CN113921133A

CN113921133A - Recommendation method and device for lung cancer treatment scheme, computer equipment and storage medium

Info

Publication number: CN113921133A
Application number: CN202111400002.6A
Authority: CN
Inventors: 贾乐成
Original assignee: Shenzhen United Imaging Research Institute of Innovative Medical Equipment
Current assignee: Shenzhen United Imaging Research Institute of Innovative Medical Equipment
Priority date: 2021-11-19
Filing date: 2021-11-19
Publication date: 2022-01-11

Abstract

The application relates to a method, an apparatus, a computer device and a storage medium for recommending a lung cancer treatment plan. The method comprises the following steps: acquiring a four-dimensional medical image of a target object, and determining a plurality of functional region images based on the four-dimensional medical image, wherein the four-dimensional medical image is an image of a lung including the target object, and the plurality of functional region images are images of a plurality of functional regions of the lung; determining a target treatment category for each functional zone based on the plurality of functional zone images, the four-dimensional medical image, and a scoring model; determining a target treatment plan for the target object based on the target treatment category for each functional zone and the four-dimensional medical image. By adopting the scheme, a proper target treatment scheme can be recommended for the target object, richer reference information can be provided for a clinician, and the clinician can be helped to determine the final treatment scheme.

Description

Recommendation method and device for lung cancer treatment scheme, computer equipment and storage medium

Technical Field

The present application relates to the field of medical image processing technology, and in particular, to a lung cancer treatment plan method, apparatus, computer device, and storage medium.

Background

Before lung cancer treatment, lung of a lung cancer patient needs to be detected, lung function of the patient needs to be known, and whether the patient is suitable for treatment such as surgery, radiotherapy and the like is determined. The existing lung detection mainly depends on a multi-conductive physiological recorder or a lung function tester to determine functional parameters of the lung, then doctors judge the tolerance degree of the lung function of patients to radiotherapy, chemotherapy or surgery according to clinical experience, and then determine the proper treatment scheme of patients. Therefore, the clinical experience of the doctor is decisive for the final clinical decision, and if the clinical experience of the doctor is less, the judgment will be affected, so that it is urgently needed to provide more abundant reference information to assist the doctor in judging.

Disclosure of Invention

In view of the above, it is necessary to provide a method, an apparatus, a computer device, and a storage medium for recommending a lung cancer treatment plan, which can recommend a lung cancer treatment plan for a doctor.

In a first aspect, the present application provides a method for recommending a lung cancer treatment regimen. The method comprises the following steps:

acquiring a four-dimensional medical image of a target object, and determining a plurality of functional region images based on the four-dimensional medical image, wherein the four-dimensional medical image is an image of a lung including the target object, and the plurality of functional region images are images of a plurality of functional regions of the lung;

determining a target treatment category for each functional zone based on the plurality of functional zone images, the four-dimensional medical image, and a scoring model;

determining a target treatment plan for the target object based on the target treatment category for each functional zone and the four-dimensional medical image.

In one embodiment, the determining a plurality of functional region images based on the four-dimensional medical image comprises:

and inputting the four-dimensional medical image into a trained delineation model to obtain images of a plurality of functional areas of the lung, wherein the trained delineation model is obtained by training the delineation model based on a plurality of training four-dimensional medical images of the lung and a reference functional area image set of each training four-dimensional medical image until the training is completed.

In one embodiment, the determining a target treatment category for each functional zone based on the plurality of functional zone images, the four-dimensional medical image, and a scoring model comprises:

determining a functional parameter of any functional area based on any functional area image;

determining a target score of any functional region based on the four-dimensional medical image, the plurality of functional region images, the functional parameter of any functional region, and the scoring model;

and determining the target treatment category of any functional area in a plurality of preset treatment categories according to the target score of any functional area.

In one embodiment, the determining a target score of any functional region based on the four-dimensional medical image, the plurality of functional region images, and the functional parameter of any functional region and the scoring model includes:

inputting the four-dimensional medical image, the plurality of functional region images and the functional parameters of any functional region into the scoring model to obtain a first scoring set of any functional region, wherein the first scoring set comprises scores of the plurality of treatment categories;

and taking the highest score in the first score set of any functional area as the target score of any functional area.

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

sketching the four-dimensional medical image to obtain a general tumor target volume image, a clinical tumor target volume image and a plurality of organs at risk images;

determining a target treatment regimen for the target subject based on the target treatment category for each functional zone, the general tumor target volume image, the clinical tumor target volume image, and the number of organ-at-risk images.

In one embodiment, the determining a target treatment plan for the target subject based on the target treatment category for each functional zone, the general tumor target volume image, the clinical tumor target volume image, and the number of organ-at-risk images comprises:

inputting the target treatment category, the gross tumor target volume image, the clinical tumor target volume image, and the number of organs-at-risk images for any functional zone into a target classification model, resulting in a second set of scores for the any functional zone, wherein the second set of scores comprises: scoring a plurality of preset protocols for the target treatment category;

taking a preset scheme corresponding to the highest score in the second score set of any functional area as a treatment scheme of any functional area;

determining a target treatment plan for the target object based on the treatment plan for each functional zone.

In one embodiment, after determining the target treatment plan for the target subject based on the target treatment category for each functional zone, the general tumor target volume image, the clinical tumor target volume image, and the number of organs-at-risk images, the method further comprises:

if the treatment plan of the target object comprises any target radiotherapy plan, determining any target radiotherapy plan according to any target radiotherapy plan, wherein the target radiotherapy plan comprises: target irradiation ray dose, target rotation angle of radiotherapy equipment and target grating position;

determining a probability of radiation pneumonitis for the any target radiotherapy plan based on the any target radiotherapy plan and the four-dimensional medical image;

and if the probability of the radiation pneumonia of any target radiotherapy plan is greater than a preset pneumonia threshold value, adjusting any target radiotherapy plan of the target object.

In one embodiment, the adjusting any target radiotherapy treatment plan of the target object includes:

for any candidate radiotherapy scheme in a plurality of preset schemes corresponding to radiotherapy, determining the candidate radiation pneumonitis probability of any candidate radiotherapy scheme according to any candidate radiotherapy preset scheme and the four-dimensional medical image;

and selecting any candidate radiation pneumonia probability smaller than the preset pneumonia threshold value from all candidate radiation pneumonia probabilities, and taking a candidate radiotherapy scheme corresponding to any selected candidate radiation pneumonia probability as a target radiotherapy scheme of the target object.

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

a functional region image determination module for acquiring a four-dimensional medical image of a target object, and determining a plurality of functional region images based on the four-dimensional medical image, wherein the four-dimensional medical image is an image of a lung including the target object, and the plurality of functional region images are images of a plurality of functional regions of the lung;

a treatment category determination module for determining a target treatment category for each functional zone based on the plurality of functional zone images, the four-dimensional medical image, and a scoring model;

and the recommendation module is used for determining a target treatment scheme of the target object based on the target treatment category of each functional area and the four-dimensional medical image.

In a third aspect, the present application also provides a computer device. The computer device comprises a memory storing a computer program and a processor implementing the following steps when executing the computer program:

In a fourth aspect, the present application further provides a computer-readable storage medium. The computer-readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps of:

In a fifth aspect, the present application further provides a computer program product. The computer program product comprising a computer program which when executed by a processor performs the steps of:

According to the method, the device, the computer equipment and the storage medium for recommending the lung cancer treatment scheme, the image of each functional area of the lung is determined according to the four-dimensional medical image of the target object, the functional areas are scored through deep learning based on the image of each functional area, the target treatment category of each functional area is obtained, namely the target treatment category is determined according to the tolerance of the functional areas, the appropriate target treatment scheme is recommended for the target object according to the four-dimensional medical image and the target treatment category of each functional area, and therefore richer reference information can be provided for a clinician to help the physician to determine the final treatment scheme.

Drawings

FIG. 1 is a schematic flow chart diagram illustrating a method for recommending a regimen for treating lung cancer in one embodiment;

FIG. 2 is a flow chart illustrating the steps of a method for recommending a regimen for treating lung cancer in an exemplary embodiment;

FIG. 3 is a block diagram of an apparatus for recommending a lung cancer treatment plan according to another embodiment;

FIG. 4 is a diagram illustrating an internal structure of a computer device according to an embodiment.

Detailed Description

In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.

In one embodiment, as shown in fig. 1, a method for recommending a lung cancer treatment plan is provided, and this embodiment is illustrated by applying the method to a terminal, it is to be understood that the method may also be applied to a server, and may also be applied to a system including a terminal and a server, and implemented by interaction between the terminal and the server. In this embodiment, the method includes the steps of:

s101, acquiring a four-dimensional medical image of a target object, and determining a plurality of functional area images based on the four-dimensional medical image.

The four-dimensional medical image is a time vector obtained by adding the fourth bit to the three-dimensional medical image, which is also called a real-time three-dimensional medical image, and the dynamic motion of the organ can be detected through the four-dimensional medical image. The four-dimensional medical image is obtained by photographing a target object by a four-dimensional medical imaging apparatus, the four-dimensional medical image is an image including a lung of the target object, the four-dimensional medical image further includes an organ at risk of the lung; the plurality of functional region images are images of a plurality of functional regions of the lungs; the plurality of functional regions are different regions divided based on a size affected by respiratory motion, and ventilation efficiencies of the plurality of functional regions are different from each other.

Specifically, a plurality of functional area images of the four-dimensional medical image can be determined through the sketching model; and inputting the multiple functional area images into the delineation model, and outputting the multiple functional area images of the lung through the delineation model.

S102, determining a target treatment category of each functional area based on the plurality of functional area images, the four-dimensional medical image and the scoring model.

Wherein the scoring model is used for obtaining a score of each treatment category of each functional zone and determining a target treatment category of each functional zone. Each of the treatment categories includes: radiotherapy, chemotherapy and surgery, wherein the target treatment category is radiotherapy, chemotherapy or surgery.

Specifically, a functional parameter is determined for each functional zone, the functional parameter reflecting the size of the functional zone affected by the respiratory motion. And for each functional area, inputting a plurality of four-dimensional medical images, a plurality of functional area images and functional parameters of the functional area into a scoring model to obtain a target treatment category of the functional area.

S103, determining a target treatment scheme of the target object based on the target treatment category of each functional area and the four-dimensional medical image.

Specifically, a general tumor target volume image, a clinical tumor target volume image and a plurality of organs-at-risk images are obtained according to the four-dimensional medical image delineation, then a treatment scheme of any functional area is determined in a plurality of preset schemes corresponding to a target treatment category of any functional area based on the general tumor target volume image, the clinical tumor target volume image, the plurality of organs-at-risk images and a classification model, the classification model is used for obtaining the score of each preset scheme, and the preset scheme with the highest score is used as the treatment scheme; and determining a target treatment plan according to the treatment plan of each functional area.

In the method for recommending the lung cancer treatment scheme, the image of each functional area of the lung is determined according to the four-dimensional medical image of the target object, each functional area is scored through deep learning based on the image of each functional area to obtain the target treatment category of each functional area, namely, the target treatment category is determined according to the tolerance of the functional area, and the appropriate target treatment scheme is recommended for the target object according to the four-dimensional medical image and the target treatment category of each functional area, so that richer reference information can be provided for a clinician, and the clinician can be helped to determine the final treatment scheme.

In one embodiment, S101 includes:

s111, inputting the four-dimensional medical image into the trained delineation model to obtain images of a plurality of functional areas of the lung.

Wherein the plurality of functional regions include: a fast region, a normal region and a slow region; the images of the plurality of functional regions are also four-dimensional images. The images of the plurality of functional regions include: fast zone image, normal zone image, slow zone image.

The functional areas are divided according to the size influenced by the respiratory motion, the fast area is greatly influenced by the respiratory motion, the slow area is less influenced by the respiratory motion, and the normal area is not influenced by the respiratory motion. The fast area is larger by the respiratory motion image, which means that any point in the fast area has larger displacement change along with the respiratory motion, the slow area is less influenced by the respiratory motion, which means that any point in the slow area has smaller displacement change along with the respiratory motion, the normal area is not influenced by the respiratory motion, and which means that any point in the normal area hardly generates displacement change along with the respiratory motion.

The trained delineation model is obtained by training the delineation model based on a plurality of training four-dimensional medical images of the lung and a reference functional area image set of each training four-dimensional medical image until the training is completed. The reference functional area image set comprises: the image of the reference fast area which is greatly influenced by respiration, the image of the reference slow area which is less influenced by respiration and the image of the reference normal area which is not influenced by respiration.

Specifically, the delineation model may be implemented by a decision tree, a random forest, or a deep neural network. During training, inputting a training four-dimensional medical image into a sketching model, outputting a training function area image set corresponding to the training four-dimensional medical image through the sketching model, calculating a first loss function value according to a reference function area image set and a training function area image set corresponding to the training four-dimensional medical image, modifying a model parameter of the sketching model according to the first loss function value, finishing one training, repeating the process of determining and outputting the training function area image set corresponding to the training medical image until the sketching model is converged, finishing the training, and obtaining the trained sketching model. Wherein the set of images of the training functional area comprises: the images of the training fast area which is greatly influenced by respiration, the images of the training slow area which is less influenced by respiration and the images of the training normal area which is not influenced by respiration.

In one embodiment, S102 includes:

and S121, determining the function parameters of any functional area based on any functional area image.

Wherein the functional parameters of any functional region at least comprise lung ventilation.

Specifically, the image of any functional region is a four-dimensional image, that is, the image of any functional region includes a plurality of respiratory images with different respiratory phases, and the functional parameter is determined according to the volume of any functional region in the plurality of respiratory images. For a plurality of respiratory images included in any functional area image, determining the volume of any functional area in each respiratory image, obtaining the maximum volume and the minimum volume, calculating the difference between the maximum volume and the minimum volume, obtaining the lung ventilation volume of any functional area, and further calculating the tidal volume, the supplementary expiratory volume and the functional residual capacity according to the lung ventilation volume.

And S122, determining a target score of any functional area based on the four-dimensional medical image, the plurality of functional area images, the functional parameters of any functional area and the scoring model.

Specifically, the four-dimensional medical image, the plurality of functional area images, and the functional parameter of any functional area are input into a scoring model to obtain a first scoring set of any functional area, and the highest score in the first scoring set is used as a target score of any functional area.

In one embodiment, S122 includes:

s1221, inputting the four-dimensional medical image, the plurality of 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.

Specifically, the four-dimensional medical image, the plurality of functional region images and the functional parameters of any functional region are input into the scoring model, and the scoring model outputs a score of each treatment category to obtain a first scoring set. That is, the first set of scores includes a score for each treatment category reflecting the probability that any one functional region will fit within each treatment category, respectively. The sum of the scores in the first score set is 1.

And S1222, using the highest score in the first score set of any functional region as the target score of any functional region.

Specifically, the highest score is the score with the largest value output by the scoring model, and the highest score is used as the target score of any functional region. For example, the first set of scores for any functional region includes: 0.1,0.3,0.6, then 0.6 is scored as the target.

And S123, determining the target treatment category of any functional area in a plurality of preset treatment categories according to the target score of any functional area.

Specifically, the first score set includes a plurality of first scores, the plurality of first scores correspond to a plurality of preset treatment categories one to one, and after the target score is determined, the treatment category corresponding to the target score is used as the target treatment category.

For example, the first set of scores for any functional region includes: 0.1,0.3,0.6, wherein 0.1 corresponds to the treatment category: radiotherapy, the treatment category corresponding to 0.3 is chemotherapy, and the treatment category corresponding to 0.6 is surgery; 0.6 score 0.6 as target, 0.6 for treatment category: surgery, as the target treatment category.

If two first scores in the first score set are the same and higher than the other first score, the treatment categories corresponding to the two same first scores can be both used as the target treatment categories to obtain the determined combined treatment category, for example: radiotherapy, chemotherapy treatment category, or preoperative radiotherapy, etc. If all three first scores in the first score set are the same, determining a combined treatment scheme of radiotherapy, chemotherapy and surgery, such as preoperative radiotherapy, postoperative chemotherapy and the like.

The scoring model is obtained by training a first preset model based on a plurality of training four-dimensional medical images, a reference function area image set of each training four-dimensional medical image, a reference function parameter set and a reference scoring set of each reference function area until training is completed. The reference functional area image set comprises a plurality of reference functional area images, and the reference functional parameter set comprises: the reference functional parameters of each reference functional area image and the reference score set of each reference functional area comprise reference scores of each reference functional area suitable for chemotherapy, radiotherapy and operation respectively, only one reference score in the reference score set of each reference functional area is not 0, and two reference scores are 0.

Specifically, the first preset model may be implemented by a decision tree, a random forest, or a deep neural network. During training, inputting a training four-dimensional medical image, an image set of a reference functional area of the training four-dimensional medical image and a reference functional parameter set into a first preset model to obtain a first training evaluation set of any reference functional area, taking the highest score in the first training evaluation set of any reference functional area as a training target score of any reference functional area, obtaining the training evaluation set of the training four-dimensional medical image according to the training target score of each reference functional area, calculating a second loss function value according to the training evaluation set of the training four-dimensional medical image and the reference evaluation set, modifying parameters of the first preset model according to the second loss function value, finishing one training, repeating the process of determining the training evaluation set until the first preset model is converged, finishing the training, and obtaining a evaluation model.

In one embodiment, S103 includes:

s311, the four-dimensional medical image is sketched to obtain a general tumor target volume image, a clinical tumor target volume image and a plurality of organs at risk images.

Wherein, the Gross Tumor target Volume (GTV) image refers to an image of the Tumor site and Tumor range which is directly visible or touchable and can be confirmed by a diagnostic examination means; a clinical tumor target volume (CTV) image, which refers to a tumor and potentially invaded tissue that have been determined to be present; organs at risk for lung cancer include: bilateral (or left or right), esophagus, heart (or pericardium), great vessel, spinal cord, trachea (or proximal bronchial tree), chest wall, ribs, skin, stomach, and liver.

In particular, the four-dimensional medical image is sketched to obtain a general tumor target volume image, a clinical tumor target volume image and a plurality of organs-at-risk images, which can be realized by the existing PV-iRT intelligent radiotherapy auxiliary system.

S312, determining a target treatment plan of the target object based on the target treatment category of each functional area, the general tumor target volume image, the clinical tumor target volume image and the several organs-at-risk images.

Specifically, the target treatment category of each functional area, the general tumor target volume image, the clinical tumor target volume image and the multiple organs-at-risk images are input into a target classification model to obtain a scheme score of each functional area, a treatment scheme corresponding to the scheme score of each functional area is determined in a preset scheme set, and a target treatment scheme is determined according to the treatment scheme of each functional area.

In one embodiment, S312 includes:

s3121, inputting the target treatment category, the gross tumor target volume image, the clinical tumor target volume image and the multiple organs-at-risk images of any functional area into a target classification model to obtain a second evaluation set of any functional area.

Specifically, the second scoring set includes: the second score set comprises a plurality of second scores, the plurality of second scores correspond to the plurality of preset schemes corresponding to the target category in a one-to-one manner, and all the second scores in the second score set are equal to 1. And the target classification model is used for determining a second score of any functional area in a plurality of preset schemes corresponding to the target treatment category.

The multiple preset schemes corresponding to radiotherapy comprise: three-dimensional conformal radiotherapy 3DCRT, conformal intensity modulated radiotherapy IMRT, volume arc intensity modulated radiotherapy technology VMRT and stereotactic body radiotherapy SBRT; multiple pre-set regimens for chemotherapy include: a plurality of chemotherapeutic agents; the plurality of preset schemes corresponding to the operation comprise: partial resection, segmental resection, lobectomy, bronchiolar lobectomy, bronchopulmonary sleeve-shaped lobectomy, carina reconstruction, and total lung resection.

For example, the target treatment category of the slow zone is radiotherapy, which correspondingly inputs the target treatment category, the general tumor target volume image, the clinical tumor target volume image and the several images of the organs at risk into the target classification model, resulting in a second set of scores for the slow zone, assuming that the second set of scores includes: 0.15,0.1,0.05,0.7, assuming that 0.15 corresponds to the predetermined scheme: the three-dimensional conformal radiotherapy 3DCRT, 0.1 corresponds to a preset scheme: IMRT, 0.05 corresponds to a preset protocol: the volume arc intensity modulated radiation therapy technology VMRT is a preset scheme corresponding to 0.7 and comprises the following steps: stereotactic body radiotherapy SBRT; the probability of the slow region being suitable for the three-dimensional conformal radiotherapy 3DCRT is 0.15, the probability of the slow region being suitable for the conformal intensity modulated radiotherapy IMRT is 0.1, the probability of the slow region being suitable for the volume arc intensity modulated radiotherapy technology VMRT is 0.05, and the probability of the slow region being suitable for the stereotactic body radiotherapy SBRT is 0.7.

The target classification model is obtained by training a second preset model based on a plurality of training data sets until training is completed, wherein each training data set comprises: training treatment categories, training gross tumor target volume images, training clinical tumor receptor volume images, training organs-at-risk images, and reference treatment protocols.

Specifically, the second preset model may be implemented by a decision tree, a random forest, or a deep neural network. During training, inputting a training treatment type, a training gross tumor target volume image, a training clinical tumor body volume image and a plurality of training organs at risk image in a training data set into a second preset model to obtain a training treatment scheme corresponding to the training data set, adjusting parameters of the second preset model according to a reference treatment precaution and the training treatment scheme to finish one training, repeating the process of determining the training treatment scheme until the second preset model converges, finishing the training to obtain a target classification model.

S313, taking the preset scheme corresponding to the highest score in the second score set of any functional area as the treatment scheme of any functional area.

And S314, determining a target treatment scheme of the target object according to the treatment scheme of each functional area.

Specifically, the target treatment regimen for the target subject includes: a treatment plan for the slow zone, a treatment plan for the fast zone, and a treatment plan for the normal zone.

For example, the target treatment category for the fast zone is radiotherapy, assuming that the second score set corresponding to the fast zone includes: 0.15,0.1,0.05,0.7, wherein the preset scheme corresponding to 0.7 is as follows: suitable for stereotactic body radiotherapy SBRT; the target treatment category for the normal zone is chemotherapy, assuming that the second scoring set for the normal zone includes: 0.2,0.05,0.75, wherein the preset scheme corresponding to 0.75 is as follows: 1 part of chemotherapeutic drugs; assuming that the target treatment category of the slow zone is radiotherapy, assuming that the second score set corresponding to the slow zone comprises: 0.05,0.05,0.15,0.75, wherein the preset scheme corresponding to 0.75 is as follows: suitable for stereotactic body radiotherapy SBRT, the target treatment plan corresponding to the target comprises: fast zone: stereotactic body radiotherapy SBRT, slow zone: stereotactic body radiotherapy SBRT, normal area: chemotherapeutic agent 1.

To facilitate understanding of the above recommendation method for a lung cancer treatment protocol, in a specific embodiment, referring to fig. 2, the recommendation method for a lung cancer treatment protocol comprises:

according to the four-dimensional medical image of the target object, a plurality of functional area images of the lung are obtained by using the trained delineation model, each target score is obtained by using the scoring model according to the plurality of functional area images and the four-dimensional medical image, and then the target treatment category of each functional area is determined in a plurality of preset treatment categories. According to the four-dimensional medical image, a general tumor target volume image, a clinical tumor target volume image and a plurality of organs at risk image are obtained through sketching, according to the general tumor target volume image, the clinical tumor target volume image and the plurality of organs at risk image, a target classification model is utilized, in a plurality of preset schemes corresponding to target treatment types of any functional area, the treatment scheme of any functional area is determined, and then a target radiotherapy scheme is obtained.

In one embodiment, if the treatment type of any functional region is radiotherapy, the probability that the target radiotherapy scheme used in any functional region causes radiation pneumonitis can be further predicted, and if the probability that the target radiotherapy scheme used in any functional region causes radiation pneumonitis is higher, the target radiotherapy scheme needs to be adjusted.

After S103, the method further comprises:

s104, if the treatment plan of the target object comprises any target radiotherapy plan, determining any target radiotherapy plan according to any target radiotherapy plan.

Specifically, the target radiotherapy treatment plan is a treatment plan of radiotherapy category of any functional region of the lung. The treatment plan of the target object comprises any target radiotherapy plan, which means that the treatment category with any functional area is radiotherapy, that is, the treatment plan with any functional area is determined in a plurality of preset plans corresponding to the radiotherapy. If the plurality of functional regions include: slow zone, fast zone, and normal zone, the treatment plan for the target object may include: one target radiotherapy plan (with either functional zone employing a target radiotherapy plan), or two target radiotherapy plans (with two functional zones employing a target radiotherapy plan), or three target radiotherapy plans (with each functional zone employing a target radiotherapy plan).

For any target radiotherapy plan of the target object, determining any target radiotherapy plan according to any target radiotherapy plan, comprising: and acquiring the dose of any target radiotherapy plan, and determining any target radiotherapy plan by the existing automatic optimization algorithm according to the dose and any target radiotherapy plan. The dose of any target radiotherapy scheme can be the reference dose of any target scheme and can be the dose set by a doctor according to any target radiotherapy scheme. The target radiotherapy plan comprises: target irradiation radiation dose, target rotation angle of radiotherapy equipment and target grating position.

If the treatment plan of the target object includes a plurality of target radiotherapy plans, a target radiotherapy plan of each target radiotherapy plan can be obtained through an existing automatic optimization algorithm.

S105, determining the probability of the radiation pneumonitis of any target radiotherapy plan based on the any target radiotherapy plan and the four-dimensional medical image.

In particular, determining a gross tumor target volume image, the clinical tumor target volume image, and the number of organ-at-risk images from the four-dimensional medical image; determining a dose distribution image from any of the target radiotherapy plan, the gross tumor target volume image, the clinical tumor target volume image, and the number of images of organs-at-risk using a monte carlo algorithm; the dose distribution image is used for reflecting 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.

And inputting the dose distribution image and the four-dimensional medical image into a pneumonia prediction model to obtain the probability of the radiation pneumonia of any target radiotherapy plan. The pneumonia prediction model is obtained by training a third preset model based on a plurality of training medical images, a training dose distribution image of each medical image and a radioactive pneumonia label until training is completed.

S106, if the probability of the radiation pneumonitis of any target radiotherapy plan is larger than a preset pneumonia threshold, any target radiotherapy scheme of the target object is adjusted.

Specifically, if the probability of radiation pneumonitis of any target radiotherapy plan is greater than the preset pneumonia threshold, it indicates that the radiation pneumonitis may be caused by the any target radiotherapy plan, and it is necessary to adjust the any target radiotherapy plan. The adjusted target radiotherapy scheme may be determined in a plurality of preset schemes corresponding to radiotherapy, so that the probability of radiation pneumonitis corresponding to the adjusted target radiotherapy scheme is smaller than a preset pneumonia threshold.

In one embodiment, S106 includes:

s161, for any candidate radiotherapy scheme in a plurality of preset schemes corresponding to radiotherapy, determining the candidate radiation pneumonitis probability of any candidate radiotherapy scheme according to any candidate radiotherapy preset scheme and the four-dimensional medical image.

Specifically, the any candidate radiotherapy treatment plan is any one preset plan except the target radiotherapy treatment plan among a plurality of corresponding preset plans. Determining any candidate radiotherapy plan according to any candidate radiotherapy plan, comprising: and acquiring the dose of any candidate radiotherapy plan, and determining any candidate radiotherapy plan by an existing automatic optimization algorithm according to the dose of any candidate radiotherapy plan and the dose scheme of any candidate radiotherapy plan. Wherein the dose of any candidate radiotherapy treatment plan can be the reference dose of any candidate radiotherapy treatment plan.

Determining a gross tumor target volume image, the clinical tumor target volume image, and the number of organ-at-risk images from the four-dimensional medical image; determining candidate dose distribution images from any of the candidate radiotherapy plans, the gross tumor target volume image, the clinical tumor target volume image, and the number of images of organs-at-risk using a monte carlo algorithm; the candidate dose distribution images are used to reflect the radiation dose at the gross tumor target, the clinical tumor target, and the organs-at-risk region under any candidate radiotherapy plan.

And inputting the candidate dose distribution image and the four-dimensional medical image into a pneumonia prediction model to obtain the candidate radiation pneumonia probability of any candidate radiotherapy plan.

And S162, selecting any candidate radiation pneumonia probability smaller than the preset pneumonia threshold value from all candidate radiation pneumonia probabilities, and taking a candidate radiotherapy scheme corresponding to any selected candidate radiation pneumonia probability as a target radiotherapy scheme of the target object.

Specifically, the candidate radiation therapy plan corresponding to the candidate radiation pneumonia probability of each candidate radiation therapy plan in a plurality of preset plans corresponding to radiation therapy is determined, and the candidate radiation therapy plan corresponding to the candidate radiation pneumonia probability smaller than a preset pneumonia threshold value is used as a target radiation therapy plan corresponding to a target.

And if the candidate radiation pneumonitis probabilities of at least two candidate radiation therapy schemes are smaller than a preset pneumonitis threshold value, determining second scores of the at least two candidate radiation therapy schemes, and taking the candidate radiation therapy scheme with the highest second score as a target radiation therapy scheme.

If the probability of the candidate radiation pneumonitis without any candidate radiotherapy scheme is smaller than a preset pneumonitis threshold value, adjusting the target treatment type; since the first score corresponding to the radiotherapy is the maximum value in the first score set, other first scores that are only smaller than the first score corresponding to the radiotherapy are determined in the first score set, the treatment categories corresponding to the determined other first scores are used as the adjusted target treatment categories, and the step S103 is continued.

For example, the preset plurality of treatment categories includes radiotherapy, chemotherapy and surgery, if the first score of the surgery is greater than the first score of the chemotherapy in the first score set, the target treatment category is adjusted to surgery and the operation is continued to be performed S103, and if the first score of the chemotherapy is greater than the first score of the surgery, the target treatment category is adjusted to chemotherapy and the operation is continued to be performed S103.

In this embodiment, according to a four-dimensional medical image of a target object, an image of each functional area of a lung is obtained by using a trained delineation model, and a functional parameter of each functional area is determined, and according to the four-dimensional medical image, a plurality of functional area images, and a functional parameter of any functional area, a target score of any functional area is obtained by using a scoring model, and then a target treatment category of any functional area is determined in a plurality of preset treatment categories. According to the four-dimensional medical image, a general tumor target volume image, a clinical tumor target volume image and a plurality of organs at risk image are obtained through sketching, according to the general tumor target volume image, the clinical tumor target volume image and the plurality of organs at risk image, a target classification model is utilized, in a plurality of preset schemes corresponding to target treatment types of any functional area, the treatment scheme of any functional area is determined, and then a target radiotherapy scheme is obtained. If the treatment scheme of any functional area is the target radiotherapy scheme, predicting the probability of the radiation pneumonia of the target radiotherapy scheme, and if the probability of the radiation pneumonia is larger than the pneumonia threshold, determining a preset scheme with the probability of the radiation pneumonia smaller than the pneumonia threshold from a plurality of preset schemes corresponding to radiotherapy to serve as the target radiotherapy scheme. By the recommendation method for the lung cancer treatment scheme, richer reference information is provided for a clinician, a target treatment scheme is recommended for the clinician, and the physician is helped to determine the final treatment scheme.

It should be understood that, although the steps in the flowcharts related to the embodiments as described above are sequentially displayed as indicated by arrows, the steps are not necessarily performed sequentially as indicated by the arrows. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least a part of the steps in the flowcharts related to the embodiments described above may include multiple steps or multiple stages, which are not necessarily performed at the same time, but may be performed at different times, and the execution order of the steps or stages is not necessarily sequential, but may be rotated or alternated with other steps or at least a part of the steps or stages in other steps.

Based on the same inventive concept, the embodiment of the present application further provides a recommendation device for a lung cancer treatment plan, which is used for implementing the recommendation method for a lung cancer treatment plan. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the method, so that specific limitations in the following embodiments of one or more recommendation devices for lung cancer treatment schemes can be referred to the limitations of the recommendation method for lung cancer treatment schemes, and are not described herein again.

In one embodiment, as shown in fig. 3, there is provided a recommendation apparatus for a lung cancer treatment plan, including: functional area image determination module, treatment category determination module and recommendation module, wherein:

In one embodiment, the functional region image determination module includes: a first delineation unit, wherein:

the first delineation unit is used for inputting the four-dimensional medical image into a trained delineation model to obtain images of a plurality of functional areas of the lung, wherein the trained delineation model is obtained by training the delineation model based on a plurality of training four-dimensional medical images of the lung and a reference functional area image set of each training four-dimensional medical image until training is completed.

In one embodiment, the treatment category determination module comprises: parameter determining unit, first scoring unit and category determining unit, wherein:

the parameter determining unit is used for determining the functional parameters of any functional area based on any functional area image;

a first scoring unit, configured to determine a target score of any functional region based on the four-dimensional medical image, the plurality of functional region images, the functional parameter of the any functional region, and the scoring model;

and the category determining unit is used for determining a target treatment category of any functional area in a plurality of preset treatment categories according to the target score of any functional area.

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

a first subunit, configured to input the four-dimensional medical image, the multiple functional region images, and the functional parameter of any functional region into the scoring model, so as to obtain a first scoring set of any functional region, where the first scoring set includes scores of the multiple treatment categories;

and the second word unit is used for taking the highest score in the first score set of any functional area as the target score of any functional area.

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

the second delineation unit is used for delineating the four-dimensional medical image to obtain a general tumor target volume image, a clinical tumor target volume image and a plurality of organs at risk images;

a protocol determination unit for determining a target treatment protocol for the target subject based on the target treatment category for each functional zone, the general tumor target volume image, the clinical tumor target volume image, and the number of organ-at-risk images.

In one embodiment, the scheme determination unit includes: a third subunit, a fourth subunit, and a fifth subunit, wherein:

a third subunit for inputting the target treatment category, the general tumor target volume image, the clinical tumor target volume image and the plurality of organs-at-risk images of any functional area into a target classification model to obtain a second scoring set of the any functional area, wherein the second scoring set comprises: scoring a plurality of preset protocols for the target treatment category;

a fourth subunit, configured to use a preset scheme corresponding to a highest score in the second score set of any functional region as a treatment scheme for the any functional region;

a fifth subunit, configured to determine a target treatment plan of the target object according to the treatment plan of each functional region.

In one embodiment, the apparatus for recommending a lung cancer treatment plan further comprises: first adjustment module, second adjustment module and third adjustment module, wherein:

a first adjusting module, configured to determine any target radiotherapy plan according to any target radiotherapy plan if the treatment plan of the target object includes any target radiotherapy plan, where the target radiotherapy plan includes: target irradiation ray dose, target rotation angle of radiotherapy equipment and target grating position;

a second adjustment module for determining a probability of radiation pneumonitis for the any target radiotherapy plan based on the any target radiotherapy plan and the four-dimensional medical image;

and the third adjusting module is used for adjusting any target radiotherapy scheme of the target object if the probability of the radiation pneumonia of any target radiotherapy plan is greater than a preset pneumonia threshold.

In one embodiment, the third adjustment module comprises: a first adjusting unit and a second adjusting unit, wherein:

the first adjusting unit is used for determining the candidate radiation pneumonitis probability of any candidate radiotherapy scheme according to any candidate radiotherapy preset scheme and the four-dimensional medical image for any candidate radiotherapy scheme in a plurality of preset schemes corresponding to radiotherapy;

and the second adjusting unit is used for selecting any candidate radiation pneumonia probability smaller than the preset pneumonia threshold value from all the candidate radiation pneumonia probabilities, and taking the candidate radiotherapy scheme corresponding to the selected any candidate radiation pneumonia probability as the target radiotherapy scheme of the target object.

The modules in the recommendation device for a lung cancer treatment scheme can be wholly or partially realized by software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.

In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as shown in fig. 4. The computer device includes a processor, a memory, a communication interface, a display screen, and an input device connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The communication interface of the computer device is used for carrying out wired or wireless communication with an external terminal, and the wireless communication can be realized through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. The computer program is executed by a processor to implement a method of recommending a lung cancer treatment plan. The display screen of the computer equipment can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer equipment can be a touch layer covered on the display screen, a key, a track ball or a touch pad arranged on the shell of the computer equipment, an external keyboard, a touch pad or a mouse and the like.

Those skilled in the art will appreciate that the architecture shown in fig. 4 is merely a block diagram of some of the structures associated with the disclosed aspects and is not intended to limit the computing devices to which the disclosed aspects apply, as particular computing devices may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.

In one embodiment, a computer device is provided, comprising a memory and a processor, the memory having a computer program stored therein, the processor implementing the following steps when executing the computer program:

In one embodiment, the processor, when executing the computer program, further performs the steps of: the determining a plurality of functional region images based on the four-dimensional medical image comprises:

In one embodiment, the processor, when executing the computer program, further performs the steps of: the determining a target treatment category for each functional zone based on the plurality of functional zone images, the four-dimensional medical image, and a scoring model comprises:

In one embodiment, the processor, when executing the computer program, further performs the steps of: the determining a target score of any functional region based on the four-dimensional medical image, the plurality of functional region images, and the functional parameter of any functional region and the scoring model comprises:

In one embodiment, the processor, when executing the computer program, further performs the steps of: the determining a target treatment plan for the target object based on the target treatment category for each functional zone and the four-dimensional medical image comprises:

In one embodiment, the processor, when executing the computer program, further performs the steps of: the determining a target treatment plan for the target subject based on the target treatment category for each functional zone, the general tumor target volume image, the clinical tumor target volume image, and the number of organ-at-risk images, comprising:

In one embodiment, the processor, when executing the computer program, further performs the steps of: after said determining a target treatment plan for the target subject based on the target treatment category for each functional zone, the general tumor target volume image, the clinical tumor target volume image, and the number of organ-at-risk images, further comprising:

In one embodiment, the processor, when executing the computer program, further performs the steps of: the adjusting any target radiotherapy plan of the target object comprises:

In one embodiment, a computer-readable storage medium is provided, having a computer program stored thereon, which when executed by a processor, performs the steps of:

In one embodiment, the computer program when executed by the processor further performs the steps of: the determining a plurality of functional region images based on the four-dimensional medical image comprises:

In one embodiment, the computer program when executed by the processor further performs the steps of: the determining a target treatment category for each functional zone based on the plurality of functional zone images, the four-dimensional medical image, and a scoring model comprises:

In one embodiment, the computer program when executed by the processor further performs the steps of: the determining a target score of any functional region based on the four-dimensional medical image, the plurality of functional region images, and the functional parameter of any functional region and the scoring model comprises:

In one embodiment, the computer program when executed by the processor further performs the steps of: the determining a target treatment plan for the target object based on the target treatment category for each functional zone and the four-dimensional medical image comprises:

In one embodiment, the computer program when executed by the processor further performs the steps of: the determining a target treatment plan for the target subject based on the target treatment category for each functional zone, the general tumor target volume image, the clinical tumor target volume image, and the number of organ-at-risk images, comprising:

In one embodiment, the computer program when executed by the processor further performs the steps of: after said determining a target treatment plan for the target subject based on the target treatment category for each functional zone, the general tumor target volume image, the clinical tumor target volume image, and the number of organ-at-risk images, further comprising:

In one embodiment, the computer program when executed by the processor further performs the steps of: the adjusting any target radiotherapy plan of the target object comprises:

In one embodiment, a computer program product is provided, comprising a computer program which, when executed by a processor, performs the steps of:

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, presented data, etc.) referred to in the present application are information and data authorized by the user or sufficiently authorized by each party.

It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, database, or other medium used in the embodiments provided herein may include at least one of non-volatile and volatile memory. The nonvolatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical Memory, high-density embedded nonvolatile Memory, resistive Random Access Memory (ReRAM), Magnetic Random Access Memory (MRAM), Ferroelectric Random Access Memory (FRAM), Phase Change Memory (PCM), graphene Memory, and the like. Volatile Memory can include Random Access Memory (RAM), external cache Memory, and the like. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM), among others. The databases referred to in various embodiments provided herein may include at least one of relational and non-relational databases. The non-relational database may include, but is not limited to, a block chain based distributed database, and the like. The processors referred to in the embodiments provided herein may be general purpose processors, central processing units, graphics processors, digital signal processors, programmable logic devices, quantum computing based data processing logic devices, etc., without limitation.

The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.

The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the present application. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection 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 treatment for lung cancer, said method comprising:

2. The method of claim 1, wherein determining a plurality of functional region images based on the four-dimensional medical image comprises:

3. The method of claim 1, wherein determining a target treatment category for each functional zone based on the plurality of functional zone images, the four-dimensional medical image, and a scoring model comprises:

4. The method of claim 3, wherein determining the target score for the any functional region based on the four-dimensional medical image, the plurality of functional region images, and the functional parameter of the any functional region and the scoring model comprises:

5. The method of claim 1, wherein determining a target treatment plan for the target object based on the target treatment category for each functional zone and the four-dimensional medical image comprises:

6. The method of claim 5, wherein determining a target treatment plan for the target subject based on the target treatment category for each functional zone, the general tumor target volume image, the clinical tumor target volume image, and the number of organ-at-risk images comprises:

7. The method of claim 5, wherein after determining the target treatment plan for the target subject based on the target treatment category for each functional zone, the general tumor target volume image, the clinical tumor target volume image, and the number of organs-at-risk images, further comprising:

8. The method of claim 7, wherein said adjusting any target radiotherapy protocol of said target object comprises:

9. An apparatus for recommending a lung cancer treatment plan, said apparatus comprising:

10. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor, when executing the computer program, implements the steps of the method of any of claims 1 to 8.

11. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 8.