneumonitis caused by the target radiotherapy regimen is high, the target radiotherapy regimen needs to be adjusted.

[0106] After S103, the method further includes:

[0107] S104: If the treatment plan of the target object includes any target radiotherapy plan, determine any target radiotherapy plan according to the any target radiotherapy plan.

[0108] Specifically, any of the target radiotherapy plans is a treatment plan for the radiotherapy category of any functional area of the lungs. The treatment plan for the target object includes any target radiotherapy plan, which means that there is a treatment category of radiotherapy for any functional area, that is, there is a treatment plan for any functional area determined from multiple preset plans corresponding to radiotherapy. If the multiple functional areas include: a slow area, a fast area, and a normal area, the treatment plan for the target object may include: one target radiotherapy plan (any functional area adopts the target radiotherapy plan), or two target radiotherapy plans (two functional areas adopt the target radiotherapy plan), or three target radiotherapy plans (each functional area adopts the target radiotherapy plan).

[0109] For any target radiotherapy plan of the target object, determining any target radiotherapy plan according to any target radiotherapy plan includes: obtaining the dose of any target radiotherapy plan, and according to the dose and any target radiotherapy plan, through an existing automatic optimization algorithm, any target radiotherapy plan can be determined. Among them, the dose of any target radiotherapy plan can be the reference dose of any target plan, or the dose set by a doctor according to any target radiotherapy plan. The target radiotherapy plan includes: the target irradiation ray dose, the target rotation angle of the radiotherapy device, and the target grating position.

[0110] If the treatment plan for the target object includes multiple target radiotherapy plans, the target radiotherapy plans for each target radiotherapy plan can be obtained through an existing automatic optimization algorithm.

[0111] S105, based on any target radiotherapy plan and the four-dimensional medical image, determine the probability of radiation pneumonitis for any target radiotherapy plan.

[0112] Specifically, determine the gross tumor target volume image, the clinical tumor target volume image, and the images of several organs at risk according to the four-dimensional medical image; according to any target radiotherapy plan, the gross tumor target volume image, the clinical tumor target volume image, and the images of several organs at risk, use the Monte Carlo algorithm to determine the dose distribution image; the dose distribution image is used to reflect the radiation dose of the gross tumor target area, the clinical tumor target area, and the organ-at-risk area under any target radiotherapy plan.

[0113] Input the dose distribution image and the four-dimensional medical image into the pneumonia prediction model to obtain the probability of radiation pneumonitis for any target radiotherapy plan. Among them, the pneumonia prediction model is obtained by training a third preset model based on multiple training medical images, the training dose distribution image of each medical image, and the radiation pneumonitis label until the training is completed.

[0114] S106, if the probability of radiation pneumonitis for any target radiotherapy plan is greater than the preset pneumonia threshold, adjust any target radiotherapy plan for the target object.

[0115] Specifically, if the probability of radiation pneumonitis of any target radiotherapy plan is greater than the preset pneumonitis threshold, it indicates that the any target radiotherapy plan may cause radiation pneumonitis and the any target radiotherapy plan needs to be adjusted. The adjusted target radiotherapy plan can be determined from multiple preset plans corresponding to radiotherapy, so that the probability of radiation pneumonitis corresponding to the adjusted target radiotherapy plan is less than the preset pneumonitis threshold.

[0116] In one embodiment, S106 includes:

[0117] S161. For any candidate radiotherapy plan among multiple preset plans corresponding to radiotherapy, according to any candidate radiotherapy preset plan and the four-dimensional medical image, determine the candidate probability of radiation pneumonitis of any candidate radiotherapy plan.

[0118] Specifically, any candidate radiotherapy plan is any one of the multiple preset plans corresponding thereto, except for the target radiotherapy plan. Determining any candidate radiotherapy plan according to any candidate radiotherapy plan includes: obtaining the dose of any candidate radiotherapy plan, and according to the dose of any candidate radiotherapy plan and the dose plan of any candidate radiotherapy plan, through an existing automatic optimization algorithm, any candidate radiotherapy plan can be determined. Wherein, the dose of any candidate radiotherapy plan may be the reference dose of any candidate plan.

[0119] Determine the gross tumor target volume image, the clinical tumor target volume image, and the images of several organs at risk according to the four-dimensional medical image; according to any candidate radiotherapy plan, the gross tumor target volume image, the clinical tumor target volume image, and the images of several organs at risk, use the Monte Carlo algorithm to determine the candidate dose distribution image; the candidate dose distribution image is used to reflect the radiation doses of the gross tumor target area, the clinical tumor target area, and the organ-at-risk area under any candidate radiotherapy plan.

[0120] Input the candidate dose distribution image and the four-dimensional medical image into the pneumonia prediction model to obtain the candidate probability of radiation pneumonitis of any candidate radiotherapy plan.

[0121] S162. Among all candidate probabilities of radiation pneumonitis, select any candidate probability of radiation pneumonitis that is less than the preset pneumonitis threshold, and use the candidate radiotherapy plan corresponding to the selected any candidate probability of radiation pneumonitis as the target radiotherapy plan of the target object.

[0122] Specifically, determine the candidate probability of radiation pneumonitis of each candidate radiotherapy plan among the multiple preset plans corresponding to radiotherapy, and use the candidate radiotherapy plan corresponding to the candidate probability of radiation pneumonitis that is less than the preset pneumonitis threshold as the target radiotherapy plan of the target.

[0123] If the candidate radiation pneumonitis probabilities of at least two candidate radiotherapy plans are both less than a preset pneumonitis threshold, determine the second score of the at least two candidate radiotherapy plans, and use the candidate radiotherapy plan with the highest second score as the target radiotherapy plan.

[0124] If there is no candidate radiotherapy plan with a candidate radiation pneumonitis probability less than the preset pneumonitis threshold, adjust the target treatment category; since the first score corresponding to radiotherapy is the maximum value in the first score set, determine other first scores in the first score set that are only less than the first score corresponding to radiotherapy, and use the treatment categories corresponding to the determined other first scores as the adjusted target treatment category, and continue to execute S103.

[0125] For example, if the preset multiple treatment categories include radiotherapy, chemotherapy, and surgery, and in the first score set, the first score of surgery is greater than the first score of chemotherapy, then adjust the target treatment category to surgery and continue to execute S103; if the first score of chemotherapy is greater than the first score of surgery, then adjust the target treatment category to chemotherapy and continue to execute S103.

[0126] In this embodiment, according to the four-dimensional medical image of the target object, using the trained delineation model, obtain the images of each functional area of the lungs, and determine the functional parameters of each functional area. According to the four-dimensional medical image, multiple functional area images, and the functional parameters of any functional area, use the scoring model to obtain the target score of any functional area, and then determine the target treatment category of any functional area among the preset multiple treatment categories. According to the four-dimensional medical image, delineate the gross tumor target volume image, clinical tumor target volume image, and several organ-at-risk images. According to the gross tumor target volume image, clinical tumor target volume image, and several organ-at-risk images, use the target classification model to determine the treatment plan for any functional area among the multiple preset plans corresponding to the target treatment category of any functional area, and then obtain the target radiotherapy plan. If the treatment plan for any functional area is the target radiotherapy plan, predict the radiation pneumonitis probability of the target radiotherapy plan. If the radiation pneumonitis probability is greater than the pneumonitis threshold, determine the preset plan with a radiation pneumonitis probability less than the pneumonitis threshold among the multiple preset plans corresponding to radiotherapy as the target radiotherapy plan. Through the recommended method for the lung cancer treatment plan, it provides richer reference information for clinicians, recommends the target treatment plan for doctors, and helps doctors decide the final treatment plan.

[0127] It should be understood that although the steps in the flowcharts involved in the above-described embodiments are sequentially shown according to the indications of the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless there is a clear indication in this article, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-described embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be executed alternately or in turn with at least a part of other steps or steps or stages in other steps.

[0128] Based on the same inventive concept, an embodiment of the present application further provides a recommendation device for a lung cancer treatment plan for implementing the recommendation method of the lung cancer treatment plan described above. The solution provided by this device to solve the problem is similar to the solution described in the above method. Therefore, the specific limitations in one or more embodiments of the recommendation device for the lung cancer treatment plan provided below can refer to the limitations on the recommendation method of the lung cancer treatment plan in the above text, and will not be repeated here.

[0129] In one embodiment, as Figure 3 shown, a recommendation device for a lung cancer treatment plan is provided, including: a functional area image determination module, a treatment category determination module, and a recommendation module, where:

[0130] The functional area image determination module is configured to obtain four-dimensional medical images of a target object, and determine a plurality of functional area images based on the four-dimensional medical images, where the four-dimensional medical images are images including the lungs of the target object, and the plurality of functional area images are images of a plurality of functional areas of the lungs;

[0131] The treatment category determination module is configured to determine the target treatment category of each functional area based on the plurality of functional area images, the four-dimensional medical images, and a scoring model;

[0132] The recommendation module is configured to determine the target treatment plan of the target object based on the target treatment category of each functional area and the four-dimensional medical images.

[0133] In one embodiment, the functional area image determination module includes: a first delineation unit, where:

[0134] The first delineation unit is configured to input the four-dimensional medical images into a trained delineation model to obtain images of a plurality of functional areas of the lungs. The trained delineation model is obtained by training the delineation model based on a plurality of training four-dimensional medical images of the lungs and a reference functional area image set of each training four-dimensional medical image until the training is completed.

[0135] In one embodiment, the treatment category determination module comprises: a parameter determination unit, a first scoring unit and a category determination unit, wherein:

[0136] A parameter determination unit, used for determining a function parameter of any function area based on any function area image;

[0137] A first scoring unit, configured to determine a target score for any functional area based on the four-dimensional medical image, the multiple functional area images, the functional parameters of any functional area, and the scoring model;

[0138] A category determination unit is used to determine a target treatment category of any functional area from a plurality of preset treatment categories according to the target score of any functional area.

[0139] In one embodiment, the first scoring unit includes: a first subunit and a second subunit, wherein:

[0140] A first subunit is configured to input the four-dimensional medical image, the multiple functional area images, and the functional parameters of any functional area into the scoring model to obtain a first scoring set of any functional area, wherein the first scoring set includes the scores of the multiple treatment categories;

[0141] The second character unit is used to set the highest score in the first score set of any functional area as the target score of any functional area.

[0142] In one embodiment, the recommendation module includes: a second delineation unit and a scheme determination unit, wherein:

[0143] A second delineation unit is used to delineate the four-dimensional medical image to obtain a gross tumor target volume image, a clinical tumor target volume image and several images of organs at risk;

[0144] A plan determination unit is used to determine a target treatment plan for the target object based on a target treatment category of each functional area, the gross tumor target volume image, the clinical tumor target volume image, and the plurality of organ-at-risk images.

[0145] In one embodiment, the solution determination unit includes: a third subunit, a fourth subunit and a fifth subunit, wherein:

[0146] A third subunit is used to input the target treatment category of any functional area, the gross tumor target volume image, the clinical tumor target volume image and the plurality of organ-at-risk images into a target classification model to obtain a second score set for any functional area, wherein the second score set includes: scores of multiple preset schemes for the target treatment category;

[0147] A fourth sub-unit, configured to use the preset plan corresponding to the highest score in the second scores of any one of the functional areas as the treatment plan for any one of the functional areas;

[0148] A fifth sub-unit, configured to determine the target treatment plan of the target object according to the treatment plans of each functional area.

[0149] In one embodiment, the recommendation device for the lung cancer treatment plan further includes: a first adjustment module, a second adjustment module, and a third adjustment module, where:

[0150] The first adjustment module is configured to, if the treatment plan of the target object includes any one of the target radiotherapy plans, determine any one of the target radiotherapy plans according to any one of the target radiotherapy plans, where the target radiotherapy plan includes: the target irradiation ray dose, the target rotation angle of the radiotherapy device, and the target grating position;

[0151] The second adjustment module is configured to determine the probability of radiation pneumonitis of any one of the target radiotherapy plans based on any one of the target radiotherapy plans and the four-dimensional medical image;

[0152] The third adjustment module is configured to, if the probability of radiation pneumonitis of any one of the target radiotherapy plans is greater than the preset pneumonitis threshold, adjust any one of the target radiotherapy plans of the target object.

[0153] In one embodiment, the third adjustment module includes: a first adjustment unit and a second adjustment unit, where:

[0154] The first adjustment unit is configured to, for any one of the candidate radiotherapy plans among the multiple preset plans corresponding to radiotherapy, determine the candidate probability of radiation pneumonitis of any one of the candidate radiotherapy preset plans according to any one of the candidate radiotherapy preset plans and the four-dimensional medical image;

[0155] The second adjustment unit is configured to select, from all the candidate probabilities of radiation pneumonitis, any one of the candidate probabilities of radiation pneumonitis that is less than the preset pneumonitis threshold, and use the candidate radiotherapy plan corresponding to any one of the selected candidate probabilities of radiation pneumonitis as the target radiotherapy plan of the target object.

[0156] Each module in the above-mentioned recommendation device for the lung cancer treatment plan can be implemented in whole or in part by software, hardware, and their combination. The above-mentioned modules can be embedded in the processor of the computer device in the form of hardware or be independent of the processor, or be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above-mentioned modules.

[0157] In one embodiment, a computer device is provided. The computer device can be a terminal, and its internal structure diagram can be as Figure 4As shown in the figure. The computer device includes a processor, a memory, a communication interface, a display screen, and an input device connected through 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 and an internal memory. The non-volatile storage medium stores an operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The communication interface of the computer device is used to communicate with external terminals in a wired or wireless manner, and the wireless manner can be achieved through WIFI, mobile cellular networks, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it realizes a method for recommending a lung cancer treatment plan. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covered on the display screen, or buttons, trackballs, or touchpads provided on the computer device housing, or an external keyboard, touchpad, or mouse, etc.

[0158] Those skilled in the art can understand that Figure 4 the structure shown in the figure is only a block diagram of some structures related to the solution of this application, and does not constitute a limitation on the computer device to which the solution of this application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine some components, or have different component arrangements.

[0159] In one embodiment, a computer device is provided, including a memory and a processor. A computer program is stored in the memory. When the processor executes the computer program, the following steps are implemented:

[0160] Obtain the four-dimensional medical image of the target object, and determine a plurality of functional area images based on the four-dimensional medical image. Among them, the four-dimensional medical image is an image including the lungs of the target object, and the plurality of functional area images are images of multiple functional areas of the lungs;

[0161] Determine the target treatment category of each functional area based on the plurality of functional area images, the four-dimensional medical image, and the scoring model;

[0162] Determine the target treatment plan of the target object based on the target treatment category of each functional area and the four-dimensional medical image.

[0163] In one of the embodiments, when the processor executes the computer program, the following steps are also implemented: The determining a plurality of functional area images based on the four-dimensional medical image includes:

[0164] Input the four-dimensional medical image into the trained delineation model to obtain images of multiple functional regions of the lungs. The trained delineation model is obtained by training the delineation model based on multiple training four-dimensional medical images of the lungs and a reference functional region image set for each training four-dimensional medical image until the training is completed.

[0165] In one embodiment, when the processor executes the computer program, the following steps are further implemented: Determining the target treatment category for each functional region based on the multiple functional region images, the four-dimensional medical image, and the scoring model includes:

[0166] Based on any one of the functional region images, determine the functional parameters of the any one functional region;

[0167] Based on the four-dimensional medical image, the multiple functional region images, the functional parameters of the any one functional region, and the scoring model, determine the target score of the any one functional region;

[0168] According to the target score of the any one functional region, determine the target treatment category of the any one functional region among a preset plurality of treatment categories.

[0169] In one embodiment, when the processor executes the computer program, the following steps are further implemented: Determining the target score of the any one functional region based on the four-dimensional medical image, the multiple functional region images, the functional parameters of the any one functional region, and the scoring model includes:

[0170] Input the four-dimensional medical image, the multiple functional region images, and the functional parameters of the any one functional region into the scoring model to obtain a first score set of the any one functional region, where the first score set includes scores for the plurality of treatment categories;

[0171] Take the highest score in the first score set of the any one functional region as the target score of the any one functional region.

[0172] In one embodiment, when the processor executes the computer program, the following steps are further implemented: Determining the target treatment plan for the target object based on the target treatment category of each functional region and the four-dimensional medical image includes:

[0173] Delineate the four-dimensional medical image to obtain a gross tumor target volume image, a clinical tumor target volume image, and several images of organs at risk;

[0174] Based on the target treatment category of each functional region, the gross tumor target volume image, the clinical tumor target volume image, and the several images of organs at risk, determine the target treatment plan for the target object.

[0175] In one embodiment, when the processor executes the computer program, the following steps are further implemented: determining the target treatment plan for the target object based on the target treatment category of each functional area, the gross tumor target volume image, the clinical tumor target volume image, and the plurality of organ-at-risk images, including:

[0176] Inputting the target treatment category of any one functional area, the gross tumor target volume image, the clinical tumor target volume image, and the plurality of organ-at-risk images into a target classification model to obtain a second score set for any one functional area, where the second score set includes: scores of multiple preset plans for the target treatment category;

[0177] Taking the preset plan corresponding to the highest score in the second score set of any one functional area as the treatment plan for any one functional area;

[0178] Determining the target treatment plan for the target object according to the treatment plans of each functional area.

[0179] In one embodiment, when the processor executes the computer program, the following steps are further implemented: after determining the target treatment plan for the target object based on the target treatment category of each functional area, the gross tumor target volume image, the clinical tumor target volume image, and the plurality of organ-at-risk images, it further includes:

[0180] If the treatment plan for the target object includes any target radiotherapy plan, determining any target radiotherapy plan according to the any target radiotherapy plan, where the target radiotherapy plan includes: the target irradiation ray dose, the target rotation angle of the radiotherapy device, and the target grating position;

[0181] Based on the any target radiotherapy plan and the four-dimensional medical image, determining the probability of radiation pneumonitis for the any target radiotherapy plan;

[0182] If the probability of radiation pneumonitis for any target radiotherapy plan is greater than a preset pneumonitis threshold, adjusting the any target radiotherapy plan for the target object.

[0183] In one embodiment, when the processor executes the computer program, the following steps are further implemented: adjusting the any target radiotherapy plan for the target object includes:

[0184] For any candidate radiotherapy plan among the multiple preset plans corresponding to radiotherapy, determining the candidate probability of radiation pneumonitis for any candidate radiotherapy plan according to the any candidate radiotherapy preset plan and the four-dimensional medical image;

[0185] Among all the candidate radiation pneumonitis probabilities, select any candidate radiation pneumonitis probability that is less than the preset pneumonitis threshold, and use the candidate radiotherapy plan corresponding to the selected candidate radiation pneumonitis probability as the target radiotherapy plan for the target object.

[0186] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the following steps are implemented:

[0187] Obtain the four-dimensional medical image of the target object, and determine a plurality of functional area images based on the four-dimensional medical image, where the four-dimensional medical image is an image including the lungs of the target object, and the plurality of functional area images are images of multiple functional areas of the lungs;

[0188] Determine the target treatment category of each functional area based on the plurality of functional area images, the four-dimensional medical image, and the scoring model;

[0189] Determine the target treatment plan of the target object based on the target treatment category of each functional area and the four-dimensional medical image.

[0190] In one of the embodiments, when the computer program is executed by the processor, the following steps are further implemented: The determining a plurality of functional area images based on the four-dimensional medical image includes:

[0191] Input the four-dimensional medical image into the trained delineation model to obtain images of multiple functional areas of the lungs. The trained delineation model is obtained by training the delineation model based on multiple training four-dimensional medical images of the lungs and the reference functional area image set of each training four-dimensional medical image until the training is completed.

[0192] In one of the embodiments, when the computer program is executed by the processor, the following steps are further implemented: The determining the target treatment category of each functional area based on the plurality of functional area images, the four-dimensional medical image, and the scoring model includes:

[0193] Based on any one of the functional area images, determine the functional parameters of the any one functional area;

[0194] Based on the four-dimensional medical image, the plurality of functional area images, the functional parameters of the any one functional area, and the scoring model, determine the target score of the any one functional area;

[0195] According to the target score of the any one functional area, determine the target treatment category of the any one functional area among a preset plurality of treatment categories.

[0196] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented: Based on the four-dimensional medical image, the multiple functional area images, the functional parameters of any one of the functional areas, and the scoring model, determining the target score of any one of the functional areas, including:

[0197] Inputting the four-dimensional medical image, the multiple functional area images, and the functional parameters of any one of the functional areas into the scoring model to obtain a first score set of any one of the functional areas, where the first score set includes scores of multiple treatment categories;

[0198] Taking the highest score in the first score set of any one of the functional areas as the target score of any one of the functional areas.

[0199] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented: Based on the target treatment category of each functional area and the four-dimensional medical image, determining the target treatment plan for the target object, including:

[0200] Outlining the four-dimensional medical image to obtain a gross tumor target volume image, a clinical tumor target volume image, and several organ-at-risk images;

[0201] Based on the target treatment category of each functional area, the gross tumor target volume image, the clinical tumor target volume image, and the several organ-at-risk images, determining the target treatment plan for the target object.

[0202] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented: Based on the target treatment category of each functional area, the gross tumor target volume image, the clinical tumor target volume image, and the several organ-at-risk images, determining the target treatment plan for the target object, including:

[0203] Inputting the target treatment category of any one of the functional areas, the gross tumor target volume image, the clinical tumor target volume image, and the several organ-at-risk images into a target classification model to obtain a second score set of any one of the functional areas, where the second score set includes: scores of multiple preset plans of the target treatment category;

[0204] Taking the preset plan corresponding to the highest score in the second score set of any one of the functional areas as the treatment plan of any one of the functional areas;

[0205] Determining the target treatment plan for the target object according to the treatment plans of each functional area.

[0206] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented: After determining the target treatment plan of the target object based on the target treatment category of each functional area, the gross tumor target volume image, the clinical tumor target volume image, and the plurality of organs at risk images, the following steps are further included:

[0207] If the treatment plan of the target object includes any target radiotherapy plan, then determine any target radiotherapy plan according to the any target radiotherapy plan, wherein the target radiotherapy plan includes: the target irradiation ray dose, the target rotation angle of the radiotherapy device, and the target grating position;

[0208] Based on the any target radiotherapy plan and the four-dimensional medical image, determine the probability of radiation pneumonitis of the any target radiotherapy plan;

[0209] If the probability of radiation pneumonitis of any target radiotherapy plan is greater than the preset pneumonitis threshold, then adjust the any target radiotherapy plan of the target object.

[0210] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented: The adjusting the any target radiotherapy plan of the target object includes:

[0211] For any candidate radiotherapy plan among the multiple preset plans corresponding to radiotherapy, determine the candidate radiation pneumonitis probability of the any candidate radiotherapy plan according to the any candidate radiotherapy preset plan and the four-dimensional medical image;

[0212] Among all the candidate radiation pneumonitis probabilities, select any candidate radiation pneumonitis probability that is less than the preset pneumonitis threshold, and use the candidate radiotherapy plan corresponding to the selected any candidate radiation pneumonitis probability as the target radiotherapy plan of the target object.

[0213] In one embodiment, a computer program product is provided, including a computer program, which when executed by a processor, implements the following steps:

[0214] Obtain the four-dimensional medical image of the target object, and determine a plurality of functional area images based on the four-dimensional medical image, wherein the four-dimensional medical image is an image including the lungs of the target object, and the plurality of functional area images are images of a plurality of functional areas of the lungs;

[0215] Based on the plurality of functional area images, the four-dimensional medical image, and a scoring model, determine the target treatment category of each functional area;

[0216] Based on the target treatment category of each functional area and the four-dimensional medical image, determine the target treatment plan of the target object.

[0217] It should be noted that the user information (including but not limited to user device 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 fully authorized by all parties.

[0218] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, database, or other medium used in the embodiments provided in this application can include at least one of non-volatile and volatile memories. Non-volatile memories can include read-only memory (ROM), magnetic tapes, floppy disks, flash memories, optical memories, high-density embedded non-volatile memories, resistive random access memories (ReRAM), magnetoresistive random access memories (MRAM), ferroelectric random access memories (FRAM), phase change memories (PCM), graphene memories, etc. Volatile memories can include random access memory (RAM) or external cache memories, etc. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc. The databases involved in the embodiments provided in this application can include at least one of relational databases and non-relational databases. Non-relational databases can include distributed databases based on blockchain, etc., without limitation. The processors involved in the embodiments provided in this application can be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, data processing logics based on quantum computing, etc., without limitation.

[0219] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered to be within the scope described in this specification.

[0220] The above-described embodiments merely represent several implementation manners of the present application. The description thereof is relatively specific and detailed, but it should not be construed as a limitation on the patent scope of the present application. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can still be made, and these all fall within the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the appended claims.

Claims

1. A method for recommending a lung cancer treatment plan, characterized in that, The method includes: Obtaining a four-dimensional medical image of a target object, and determining a plurality of functional area images based on the four-dimensional medical image, wherein the four-dimensional medical image is an image including the lungs of the target object, and the plurality of functional area images are images of a plurality of functional areas of the lungs; Based on any one of the functional areas, inputting the plurality of functional area images, the four-dimensional medical image, and the functional parameters of the any one of the functional areas into a scoring model to determine the target treatment category of the any one of the functional areas; the functional parameters of the any one of the functional areas are used to reflect the magnitude of the influence of respiratory motion on the any one of the functional areas; Based on the target treatment category of each functional area and the four-dimensional medical image, determining the target treatment plan for the target object.

2. The method according to claim 1, wherein The determining a plurality of functional area images based on the four-dimensional medical image includes: Inputting the four-dimensional medical image into a trained delineation model to obtain images of a plurality of functional areas of the lungs, and the trained delineation model is obtained by training the delineation model based on a plurality of training four-dimensional medical images of the lungs and a reference functional area image set of each training four-dimensional medical image until the training is completed.

3. The method according to claim 1, characterized in that The inputting the plurality of functional area images, the four-dimensional medical image, and the functional parameters of the any one of the functional areas into a scoring model to determine the target treatment category of the any one of the functional areas includes: Inputting the plurality of functional area images, the four-dimensional medical image, and the functional parameters of the any one of the functional areas into a scoring model to determine the target score of the any one of the functional areas; According to the target score of the any one of the functional areas, determining the target treatment category of the any one of the functional areas among a plurality of preset treatment categories.

4. The method according to claim 3, wherein The inputting the plurality of functional area images, the four-dimensional medical image, and the functional parameters of the any one of the functional areas into a scoring model to determine the target score of the any one of the functional areas includes: Inputting the four-dimensional medical image, the plurality of functional area images, and the functional parameters of the any one of the functional areas into the scoring model to obtain a first score set of the any one of the functional areas, wherein the first score set includes scores of the plurality of treatment categories; Taking the highest score in the first score set of the any one of the functional areas as the target score of the any one of the functional areas.

5. The method according to claim 1, characterized in that The determining the target treatment plan for the target object based on the target treatment category of each functional area and the four-dimensional medical image includes: Delineating the four-dimensional medical image to obtain a gross tumor target volume image, a clinical tumor target volume image, and several images of organs at risk; Based on the target treatment category of each functional area, the gross tumor target volume image, the clinical tumor target volume image, and the several images of organs at risk, determining the target treatment plan for the target object.

6. The method according to claim 5, wherein The determining the target treatment plan for the target object based on the target treatment category of each functional area, the gross tumor target volume image, the clinical tumor target volume image, and the several images of organs at risk includes: Input the target treatment category of any functional area, the gross tumor target volume image, the clinical tumor target volume image, and the images of several organs at risk into the target classification model to obtain a second scoring set for any functional area, where the second scoring set includes: scores of multiple preset plans for the target treatment category; Use the preset plan corresponding to the highest score in the second scoring set of any functional area as the treatment plan for any functional area; Determine the target treatment plan for the target object according to the treatment plans of each functional area.

7. The method according to claim 5, wherein After determining the target treatment plan for the target object based on the target treatment category of each functional area, the gross tumor target volume image, the clinical tumor target volume image, and the images of several organs at risk, it further includes: If the treatment plan of the target object includes any target radiotherapy plan, determine any target radiotherapy plan according to the any target radiotherapy plan, where the target radiotherapy plan includes: the target irradiation ray dose, the target rotation angle of the radiotherapy device, and the target grating position; Based on the any target radiotherapy plan and the four-dimensional medical image, determine the probability of radiation pneumonitis of the any target radiotherapy plan; If the probability of radiation pneumonitis of any target radiotherapy plan is greater than the preset pneumonitis threshold, adjust the any target radiotherapy plan of the target object.

8. The method according to claim 7, wherein The adjustment of the any target radiotherapy plan of the target object includes: For any candidate radiotherapy plan among the multiple preset plans corresponding to radiotherapy, determine the candidate probability of radiation pneumonitis of any candidate radiotherapy plan according to the any candidate radiotherapy preset plan and the four-dimensional medical image; Among all candidate probabilities of radiation pneumonitis, select any candidate probability of radiation pneumonitis that is less than the preset pneumonitis threshold, and use the candidate radiotherapy plan corresponding to the selected any candidate probability of radiation pneumonitis as the target radiotherapy plan of the target object.

9. A device for recommending a lung cancer treatment plan, characterized in that, The device includes: A functional area image determination module, configured to obtain a four-dimensional medical image of a target object and determine multiple functional area images based on the four-dimensional medical image, where the four-dimensional medical image is an image including the lungs of the target object, and the multiple functional area images are images of multiple functional areas of the lungs; A treatment category determination module, configured to input the multiple functional area images, the four-dimensional medical image, and the functional parameters of any functional area into a scoring model based on any functional area to determine the target treatment category of any functional area; the functional parameters of any functional area are used to reflect the degree of influence of any functional area by respiratory movement; A recommendation module, configured to determine the target treatment plan for the target object based on the target treatment category of each functional area and the four-dimensional medical image.

10. A computer device, comprising a memory and a processor, the memory storing a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 8.

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

Images