Lung development evaluation system, method and equipment based on CT (Computed Tomography) image and medium

Abstract

The invention relates to the technical field of medical detection, and discloses a lung development evaluation system, method and device based on a CT image and a medium. The system comprises an image segmentation device used for segmenting and marking a lung CT image sample by using an image segmentation model to obtain a lung image mask and sending the lung image mask to a data processing device, and the data processing device used for obtaining shooting parameters corresponding to a lung CT image, lung parameters corresponding to the lung CT image are obtained through calculation according to the shooting parameters and the lung image mask, the lung parameters are sent to model construction equipment, the model construction equipment is used for obtaining physiological information corresponding to the lung CT image, a lung development evaluation model is constructed according to the lung parameters and the physiological information, and the lung development evaluation model is used for evaluating lung development. And the development evaluation equipment is used for processing the to-be-evaluated lung CT image by using the image segmentation model and the lung development evaluation model to obtain an evaluation result. Various parameters of the lung can be automatically measured, and the lung development condition can be accurately evaluated.

Description

Technical Field

[0001] The present invention relates to the technical field of medical detection, and particularly to a lung development evaluation system, method, device and medium based on CT images. Background Art

[0002] In the research on children's lung development, traditional research methods mainly rely on doctors to roughly measure a small number of lung parameters manually on two-dimensional X-ray images or limited CT scan slices, such as basic dimensions like simple lung lobe length and width. For the evaluation of children's lung development, there is no unified standard. Most of the existing children's lung development evaluation methods directly apply the adult lung image analysis model to children's data for rough analysis and evaluation, without considering the dynamic change characteristics of children's lung development and the correlation between children's unique physiological characteristics and lung growth. The evaluation results are often not accurate enough. Summary of the Invention

[0003] To solve the problem that the existing lung development evaluation methods are not accurate enough for the evaluation of children's lung development, the present application provides a lung development evaluation system, method, device and medium based on CT images.

[0004] In a first aspect, the present invention provides a lung development evaluation system based on CT images, including:

[0005] An image segmentation device, a data processing device, a model construction device and a development evaluation device. The image segmentation device is communicatively connected to the data processing device, the data processing device is communicatively connected to the model construction device, and the model construction device is communicatively connected to the development evaluation device;

[0006] The image segmentation device is configured to segment and label a lung CT image sample using an image segmentation model to obtain a lung image mask, and send the lung image mask to the data processing device;

[0007] The data processing device is configured to obtain the shooting parameters corresponding to the lung CT image, calculate the lung parameters corresponding to the lung CT image according to the shooting parameters and the lung image mask, and send the lung parameters to the model construction device;

[0008] The model construction device is configured to obtain the physiological information corresponding to the lung CT image, and construct a lung development evaluation model according to the lung parameters and the physiological information;

[0009] The development evaluation device is configured to process the to-be-evaluated lung CT image using the image segmentation model and the lung development evaluation model to obtain an evaluation result.

[0010] In an alternative embodiment, the evaluation system further includes a model training device, which is configured to perform the following training steps:

[0011] Step 1: Obtain multiple lung CT images, crop each of the lung CT images to a preset size, and perform annotation to obtain annotated images;

[0012] Step 2: Input the annotated images into an initial segmentation model, and obtain the output result of the initial segmentation model;

[0013] Step 3: Use a loss function to calculate the loss value between the actual mask corresponding to the annotated image and the output result;

[0014] Step 4: If the loss value is greater than or equal to a loss value threshold, optimize the parameters of the initial segmentation model to obtain an intermediate segmentation model;

[0015] Repeat steps 2 to 4 until the loss value is less than the loss value threshold, and use the intermediate segmentation model corresponding to the loss value less than the loss value threshold as the image segmentation model.

[0016] In an alternative embodiment, the model training device is further configured to construct the loss function;

[0017] The loss function is:

[0018]

[0019] wherein, there are N samples in the annotated image, i represents the i-th sample, C represents the mask type included in each sample, each pixel in the sample corresponds to a mask type, and x ic represents the score corresponding to each mask type in the i-th sample.

[0020] In an alternative embodiment, the data processing device is further configured to:

[0021] Obtain the shooting parameters corresponding to the lung CT image and the mask parameters corresponding to the lung image mask, where the shooting parameters include pixel height, pixel width, and the thickness of the transverse plane;

[0022] Calculate the lung parameters according to the pixel height, the pixel width, the thickness of the transverse plane, and the mask parameters.

[0023] In an alternative embodiment, the lung parameters include the left lung volume and the right lung volume;

[0024] The left lung volume V l The calculation formula is:

[0025] Vl = N l × h s × w s × s s

[0026] The right lung volume V r is calculated by the formula:

[0027] V r = N r × h s × w s × s s

[0028] where h s represents the height corresponding to each pixel, w s represents the width corresponding to each pixel, s s represents the thickness of the transverse plane, and N l represents the number of pixels in the left lung region, and N r represents the number of pixels in the right lung region.

[0029] In an alternative embodiment, the model construction device is further configured to:

[0030] Obtain the physiological information corresponding to the lung CT image, where the physiological information includes at least one of the BMI index, height, weight, body surface area, gender, and number of days since birth;

[0031] Correlate and match the lung parameters and the physiological information to obtain a matching result, and construct the lung development evaluation model according to the matching result.

[0032] In an alternative embodiment, the development evaluation device is further configured to:

[0033] Use the image segmentation model to segment the lung CT image to be evaluated, and obtain a lung image mask to be evaluated;

[0034] Obtain the physiological information to be evaluated corresponding to the lung CT image to be evaluated, and input the lung image mask to be evaluated and the physiological information to be evaluated into the lung development evaluation model to obtain the evaluation result.

[0035] In a second aspect, the present invention provides a method for evaluating lung development based on CT images, including:

[0036] Use an image segmentation model to segment and label a lung CT image sample to obtain a lung image mask;

[0037] Obtain the shooting parameters corresponding to the lung CT image, and calculate the lung parameters corresponding to the lung CT image according to the shooting parameters and the lung image mask;

[0038] Obtain the physiological information corresponding to the lung CT image, and construct a lung development evaluation model according to the lung parameters and the physiological information;

[0039] Use the image segmentation model and the lung development evaluation model to evaluate the lung CT image to be evaluated to obtain an evaluation result.

[0040] In a third aspect, the present invention provides a computer device, which includes a processor and a memory. The memory stores a computer program, and the processor is used to execute the computer program to implement the lung development evaluation method based on CT images described in the foregoing embodiments.

[0041] In a fourth aspect, the present invention provides a computer storage medium, which stores a computer program. When the computer program is executed on a processor, it implements the lung development evaluation method based on CT images described in the foregoing embodiments.

[0042] The embodiments of the present invention have the following beneficial effects:

[0043] The lung development evaluation system based on CT images provided by the present invention, by using a deep learning model combined with a traditional measurement correction method, accurately segments the images, automatically and precisely measures various lung parameters, greatly reduces the errors introduced by manual operations, and then constructs a lung development evaluation model by combining lung parameters and physiological parameters, providing a solid data basis for children's lung development research and clinical applications, and ensuring that the obtained parameters truthfully reflect the real state of the lungs. Description of the Drawings

[0044] In order to more clearly illustrate the technical solutions of the present invention, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention, and therefore should not be regarded as limiting the protection scope of the present invention. For those of ordinary skill in the art, other related drawings can also be obtained based on these drawings without creative efforts.

[0045] Figure 1 Shows a schematic structural diagram of a lung development evaluation system based on CT images provided by an embodiment of the present application;

[0046] Figure 2 Shows a schematic diagram of a transverse plane lung image provided by an embodiment of the present application;

[0047] Figure 3 Shows a schematic diagram of a lung segmentation mask image provided by an embodiment of the present application;

[0048] Figure 4Shows a schematic diagram of a bone segmentation mask image provided by an embodiment of the present application;

[0049] Figure 5 Shows a schematic diagram of a thoracic cavity segmentation mask image provided by an embodiment of the present application;

[0050] Figure 6 Shows a schematic diagram of a curve of the development change of the lung volume of children provided by an embodiment of the present application;

[0051] Figure 7 Shows a schematic diagram of the process of a lung development evaluation method based on CT images provided by an embodiment of the present application.

[0052] Description of main component symbols:

[0053] 100, lung development evaluation system based on CT images; 110, image segmentation device; 120, data processing device; 130, model construction device; 140, development evaluation device. Specific implementation manners

[0054] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments.

[0055] Generally, the components of the embodiments of the present invention described and illustrated herein can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed present invention, but merely represents selected embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative efforts fall within the scope of protection of the present invention.

[0056] Hereinafter, the terms "including", "having" and their cognates that can be used in various embodiments of the present invention are only intended to indicate specific features, numbers, steps, operations, elements, components or combinations of the foregoing items, and should not be construed as first excluding the existence of one or more other features, numbers, steps, operations, elements, components or combinations of the foregoing items or increasing the possibility of one or more features, numbers, steps, operations, elements, components or combinations of the foregoing items.

[0057] In addition, the terms "first", "second", "third", etc. are only used for distinguishing descriptions and cannot be understood as indicating or implying relative importance.

[0058] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which various embodiments of the present invention belong. The terms (such as those defined in a commonly used dictionary) will be interpreted as having the same meaning as the contextual meaning in the relevant technical field and will not be interpreted as having an idealized meaning or an overly formal meaning unless clearly defined in various embodiments of the present invention.

[0059] The following will, with reference to the accompanying drawings, elaborate on some embodiments of the present invention. Without conflict, the following embodiments and the features in the embodiments may be combined with each other.

[0060] Refer to Figure 1 , Figure 1 which is a schematic structural diagram of a lung development evaluation system 100 provided for this embodiment. The lung development evaluation system 100 includes:

[0061] an image segmentation device 110, a data processing device 120, a model construction device 130, and a development evaluation device 140. The image segmentation device 110 is communicatively connected to the data processing device 120, the data processing device 120 is communicatively connected to the model construction device 130, and the model construction device 130 is communicatively connected to the development evaluation device 140.

[0062] The image segmentation device 110 is configured to segment and label a pulmonary CT image sample by using an image segmentation model to obtain a pulmonary image mask, and send the pulmonary image mask to the data processing device 120.

[0063] First, it is necessary to obtain pulmonary CT images. The pulmonary CT images can be pulmonary CT images collected from a hospital database or obtained from other databases. The pulmonary CT images can cover various types of CT images of different ages, different heights and weights, different genders, etc., making the obtained pulmonary CT images more universal. Due to differences in age and gender, the lung development of children or minors will vary greatly, while the lung development of adults generally does not change significantly with factors such as age, and the evaluation technology for the lung conditions of adults is also relatively perfect. Therefore, the solution of this application is mainly for evaluating the lung development of children or minors.

[0064] The pulmonary CT images can be pulmonary CT images of children with normal development. Then, preprocess the pulmonary CT images, such as gray normalization processing and noise reduction processing, etc., to improve the quality of the pulmonary CT images and lay a foundation for the accurate evaluation of subsequent lung development. Remove the images that do not meet the requirements in the preprocessed pulmonary CT images, and use the remaining images as pulmonary CT image samples.

[0065] Then, the lung CT image samples are input into the image segmentation device 110, and the pre-trained image segmentation model in the image segmentation device 110 is used to segment and label the lung CT image samples. There can be multiple lung CT image samples. After each lung CT image sample is segmented and labeled, a lung image mask is obtained, and then the lung image mask is sent to the data processing device 120.

[0066] The data processing device 120 is configured to obtain the shooting parameters corresponding to the lung CT image, calculate the lung parameters corresponding to the lung CT image according to the shooting parameters and the lung image mask, and send the lung parameters to the model construction device 130.

[0067] Lung CT images are usually taken by a CT machine, which is also known as a computed tomography scanner. When taking images, the CT machine will set corresponding shooting parameters, which can include pixel spatial pitch, slice thickness, etc. Then, the lung parameters corresponding to the lung CT image can be calculated through the shooting parameters and the lung image mask. The lung parameters can include key structural parameters such as the cross-sectional plane area, volume, anteroposterior diameter, transverse diameter, superior-inferior diameter, and bronchial volume of the left and right lungs.

[0068] The model construction device 130 is configured to obtain the physiological information corresponding to the lung CT image and construct a lung development evaluation model according to the lung parameters and the physiological information.

[0069] Each lung CT image corresponds to a child. Therefore, the physiological information of each child, such as BMI index, height, weight, body surface area, gender, days of birth, etc., may be different. Therefore, a lung development evaluation model can be constructed based on the lung parameters and physiological information. For example, the physiological information and lung parameters of children of the same age and gender are linearly fitted to obtain an average value of lung parameters for that age group. Then, the fluctuation range of the average value of lung parameters is determined according to other lung parameters, and values within the fluctuation range can be considered normal lung parameters. Or other methods can also be used to construct the lung development evaluation model.

[0070] The development evaluation device 140 is configured to process the lung CT image to be evaluated using the image segmentation model and the lung development evaluation model to obtain an evaluation result.

[0071] The development evaluation device 140 is mainly used to input the lung CT image to be evaluated and the physiological information corresponding to the lung CT image to be evaluated, and output the corresponding evaluation result. The evaluation result can be normal development, underdevelopment, etc.

[0072] In this embodiment, by using a deep learning model in combination with traditional measurement correction methods, the images are accurately segmented, various lung parameters are automatically and accurately measured, greatly reducing the errors introduced by manual operations. Then, a lung development evaluation model is constructed by combining the lung parameters and physiological parameters, providing a solid data basis for children's lung development research and clinical applications, and ensuring that the obtained parameters truthfully reflect the real state of the lungs.

[0073] In one implementation, the evaluation system further includes a model training device, and the model training device is used to perform the following training steps:

[0074] Step 1: Obtain multiple lung CT images, crop each of the lung CT images to a preset size and perform annotation to obtain annotated images;

[0075] Step 2: Input the annotated images into an initial segmentation model, and obtain the output result of the initial segmentation model;

[0076] Step 3: Use a loss function to calculate the loss value between the actual mask corresponding to the annotated image and the output result;

[0077] Step 4: If the loss value is greater than or equal to the loss value threshold, optimize the parameters of the initial segmentation model to obtain an intermediate segmentation model;

[0078] Repeat steps 2 to 4 until the loss value is less than the loss value threshold, and use the intermediate segmentation model corresponding to the loss value less than the loss value threshold as the image segmentation model.

[0079] The initial segmentation model can adopt a deep convolutional neural network, such as networks like ResNet, U-Net, etc. Then, the obtained lung CT images are cropped to a preset size and annotated to obtain annotated images.

[0080] The annotated images corresponding to the lung CT images can refer to Figures 2 to 5 as shown.

[0081] Figure 2 This is a schematic diagram of a transverse plane lung image provided in this embodiment.

[0082] Figure 3 This is a schematic diagram of a lung segmentation mask image provided in this embodiment.

[0083] Figure 4 This is a schematic diagram of a bone segmentation mask image provided in this embodiment.

[0084] Figure 5 This is a schematic diagram of a thoracic cage segmentation mask image provided in this embodiment.

[0085] Specifically, Figures 2 - 5The black part in it represents the background, Figure 3 the white part represents the left lung, and the gray part represents the right lung, Figure 4 the white part represents the bones, Figure 5 the white part represents the chest wall. Through the above segmentation and annotation methods, masks of each part in the lung CT image can be obtained.

[0086] The preset size can be h×w pixels, where h is the image height and w is the image width, so as to obtain multiple lung CT image samples. The multiple lung CT image samples are divided into a training set and a validation set. For example, 80% of multiple randomly selected lung CT image samples are used as the training set, and the other 20% are used as the validation set.

[0087] The training set is input into the initial segmentation model. In the initial stage of training, the output of the initial segmentation model usually has a large gap from the lung mask corresponding to the training set. Therefore, it is necessary to adjust the parameters of the initial segmentation model, and then continue to train using the training set. After repeating the training for multiple rounds, an image segmentation model can be obtained. The output of the image segmentation model is the lung mask corresponding to the training set.

[0088] The ADAM optimizer can be used in the training process. The optimization parameters are set according to actual needs. For example, the learning rate is set to 0.0001, the number of images per batch is 8, and a total of 30 rounds of training are performed. After each round of training is completed, it is tested on the validation set until the trained model meets the preset conditions.

[0089] In this embodiment, the lung CT images are cropped to the preset size and annotated, then the initial segmentation model is trained, and the optimizer is used to optimize the parameters of the model, so that the finally obtained image segmentation model can accurately annotate and segment the lung CT images.

[0090] In one implementation manner, the model training device is further configured to construct the loss function;

[0091] The loss function is:

[0092]

[0093] wherein, there are N samples in the annotated image, i represents the i-th sample, C represents the mask types included in each sample, each pixel in the sample corresponds to a mask type, and x ic represents the score corresponding to each mask type in the i-th sample.

[0094] When annotating lung CT images, different masks can be used to annotate pixels in different parts. For example, 0 represents the background, 1 represents the left lung, and 2 represents the right lung. At this time, C = 3. For a sample, it will at least contain one of the 3 mask types. For the i-th sample, its true annotation is y i , and the true annotation represents the complete mask of the sample. The output of the model is {x i1 , x i2 , …, x iC}, where x ic represents the score corresponding to each mask type in the i-th sample. When C = 3, x i1 represents the score of the i-th sample belonging to the background, x i2 represents the score of the i-th sample belonging to the left lung, and x i3 represents the score of the i-th sample belonging to the right lung. Then, the type of the sample is determined according to the scores, and the loss value between the sample and the true annotation is calculated.

[0095] In this embodiment, by classifying and annotating lung CT images, the type of each CT image sample can be determined, and then the loss value between it and the true annotation is calculated. According to the loss value, it is judged whether the image segmentation model needs to be continuously trained and optimized, improving the accuracy of the image segmentation model.

[0096] In one implementation, the data processing device 120 is further configured to:

[0097] Obtain the shooting parameters corresponding to the lung CT image and the mask parameters corresponding to the lung image mask. The shooting parameters include the pixel height, pixel width, and the thickness of the transverse plane.

[0098] Calculate the lung parameters according to the pixel height, the pixel width, the thickness of the transverse plane, and the mask parameters.

[0099] For each lung CT image, after it is segmented and annotated by the image segmentation model, the pixels with the annotation value of 1 correspond to the left lung area, and the pixels with the annotation value of 2 correspond to the right lung area. Then, according to the annotation of each lung CT image, the corresponding mask parameters can be determined. The mask parameters are mainly the number of pixels corresponding to the mask, etc. Then, the lung parameters can be calculated according to the mask parameters and the shooting parameters.

[0100] The lung parameters can include parameters such as the left lung volume, the right lung volume, the left lung transverse plane area, and the right lung transverse plane area. For different lung parameters, their corresponding calculation methods are also different.

[0101] Specifically, the calculation formula for the left lung volume V l is:

[0102] V l = N l × h s × w s × s s

[0103] The volume V of the right lung r is calculated by the formula:

[0104] V r = N r × h s × w s × s s

[0105] where h s represents the height corresponding to each pixel, w s represents the width corresponding to each pixel, s s represents the thickness of the transverse plane, and N l represents the number of pixels in the left lung region, and N r represents the number of pixels in the right lung region.

[0106] The area A of the transverse plane of the left lung l is calculated by the formula:

[0107] A l = n l × h s × w s × s s

[0108] From the formula for the area A of the transverse plane of the lung r is:

[0109] A r = n r × h s × w s × s s

[0110] where n l represents the number of pixels in the left lung region, and n r represents the number of pixels in the right lung region.

[0111] Similarly, similar methods can also be used to calculate other lung parameters, such as the anteroposterior diameters of the left and right lungs, the transverse diameters of the left and right lungs, the transverse diameter of the chest, the volume of the trachea, etc.

[0112] In this embodiment, various parameters of the lung are calculated through the shooting parameters and mask parameters corresponding to the lung CT image, providing a data basis for subsequent lung development evaluation.

[0113] In one implementation, the model construction device 130 is further configured to:

[0114] Obtain the physiological information corresponding to the lung CT image, where the physiological information includes at least one of BMI index, height, weight, body surface area, gender, and number of days since birth;

[0115] Perform correlation matching between the lung parameters and the physiological information to obtain a matching result, and construct the lung development evaluation model according to the matching result.

[0116] Since the lung CT images are collected for the lungs of children, each lung CT image corresponds to a child, and the physiological information of each child may be different. Therefore, after obtaining the lung parameters in each lung CT image, it is also necessary to perform multi-dimensional physiological data integration in combination with the physiological information of the child corresponding to the lung CT image.

[0117] Specifically, perform correlation matching between parameters such as the BMI index, height, weight, body surface area, gender, and number of days since birth of children and the lung parameters, and establish a corresponding comprehensive database. The database records the lung CT images corresponding to each child collected, the various lung parameters obtained through the second calculation, such as the left and right lung volumes, the left and right lung areas on each transverse plane, the thoracic volume, the anteroposterior and superior-inferior diameters of the left and right lungs, etc., as well as information such as the BMI index, height, weight, body surface area, gender, and number of days since birth. When accessing the database, through the identity ID of the child, such as the sample number corresponding to the database, the above-mentioned information parameters can be freely retrieved and extracted.

[0118] A corresponding lung development evaluation model can also be established based on this comprehensive database, and the lung development of children can be evaluated through the lung development evaluation model.

[0119] For example, use statistical methods to model the comprehensive data, draw the curves of children's lung development changes with age and physical development, and quantify the lung development characteristics at each stage.

[0120] Based on the comprehensive database of children's lungs collected, taking the number of days since birth as the benchmark, divide the number of days since children's birth into multiple time periods, such as within one month after birth, between one month and two months, between two months and three months, between three months and six months, between six months and one year, between one year and two years, between two years and three years, between three years and four years, and so on until between eleven years and twelve years. Statistically calculate the average value and variance of the left and right lung volumes of children in different birth day periods to obtain the curve of the development change of children's lung volume.

[0121] Refer to Figure 6 , Figure 6 which is a schematic diagram of the development change curve of children's lung volume provided in this embodiment.

[0122] This curve can be used to reveal the variation law of the lung volume of normal children with the increase of the number of days after birth. Further, by combining information such as gender and region, the variation curves of the lung volume development of children of different genders and regions with the increase of the number of days after birth can be refined, and a more refined normal lung development model for children can be constructed.

[0123] In this embodiment, the lung parameters and physiological parameters corresponding to multiple children's lung CT images are used as the standard parameters or normal parameters for lung development evaluation, and then a corresponding comprehensive database and a lung development evaluation model are constructed using these data, so as to achieve an accurate evaluation of the subsequent lung CT images to be evaluated.

[0124] In one implementation manner, the development evaluation device 140 is further configured to:

[0125] Segment the lung CT image to be evaluated using the image segmentation model to obtain a lung image mask to be evaluated;

[0126] Obtain the physiological information to be evaluated corresponding to the lung CT image to be evaluated, and input the lung image mask to be evaluated and the physiological information to be evaluated into the lung development evaluation model to obtain the evaluation result.

[0127] Specifically, by inputting the children's lung CT image and the corresponding physiological information through the development evaluation device 140, and comparing the lung CT image and the physiological information, it can be determined whether the lung parameters of the children are normal, and the corresponding evaluation result is output.

[0128] Refer to Figure 7 , Figure 7 which is a schematic flowchart of a lung development evaluation method based on CT images provided in this embodiment. The method includes:

[0129] S701. Segment and label the lung CT image sample using the image segmentation model to obtain a lung image mask.

[0130] S702. Obtain the shooting parameters corresponding to the lung CT image, and calculate the lung parameters corresponding to the lung CT image according to the shooting parameters and the lung image mask.

[0131] S703. Obtain the physiological information corresponding to the lung CT image, and construct a lung development evaluation model according to the lung parameters and the physiological information.

[0132] S704. Use the image segmentation model and the lung development evaluation model to evaluate the lung CT image to be evaluated to obtain an evaluation result.

[0133] It can be understood that the method of this embodiment corresponds to the CT image-based lung development evaluation system 100 in the above embodiment. The optional items in the above embodiment are also applicable to this embodiment, so they will not be described repeatedly here.

[0134] The present invention also provides a computer device. Exemplarily, the computer device includes a processor and a memory. Among them, the memory stores a computer program, and the processor runs the computer program to enable the computer device to execute the above CT image-based lung development evaluation method or the functions of each module in the above CT image-based lung development evaluation system 100.

[0135] Among them, the processor can be an integrated circuit chip with signal processing capabilities. The processor can be a general-purpose processor, including at least one of a central processing unit (CPU), a graphics processing unit (GPU), a network processor (NP), a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, and discrete hardware components. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc., which can implement or execute the various methods, steps, and logic block diagrams disclosed in the embodiments of the present invention.

[0136] The memory can be, but is not limited to, a random access memory (RAM), a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), etc. Among them, the memory is used to store the computer program, and after receiving the execution instruction, the processor can execute the computer program accordingly.

[0137] The present invention also provides a computer storage medium for storing the computer program used in the above computer device. Wherein, the computer storage medium can be a readable storage medium, a non-volatile storage medium or a volatile storage medium. For example, the computer storage medium may include, but is not limited to: USB flash drives, mobile hard disks, read-only memories (ROMs), random access memories (RAMs), magnetic disks or optical discs, etc., which are all media capable of storing program codes.

[0138] In several embodiments provided by the present invention, it should be understood that the disclosed apparatus and method can also be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the flowcharts and structure diagrams in the accompanying drawings show the possible architectures, functions, and operations of the apparatus, method, and computer program product according to multiple embodiments of the present invention. In this regard, each block in the flowchart or block diagram may represent a module, a program segment, or a part of code, and the module, program segment, or part of code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in alternative implementations, the functions marked in the blocks may occur in a different order than marked in the accompanying drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the structure diagram and / or flowchart, as well as the combination of blocks in the structure diagram and / or flowchart, can be implemented by a dedicated hardware-based system for performing the specified functions or actions, or can be implemented by a combination of dedicated hardware and computer instructions.

[0139] In addition, in each embodiment of the present invention, the various functional modules or units may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.

[0140] If the described functions are implemented in the form of software functional modules and sold or used as an independent product, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a smart phone, a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention.

[0141] The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily conceive of changes or substitutions, which should all be covered within the protection scope of the present invention.

Claims

1. A lung development evaluation system based on CT images, characterized in that, Including: An image segmentation device, a data processing device, a model construction device, and a development evaluation device. The image segmentation device is communicatively connected to the data processing device, the data processing device is communicatively connected to the model construction device, and the model construction device is communicatively connected to the development evaluation device; The image segmentation device is configured to segment and label a lung CT image sample using an image segmentation model to obtain a lung image mask, and send the lung image mask to the data processing device; The data processing device is configured to obtain the shooting parameters corresponding to the lung CT image, calculate the lung parameters corresponding to the lung CT image according to the shooting parameters and the lung image mask, and send the lung parameters to the model construction device; The model construction device is configured to obtain the physiological information corresponding to the lung CT image, and construct a lung development evaluation model according to the lung parameters and the physiological information; The development evaluation device is configured to process the lung CT image to be evaluated using the image segmentation model and the lung development evaluation model to obtain an evaluation result.

2. The lung development evaluation system based on CT images according to claim 1, wherein The evaluation system further includes a model training device, and the model training device is configured to perform the following training steps: Step 1: Obtain multiple lung CT images, crop each lung CT image to a preset size and perform annotation to obtain an annotated image; Step 2: Input the annotated image into an initial segmentation model, and obtain the output result of the initial segmentation model; Step 3: Calculate the loss value between the actual mask corresponding to the annotated image and the output result using a loss function; Step 4: If the loss value is greater than or equal to the loss value threshold, optimize the parameters of the initial segmentation model to obtain an intermediate segmentation model; Repeat steps 2 to 4 until the loss value is less than the loss value threshold, and use the intermediate segmentation model corresponding to the loss value less than the loss value threshold as the image segmentation model.

3. The lung development evaluation system based on CT images according to claim 2, wherein The model training device is further configured to construct the loss function; The loss function is: Among them, the labeled image includes N samples, i represents the i-th sample, C represents the type of mask included in each sample, each pixel in the sample corresponds to a mask type, and x ic represents the score corresponding to each mask type in the i-th sample.

4. The lung development evaluation system based on CT images according to claim 1, wherein, The data processing device is further configured to: Obtain the shooting parameters corresponding to the lung CT image and the mask parameters corresponding to the lung image mask, where the shooting parameters include the pixel height, pixel width, and thickness of the transverse plane; Calculate the lung parameters according to the pixel height, pixel width, thickness of the transverse plane, and the mask parameters.

5. The lung development evaluation system based on CT images according to claim 4, wherein The lung parameters include the left lung volume and the right lung volume; The left lung volume V l is calculated by the formula: V l = N l × h s × w s × s s The volume V of the right lung r is calculated by the following formula: V r = N r × h s × w s × s s Among them, h s represents the height corresponding to each pixel, w s represents the width corresponding to each pixel, s s represents the thickness of the transverse bit plane, N l represents the number of pixels in the left lung region, N r represents the number of pixels in the right lung region.

6. The lung development evaluation system based on CT images according to claim 1, wherein, The model construction device is further configured to: Obtain the physiological information corresponding to the lung CT image, where the physiological information includes at least one of the BMI index, height, weight, body surface area, gender, and number of days since birth; Perform association matching on the lung parameters and the physiological information to obtain a matching result, and construct the lung development evaluation model according to the matching result.

7. The lung development evaluation system based on CT images according to claim 6, characterized in that, The development evaluation device is further configured to: Segment the lung CT image to be evaluated using the image segmentation model to obtain a lung image mask to be evaluated; Obtain the physiological information to be evaluated corresponding to the lung CT image to be evaluated, and input the lung image mask to be evaluated and the physiological information to be evaluated into the lung development evaluation model to obtain the evaluation result.

8. A method for evaluating lung development based on CT images, characterized in that, Including: Use an image segmentation model to segment and label the lung CT image sample to obtain a lung image mask; Obtain the shooting parameters corresponding to the lung CT image, and calculate the lung parameters corresponding to the lung CT image according to the shooting parameters and the lung image mask; Obtain the physiological information corresponding to the lung CT image, and construct a lung development evaluation model according to the lung parameters and the physiological information; Use the image segmentation model and the lung development evaluation model to evaluate the lung CT image to be evaluated to obtain an evaluation result.

9. A computer device, characterized in that, The computer device includes a processor and a memory, the memory stores a computer program, and the processor is configured to execute the computer program to implement the lung development evaluation method based on CT images according to claim 8.

10. A computer storage medium, characterized in that, It stores a computer program, and when the computer program is executed on a processor, it implements the lung development evaluation method based on CT images according to claim 8.

Images