Ophthalmic disease onset risk prediction method and system

Abstract

The invention relates to the technical field of ophthalmic disease diagnosis, in particular to an ophthalmic disease onset risk prediction method and system. Through combination of multiple groups of indexes and image data, the onset condition of the ophthalmic disease can be effectively evaluated, score data reflecting the onset state of the disease can be obtained, and the onset probability and the staging result degree of the ophthalmic disease can be more intuitively determined through the score data. Therefore, the overall automatic prediction of the onset of ophthalmic diseases, especially thyroid ophthalmic diseases, is realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of ophthalmic disease diagnosis, in particular to an ophthalmic disease onset risk prediction method and system. BACKGROUND

[0002] Thyroid-associated ophthalmopathy, also known as Graves ophthalmopathy, is an autoimmune disease related to hyperthyroidism, affecting the soft tissues of the orbit. The disease accounts for up to 20% of adult orbital diseases and is the main cause of exophthalmos and diplopia in adults.

[0003] Early diagnosis and treatment of Graves disease is very important, because timely correction of thyroid dysfunction can reduce the risk of patients developing or mildly progressing, and the longer the duration of the disease, the more difficult it is for patients with moderate to severe to benefit from immunosuppressive therapy. Currently, the clinical activity score (CAS) and EUGOGO classification are mainly used in clinical practice. The CAS score is easy to operate and easy to promote, but it is difficult to distinguish whether the changes in the eye are caused by inflammation or congestion, and junior doctors are prone to errors in actual operation. It is difficult to accurately judge the severity, but the EUGOGO classification can classify patients into mild, moderate to severe, and vision-threatening categories according to different combinations of symptoms. In clinical diagnosis and treatment, the use of CAS and EUGOGO classification is difficult to avoid a certain subjectivity and is easily influenced by the experience of clinicians, so it cannot ensure consistency and accuracy. SUMMARY

[0004] In order to solve the technical problems existing in the prior art, the embodiments of the present application provide an auxiliary diagnosis method and system for combining computer medical images and multiple clinical indicators in the diagnosis of ophthalmic diseases. The features extracted from the images are automatically evaluated based on a prediction model by feature extraction of the images.

[0005] In order to achieve the above purpose, the technical solutions adopted by the embodiments of the present application are as follows:

[0006] In a first aspect, an ophthalmic disease onset risk prediction method is provided, which comprises: obtaining clinical data of a patient to be diagnosed and image set data of a target region of the patient to be diagnosed; the image set data is fat fraction, water value, anisotropy fraction and apparent diffusion coefficient obtained based on MRI imaging, and the clinical data includes age, total cholesterol value, free triiodothyronine and neutrophil / lymphocyte ratio; the clinical data and the image set data are input into a risk prediction model to calculate a risk value; the risk prediction model is a nomogram model.

[0007] Furthermore, obtaining the fat fraction includes: obtaining an image obtained by scanning the orbital tissue under coronal mDixon-quant sequence conditions, and segmenting the left and right orbital fat in the image, and obtaining a first thickness corresponding to the left orbital fat and a second thickness corresponding to the right orbital fat based on the segmentation result, and determining the average of the first thickness and the second thickness as the fat fraction.

[0008] Furthermore, obtaining the water value includes: obtaining an image obtained by scanning the orbital tissue under the coronal mDixon-quant sequence conditions, and extracting the image through a water-fat separation algorithm to obtain a pure water signal image of the left orbit and a pure water signal image of the right orbit, obtaining a corresponding first water value and a second water value based on the pure water signal image of the left orbit and the pure water signal image of the right orbit, and determining the average value of the first water value and the second water value as the water value.

[0009] Furthermore, obtaining the anisotropy fraction includes: obtaining an MRI image corresponding to the orbital tissue, and obtaining an anisotropy fraction map obtained by scanning the orbital tissue under coronal DTI sequence conditions, segmenting the extraocular muscle-related area to obtain an extraocular segmentation mask, and superimposing the extraocular segmentation mask on the anisotropy fraction map, directly extracting the anisotropy fraction values ​​of all voxels in the mask area and the average anisotropy fraction as the final value.

[0010] Furthermore, the apparent diffusion coefficient is obtained, including: obtaining an MRI image corresponding to the orbital tissue, and obtaining an apparent diffusion coefficient map obtained by scanning the orbital tissue under coronal DWI sequence conditions, segmenting the extraocular muscle-related area to obtain an extraocular segmentation mask, and superimposing the extraocular segmentation mask on the apparent diffusion coefficient map, directly extracting the apparent diffusion coefficient values ​​of all voxels in the mask area and the average apparent diffusion coefficient value as the final value.

[0011] Furthermore, the clinical data and the imaging group data are input into a risk prediction model to calculate the risk value, including: obtaining the first score, second score, third score, fourth score and fifth score corresponding to the age, total cholesterol value, free triiodothyronine, neutrophil / lymphocyte ratio and the imaging group data based on the nomogram model, as well as the corresponding total score.

[0012] Furthermore, obtaining the fifth score corresponding to the image group data includes: obtaining sub-scores corresponding to the fat fraction, water value, anisotropy fraction and apparent diffusion coefficient respectively, and updating the sub-scores according to the calculation weights corresponding to the above-mentioned image group data and weighting them to obtain the fifth score.

[0013] In a second aspect, a system for predicting the risk of developing ophthalmic diseases is provided, the system comprising: a data acquisition unit for acquiring clinical data of a patient to be diagnosed and imaging group data of a target area of ​​the patient to be diagnosed; a risk prediction unit for inputting the clinical data and the imaging group data into a risk prediction model to calculate a risk value; the risk prediction model is a nomogram model.

[0014] Furthermore, the risk prediction module includes a first score calculation unit, a second score calculation unit, a third score calculation unit, a fourth score calculation unit and a fifth score calculation unit.

[0015] Furthermore, the first score calculation unit is used to obtain a first score corresponding to age; the second score calculation unit is used to obtain a second score corresponding to the total cholesterol value; the third score calculation unit is used to obtain a third score corresponding to the free triiodothyronine; the fourth score calculation unit is used to obtain a fourth score corresponding to the neutrophil / lymphocyte ratio; and the fifth score calculation unit is used to obtain a fifth score corresponding to the imaging group data.

[0016] In the technical solution provided in the embodiments of the present application, by combining multiple groups of indicators and imaging data, ophthalmic diseases can be effectively predicted and evaluated, and score data reflecting the onset status of ophthalmic diseases can be obtained. The probability of onset of ophthalmic diseases, especially thyroid eye diseases, can be determined more intuitively through this score data, thereby realizing overall automated prediction of the onset of ophthalmic diseases, especially thyroid eye diseases. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0018] The methods, systems, and / or programs in the accompanying drawings will be further described according to exemplary embodiments. These exemplary embodiments will be described in detail with reference to the drawings. These exemplary embodiments are non-limiting exemplary embodiments, wherein example numerals represent similar structures in the various views of the drawings.

[0019] Figure 1 This is a flow chart of a method for predicting the onset of ophthalmic diseases provided in an embodiment of the present application.

[0020] Figure 2 It is a schematic diagram of the nomogram model in the embodiment of the present application.

[0021] Figure 3Fig. 1 is a schematic diagram of a system structure according to an embodiment of the present application.

[0022] Figure 4 Fig. 2 is a schematic diagram of a terminal device structure according to an embodiment of the present application. DETAILED DESCRIPTION

[0023] In order to better understand the above technical solutions, the technical solutions of the present application will be described in detail below through the accompanying drawings and specific embodiments. It should be understood that the specific features in the embodiments of the present application and the specific embodiments are detailed descriptions of the technical solutions of the present application, and are not limitations of the technical solutions of the present application. In the case of no conflict, the technical features in the embodiments of the present application and the specific embodiments can be combined with each other.

[0024] In the following detailed description, many specific details are set forth in order to provide a thorough understanding of the relevant teachings. However, it will be apparent to one skilled in the art that the present application can be practiced without these details. In other instances, well-known methods, procedures, systems, components, and / or circuits have been described at a relatively high level, without detail, in order to avoid unnecessarily obscuring aspects of the present application.

[0025] The flowcharts in the present application illustrate the execution processes performed by the system according to the embodiments of the present application. It should be clearly understood that the execution processes of the flowcharts can not be executed in sequence. On the contrary, these execution processes can be executed in reverse order or simultaneously. In addition, at least one other execution process can be added to the flowchart. One or more execution processes can be deleted from the flowchart.

[0026] Before the embodiments of the present application are further described in detail, the terms and phrases involved in the embodiments of the present application are explained, and the terms and phrases involved in the embodiments of the present application are applicable to the following explanations.

[0027] (1) In response to, used to indicate the condition or state on which the executed operation depends, when the dependent condition or state is met, one or more operations executed can be real-time or have a set delay; in the absence of special instructions, there is no restriction on the execution order of multiple operations executed.

[0028] (2) Based on, used to indicate the condition or state on which the executed operation depends, when the dependent condition or state is met, one or more operations executed can be real-time or have a set delay; in the absence of special instructions, there is no restriction on the execution order of multiple operations executed.

[0029] An embodiment of the present application provides a method for predicting the risk of developing ophthalmic diseases, which comprises performing magnetic resonance imaging on a target area of ​​a patient to obtain an MRI image, extracting and screening image features in the MRI image to obtain a plurality of image feature data, and processing the plurality of image feature data and a plurality of clinical data based on a joint scoring model to obtain a risk prediction score for the risk of developing ophthalmic diseases, especially thyroid eye disease, and in particular for the onset of ophthalmic diseases.

[0030] This embodiment provides an automated method for predicting the risk of ophthalmic diseases, which uses MRI images and multiple clinical indicators to establish a joint evaluation model. Based on this joint evaluation model, ophthalmic diseases are predicted and their corresponding severity is judged, with a high detection rate and accuracy.

[0031] For more information on this forecasting method, see Figure 1 , including the following steps:

[0032] Step S11: Acquire clinical data of the patient to be diagnosed and image data of the target area of ​​the patient to be diagnosed.

[0033] In this embodiment, this method adopts a combined prediction method to predict and judge the degree of ophthalmic diseases, especially thyroid eye disease. The logic of the combined prediction method is to obtain the score corresponding to each indicator based on the nomogram through the prediction model, and then obtain the total score. The total score is used to determine the possibility and degree of thyroid eye disease.

[0034] The indicators in this embodiment include clinical data and MRI imaging data. The imaging data includes fat fraction, water value, fractional anisotropy, and apparent diffusion coefficient obtained based on MRI imaging, and the clinical data includes age, total cholesterol value, free triiodothyronine, and neutrophil / lymphocyte ratio.

[0035] In this embodiment, the fat fraction is obtained by first segmenting the target area image, and then determining the fat fraction based on the segmented image. Specifically, an image of the orbital tissue scanned using a coronal mDixon-quant sequence is acquired, and the left and right orbital fat in the image are segmented. Based on the segmentation results, a first thickness corresponding to the left orbital fat and a second thickness corresponding to the right orbital fat are obtained. The average of the first and second thicknesses is then determined as the fat fraction.

[0036] In this embodiment, computer technology is used for segmentation, and the segmentation of the regional image is achieved through automated segmentation means.

[0037] Specifically, in the embodiment, a segmentation model is configured, and segmentation of a tumor region in an MRI image is completed based on the segmentation model. The segmentation model is constructed based on a U-Net network, and includes an encoder and a decoder, and the encoder and the decoder are connected in a skip connection manner. A cavity convolution module is configured in the encoder module to improve a receptive field and capture extensive context information, and a multi-scale fusion module is configured in the decoder to integrate feature information at different levels.

[0038] Specifically, the encoder in the embodiment adopts a VGG16 network as a backbone network, and a cavity convolution with different cavity rates is configured at the last convolution layer of the encoder. The feature extraction process of the encoder is specifically as follows: an image with a size of 512*512*3 is subjected to two rounds of 3*3 convolution processing with 64 channels to generate a preliminary feature map with a size of 512*512*64. Then, the feature map is reduced to 256*256*64 through a 2*2 maximum pooling layer. Then, two rounds of 3*3 convolution with 128 channels are performed to further refine the features to obtain a feature map with a size of 256*256*128, which is halved to 128*128*128 after 2*2 maximum pooling. Then, three rounds of 3*3 convolution with 256 channels are performed to expand the feature representation to form a feature map with a size of 128*128*256, and then 2*2 maximum pooling is performed to reduce the dimension to 64*64*256. The same process continues, three rounds of 3*3 convolution with 512 channels are performed to output a feature map with a size of 64*64*512, and then 2*2 maximum pooling is performed to obtain a feature map with a size of 32*32*512, and then three rounds of 3*3 convolution with 512 channels are performed to obtain a feature map with a size of 32*32*512, which is then input into an ASPP module, in which a 1*1 convolution is performed for dimension adjustment. In parallel, three 3*3 convolution layers with different expansion rates (6, 12, and 18) capture spatial information at different scales, and a global average pooling layer is used to extract global context information. The outputs of these parallel steps are unified in size and fused in the depth direction. Finally, a comprehensive feature map with a size of 32*32*512 is output after 1*1 convolution processing, which integrates rich local and global feature information.

[0039] The decoder adopts deconvolution for up-sampling, and the structure of the decoder is also configured with a fusion feature pyramid network in the embodiment, which gradually up-samples and transversely connects each layer in the decoder from the low end to the top end to realize feature fusion at different scales.

[0040] And, in the embodiment, shallow detail information captured in the encoder can be directly fused with deep semantic information in the decoder by introducing the skip connection. Moreover, rich information support is provided for the generation of the semantic segmentation map, and the performance of image semantic segmentation is significantly improved. The attention mechanism is added in the skip connection, which can make the semantic information fusion better. The channel attention mechanism (SE) is added in each skip connection part. After two convolution operations in the first layer, the output picture size is 512x512x64. Then in the skip connection part, feature conversion is performed first. This step does not change the number of channels, and Ftr maintains the dimension. The output dimension will still be 512x512x64. Next, compression operation is performed. Global average pooling compresses each 512x512 feature map into a single value, which provides a feature description for each channel. The result is a 1x1x64 tensor. Then, the excitation operation is performed, that is, two fully connected layers transform the features. Usually, the first fully connected layer reduces the dimension (from 64 to 16), and the second fully connected layer restores the dimension (from 16 to 64). After the activation function, a 1x1x64 tensor is obtained, and each channel has an excitation coefficient. Finally, the Scale realizes the rescaling, and each input channel is multiplied by its corresponding excitation coefficient. The final output size is 512x512x64 image, which is then fused with the image in the last layer. This operation enables the network to sort the importance of each channel of the input data, and while maintaining the original spatial structure, it adjusts the weight distribution between channels to achieve high-precision segmentation of endometrial cancer MRI images.

[0041] For the water value, first, the image obtained by scanning the orbital tissue under the condition of the coronal mDixon-quant sequence is obtained, and the left and right orbital pure water signal images are extracted from the image by water-fat separation algorithm. The first water value and the second water value are obtained based on the left and right orbital pure water signal images, and the average of the first water value and the second water value is determined as the water value.

[0042] For the anisotropy fraction, first, the MRI image corresponding to the orbital tissue is obtained, and the anisotropy fraction image obtained by scanning the orbital tissue under the condition of the coronal DTI sequence is obtained. The extraocular muscle related region is segmented to obtain an extraocular segmentation mask, and the extraocular segmentation mask is superimposed on the anisotropy fraction image to directly extract the anisotropy fraction value of all voxels in the mask region and the average anisotropy fraction as the final value.

[0043] For the diffusion coefficient, the apparent diffusion coefficient map obtained by scanning the orbital tissue under the condition of coronal DWI sequence is used to segment the extraocular muscle related region to obtain an extraocular segmentation mask, and the extraocular segmentation mask is superimposed on the apparent diffusion coefficient map to directly extract the apparent diffusion coefficient values of all voxels in the mask region and the average apparent diffusion coefficient value as the final value.

[0044] Step S12. Input the clinical data and the image data into the risk prediction model to calculate a risk value.

[0045] In this embodiment, the risk prediction model is a nomogram model. The nomogram model obtains a total score based on the scores corresponding to each index and image data obtained by mapping the multiple clinical data and image data obtained in step S11. The total score is the risk value in this embodiment, which is used to represent the classification and degree of the risk of ophthalmic diseases.

[0046] Specifically, the first score, the second score, the third score, the fourth score, and the fifth score corresponding to the age, the total cholesterol value (TC), the free triiodothyronine (FT3), the neutrophil / lymphocyte ratio (NLR), and the image data are obtained, as well as the corresponding total score.

[0047] The calculation result of the nomogram model can be referred to Figure 2 , wherein radsocre is the overall calculation score of the image data, i.e., the sixth score value. The calculation of the radsocre score is based on the following formula: Radscore = 0.562*fat fraction + 0.477*water value + 0.134*anisotropy fraction + 0.493*apparent diffusion coefficient.

[0048] The specific calculation process of the nomogram in this embodiment can be directly referred to Figure 2 As shown in the figure, the calculation logic of the nomogram can be implemented, and in this embodiment, it will not be described again.

[0049] In summary, in this embodiment, by combining the above multiple indexes, the possibility and degree of the onset of ophthalmic diseases can be effectively evaluated in the early stage, and the score data reflecting the risk of ophthalmic diseases can be obtained. The probability and degree of the onset of ophthalmic diseases can be more directly determined through the score data, thereby realizing the overall automatic prediction of the onset of ophthalmic diseases.

[0050] Referring to Figure 3 , for the method provided in steps S11-S12, in this embodiment, a prediction system 30 is also provided, which comprises:

[0051] The data acquisition unit 31 is configured to acquire clinical data of a patient to be diagnosed and image data of a target region of the patient to be diagnosed.

[0052] The risk prediction unit 32 is configured to input the clinical data and the image data into a risk prediction model to calculate a risk value.

[0053] In some embodiments, the risk prediction model comprises a first score calculation unit, a second score calculation unit, a third score calculation unit, a fourth score calculation unit and a fifth score calculation unit.

[0054] In some embodiments, the first score calculation unit is configured to acquire a first score corresponding to an age, the second score calculation unit is configured to acquire a second score corresponding to a total cholesterol value, the third score calculation unit is configured to acquire a third score corresponding to a free triiodothyronine value, the fourth score calculation unit is configured to acquire a fourth score corresponding to a neutrophil / lymphocyte ratio, and the fifth score calculation unit is configured to acquire a fifth score corresponding to the image data.

[0055] In some embodiments, the risk prediction model is a nomogram model.

[0056] Referring to Figure 4 The above method can also be integrated into the terminal device 40 provided in the present disclosure. Due to different configurations or performances, the terminal device can include one or more processors 401 and memories 402. The memories 402 can store one or more applications or data. The memories 402 can be temporary memories or persistent memories. The applications stored in the memories 402 can include one or more modules (not shown in the figure), and each module can include a series of computer executable instructions in the terminal device. Furthermore, the processor 401 can be configured to communicate with the memories 402, and execute a series of computer executable instructions in the memories 402. The terminal device can further include one or more power supplies 403, one or more wired or wireless network interfaces 404, one or more input / output interfaces 405, one or more keyboards 406, and the like.

[0057] In a specific embodiment, the terminal device includes a memory and one or more programs, wherein one or more programs are stored in the memory, and the one or more programs can include one or more modules, and each module can include a series of computer executable instructions in the terminal device, and the one or more processors are configured to execute the one or more programs, and the one or more programs include computer executable instructions for:

[0058] Obtaining clinical data of a patient to be diagnosed and image feature data of a target region of the patient to be diagnosed;

[0059] Inputting the clinical data and the image data into a risk prediction model to calculate a risk value.

[0060] The various components of the processor are described in detail as follows:

[0061] In this embodiment, the processor is an application specific integrated circuit (ASIC) or one or more integrated circuits configured to implement one or more embodiments of the present application, for example, one or more digital signal processors (DSPs), or one or more field programmable gate arrays (FPGAs).

[0062] Alternatively, the processor can perform various functions by running or executing software programs stored in the memory and calling data stored in the memory, such as executing the above Figure 1 method.

[0063] In a specific implementation, as an embodiment, the processor can include one or more microprocessors.

[0064] The memory is used to store software programs for implementing the scheme of the present application and is controlled by the processor to execute, and the specific implementation can refer to the above method embodiments, which will not be described here.

[0065] Alternatively, the memory may be a read-only memory (ROM) or other type of static storage device that can store static information and instructions, a random access memory (RAM) or other type of dynamic storage device that can store information and instructions, or an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical disc, laser disc, optical disc, digital versatile disc, Blu-ray disc, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto. The memory may be integrated with the processor or exist independently and be coupled to the processing unit through the interface circuit of the processor, and the embodiments of the present application do not specifically limit this.

[0066] It should be noted that the structure of the processor shown in this embodiment does not constitute a limitation on the device. The actual device may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently.

[0067] In addition, the technical effects of the processor can refer to the technical effects of the method described in the above method embodiment, and will not be repeated here.

[0068] It should be understood that the processor in the embodiments of the present application may be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field programmable gate arrays (FPGAs) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc.

[0069] It should also be understood that the memory in the embodiments of the present application can be a volatile memory or a nonvolatile memory, or can include both volatile and nonvolatile memory. Among them, the nonvolatile memory can be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically EPROM (EEPROM), or a flash memory. The volatile memory can be a random access memory (RAM) used as an external cache. By way of example, and not limitation, many forms of random access memory (RAM) are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchlink DRAM (SLDRAM), and direct rambus RAM (DR RAM).

[0070] The above-described embodiments can be implemented in part or in whole through software, hardware (e.g., circuitry), firmware, or any combination thereof. When implemented in software, the above-described embodiments can be implemented in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When loaded and executed by a computer, the computer instructions or computer programs can produce the processes or functions described above in accordance with the embodiments of the present application. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable apparatus. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium, such as from a website, a computer, a server, or a data center to another website, computer, server, or data center through a wired (e.g., infrared, wireless, microwave, etc.) manner. The computer-readable storage medium can be any available medium or a collection of medium accessible by a computer or a data storage device such as a server, data center, etc. that contains one or more medium. The medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a magnetic tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid state disk.

[0071] In this application, "at least one" means one or more, and "multiple" means two or more. "At least one of the following (one)" or the like means any combination of the items, including a single item (one) or a combination of multiple items. For example, at least one of a, b, or c can mean a, b, c, a-b, a-c, b-c, or a-b-c, where a, b, and c can be single or multiple.

[0072] It should be understood that the size of the sequence number of the above-described processes in various embodiments of the present application does not mean the order of execution, and the execution order of the processes should be determined by their functions and inherent logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0073] Those of ordinary skill in the art can realize that the units and algorithm steps of the examples described in conjunction with the embodiments disclosed herein can be realized in electronic hardware, or a combination of computer software and electronic hardware. Whether the functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. A person skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0074] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0075] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0076] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0077] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0078] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or the part of the 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 enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0079] The above merely provides the specific implementation of the present application, but the protection scope of the present application is not limited to this. Any person skilled in the art can easily think of the changes or replacements within the technical range disclosed by the present application, which should be covered in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A method for predicting the risk of developing an ophthalmic disease, characterized in that: The method comprises: Acquiring clinical data of a patient to be diagnosed and imaging data of a target area of ​​the patient to be diagnosed; the imaging data includes fat fraction, water value, fractional anisotropy, and apparent diffusion coefficient obtained based on MRI imaging; and the clinical data includes age, total cholesterol value, free triiodothyronine, and neutrophil / lymphocyte ratio; The clinical data and the imaging data are input into a risk prediction model to calculate the risk value; the risk prediction model is a nomogram model.

2. The method for predicting the risk of ophthalmic diseases according to claim 1, wherein: Obtaining the fat fraction includes: obtaining an image obtained by scanning orbital tissue based on a coronal mDixon-quant sequence, segmenting the left and right orbital fat in the image, obtaining a first thickness corresponding to the left orbital fat and a second thickness corresponding to the right orbital fat based on the segmentation result, and determining an average of the first thickness and the second thickness as the fat fraction.

3. The method for predicting the risk of ophthalmic diseases according to claim 1, wherein: Obtaining the water value includes: obtaining an image obtained by scanning the orbital tissue under coronal mDixon-quant sequence conditions, and extracting the image through a water-fat separation algorithm to obtain a pure water signal image of the left orbit and a pure water signal image of the right orbit, obtaining a corresponding first water value and a second water value based on the pure water signal image of the left orbit and the pure water signal image of the right orbit, and determining the average of the first water value and the second water value as the water value.

4. The method for predicting the risk of ophthalmic diseases according to claim 1, wherein: Obtaining the anisotropy fraction includes: obtaining an MRI image corresponding to the orbital tissue, and obtaining an anisotropy fraction map obtained by scanning the orbital tissue under a coronal DTI sequence, segmenting the extraocular muscle-related area to obtain an extraocular segmentation mask, superimposing the extraocular segmentation mask on the anisotropy fraction map, and directly extracting the anisotropy fraction values ​​of all voxels in the mask area and the average anisotropy fraction as a final value.

5. The method for predicting the risk of ophthalmic diseases according to claim 1, wherein: Obtaining the apparent diffusion coefficient includes: obtaining an MRI image corresponding to the orbital tissue, and obtaining an apparent diffusion coefficient map obtained by scanning the orbital tissue under a coronal DWI sequence, segmenting the extraocular muscle-related area to obtain an extraocular segmentation mask, and superimposing the extraocular segmentation mask on the apparent diffusion coefficient map, directly extracting the apparent diffusion coefficient values ​​of all voxels in the mask area and the average apparent diffusion coefficient value as a final value.

6. The method for predicting the risk of ophthalmic diseases according to claim 1, wherein: The step of inputting the clinical data and the imaging group data into the risk prediction model to calculate the risk value includes: obtaining the first score, second score, third score, fourth score and fifth score corresponding to the age, total cholesterol value, free triiodothyronine, neutrophil / lymphocyte ratio and the imaging data based on the nomogram model, as well as the corresponding total score.

7. The method for predicting the risk of ophthalmic diseases according to claim 6, wherein: Obtaining the fifth score corresponding to the image group data includes: respectively obtaining sub-scores corresponding to the fat fraction, water value, anisotropy fraction, and apparent diffusion coefficient, and updating the sub-scores according to the calculation weights corresponding to the above-mentioned image group data and weighting them to obtain the fifth score.

8. An ophthalmic disease risk prediction system, characterized in that: The system comprises: a data acquisition unit, configured to acquire clinical data of a patient to be diagnosed and image data of a target area of ​​the patient to be diagnosed; The risk prediction unit is used to input the clinical data and the imaging data into a risk prediction model to calculate the risk value; the risk prediction model is a nomogram model.

9. The ophthalmic disease risk prediction system according to claim 8, characterized in that: The risk prediction module includes a first score calculation unit, a second score calculation unit, a third score calculation unit, a fourth score calculation unit, and a fifth score calculation unit.

10. The ophthalmic disease risk prediction system according to claim 9, characterized in that: The first score calculation unit is used to obtain a first score corresponding to the age; the second score calculation unit is used to obtain a second score corresponding to the total cholesterol value; The third score calculation unit is used to obtain a third score corresponding to the free triiodothyronine; The fourth score calculation unit is used to obtain a fourth score corresponding to the neutrophil / lymphocyte ratio; The fifth score calculation unit is configured to obtain a fifth score corresponding to the image data.

Images