Establishment method, equipment and program product of eyeball indentation lesion quantity evaluation model

Abstract

The invention belongs to the field of intelligent medical treatment, and particularly relates to a construction method and device of an eyeball retraction lesion quantity evaluation model and a program product. The method comprises the following steps: acquiring an image set of an eyeball indentation patient; marking the image set to obtain orbital cavity tissue area boundaries of the uninjured side and the affected side, wherein orbital cavity tissue comprises eyeballs, orbital fat and any one or more of the following components: extraocular muscle, optic nerve, cornea and lacrimal gland; on the basis of the orbital cavity tissue region boundaries, orbital cavity tissue volumes of the uninjured side and the affected side are extracted respectively, and orbital cavity tissue volume variations of the affected side and the uninjured side are calculated; and inputting the image set as a training data set and the orbital cavity tissue volume variation as a prediction target into a neural network model for training to obtain an eyeball retraction lesion volume evaluation model. According to the method, the volume change of the finely divided orbital cavity tissue is considered, and meanwhile, the relative volume change is synthesized, so that the quantification of the lesion quantity of the eyeball retraction is more accurate, and a better correction effect can be achieved after an operation.

Description

Technical Field

[0001] The present invention relates to the field of intelligent medicine, and more specifically, to a method, device, medium, and program product for constructing an evaluation model of enophthalmos variable amount. Background Art

[0002] Post-traumatic Enophthalmos (PE) refers to the condition where, due to injuries, orbital tumor resection / radiotherapy, etc., partial tissue necrosis and adhesion occur in the orbital cavity, resulting in the posterior and outward movement of the eyeball at the equator due to the loss of orbital content and / or the increase in orbital cavity volume, and the eyeball visually appears sunken and retracted. Currently, this condition is usually confirmed in ophthalmological evaluations through the following diagnostic criteria: (1) Clinical examination: observing the appearance and manually detecting the position of the eyeball; (2) Imaging examination: mainly evaluating the structural changes in the orbit and the relative position of the eyeball through CT or MRI scans; (3) Quantitative measurement: using an exophthalmometer to quantify the protrusion of the eyeball. A recession outside the normal value range is considered a depression. Generally, a recession of more than 2 mm compared to the contralateral eyeball is considered clinically significant.

[0003] Currently, the treatment of eyeball depression usually involves injecting fillers to correct the protrusion of the eyeball; at this time, a necessary technical problem is how to determine the filling amount of the filler.

[0004] In the prior art, CT imaging computer assistance is used to analyze the orbital cavity volume of patients with orbital bone fractures and calculate an empirical formula for orbital cavity volume filling and eyeball protrusion correction. Research shows that there is a direct correlation between orbital volume and enophthalmos, mainly based on the Orbital Volume Ratio (OVR), that is, the ratio of the volume of the damaged orbit to the volume of the normal orbit and the degree of enophthalmos (Post-traumatic Enophthalmos, PE), namely the empirical formula method: the more the OV increases, the more severe the PE. On average, an increase in OV of 1 cm 3 will cause an increase in PE of 0.80 mm. This situation mainly applies to eyeball depression caused by an increase in the bony orbital volume due to orbital fractures. At this time, the filling amount is determined based on the measured increase in the bony orbital volume, and an increase in the orbital cavity volume of 1 cm 3It will cause the exophthalmos to decrease by about 0.8 mm, and the amount of injection is determined by calculating the degree of increase in orbital cavity volume based on the preoperatively measured exophthalmos. However, scholars have also found the limitations of following empirical formulas: due to the large individual differences in enophthalmos, the variability among observers in clinical evaluation is greater than that in radiological evaluation, which has certain limitations and large variability in clinical evaluation. At the same time, the current technology relies on traditional formulas and is difficult to adapt to complex trauma situations (such as surgery, radiotherapy, etc.), and is not applicable to patients with both increased bony orbital cavity volume and atrophy of orbital soft tissue volume. Summary of the Invention

[0005] In view of the above problems, the present invention provides a method for constructing an evaluation model of enophthalmos volume change, which accurately extracts the volume of orbital cavity tissues affecting enophthalmos, and simultaneously considers the changes in orbital volume and orbital cavity tissue volume, so as to realize a more accurate quantification of the ideal filling volume of enophthalmos during the filling treatment of enophthalmos patients.

[0006] The present application (in the first aspect) discloses a method for constructing an evaluation model of enophthalmos lesion variables, including:

[0007] S1: Obtain an image set of enophthalmos patients, where the images in the image set include the healthy eye and the affected eye; S2: Label the image set to obtain the boundary of the orbital cavity tissue region on the healthy side and the affected side, and the orbital cavity tissue includes the eyeball, orbital fat, and any one or more of the following: extraocular muscles, optic nerve, cornea, lacrimal gland;

[0008] S3: Based on the boundary of the orbital cavity tissue region, respectively extract the volume of the orbital cavity tissue on the healthy side and the affected side, and calculate the change in the volume of the orbital cavity tissue on the affected side and the healthy side;

[0009] S4: Use the image set as a training data set, and use the change in the volume of the orbital cavity tissue as a prediction target to input into a neural network model for training to obtain an evaluation model of enophthalmos lesion volume.

[0010] Further, the extraocular muscles include one or more of the following: medial rectus muscle, lateral rectus muscle, superior rectus muscle, inferior rectus muscle, superior oblique muscle, inferior oblique muscle.

[0011] Further, the method further includes: performing three-dimensional reconstruction based on the image set and the labeled boundary of the orbital cavity tissue region to obtain the three-dimensional boundary of the orbital cavity tissue respectively; based on the three-dimensional boundary of the orbital cavity tissue, respectively extract the volume of the orbital cavity tissue on the healthy side and the affected side, and calculate the change in the volume of the orbital cavity tissue on the affected side and the healthy side.

[0012] Further, the method further includes: marking the orbital boundary during marking, and respectively extracting the orbital volume and the volume of the orbital cavity tissue on the healthy side and the affected side based on the orbital boundary and the boundary of the orbital cavity tissue region, and calculating the change in the orbital volume and the change in the volume of the orbital cavity tissue on the affected side and the healthy side;

[0013] Calculate the relative volume change of the affected side compared to the healthy side based on the orbital volume change and the orbital cavity tissue volume change;

[0014] Use the image set as the training data set and the relative volume change as the prediction target to input into the neural network model for training to obtain an evaluation model for the relative volume change of eye lesions.

[0015] Furthermore, the relative volume change is expressed as:

[0016] RVD = ΔV orbital -ΔV content

[0017] where RVD represents the relative volume change, and ΔV orbital represents the orbital volume change; ΔV content represents the orbital cavity tissue volume change,

[0018] ΔV orbital The calculation method is expressed as:

[0019] ΔV orbital = V affected -V healthy

[0020] where ΔV orbital represents the orbital volume change, V affected represents the orbital volume of the affected side, V healthy represents the orbital volume of the healthy side;

[0021] ΔV content The calculation method is expressed as:

[0022] ΔV content = V affected-content -V healthy-content

[0023] where ΔV content represents the orbital cavity tissue volume change, V affected represents the orbital cavity tissue volume of the affected side, V healthy represents the orbital cavity tissue volume of the healthy side.

[0024] The second aspect of this application discloses a construction system for an evaluation model of enophthalmos variables, including:

[0025] An acquisition module: used to acquire an image set of enophthalmos patients, and the images in the image set include the orbits of the healthy side and the affected side;

[0026] Annotation module: used to annotate the image set to obtain the boundary of the orbital cavity tissue regions on the healthy side and the affected side, where the orbital cavity tissue includes the eyeball, orbital fat, and any one or more of the following: extraocular muscles, optic nerve, cornea, lacrimal gland;

[0027] Disease variable calculation module: used to respectively extract the volumes of the orbital cavity tissues on the healthy side and the affected side based on the boundary of the orbital cavity tissue region, and calculate the change in volume of the orbital cavity tissues on the affected side and the healthy side;

[0028] Model training module: used to use the image set as the training data set, and the change in volume of the orbital cavity tissue as the prediction target to input into the neural network model for training to obtain an evaluation model for the change in volume of the orbital cavity tissue, and the evaluation model for the change in volume of the orbital cavity tissue is an evaluation model for the volume of enophthalmos lesions.

[0029] Meanwhile, the present application discloses an application method of the constructed model:

[0030] A method for calculating the disease variable in the eyeball, characterized in that the method includes:

[0031] Obtain the image of a patient with enophthalmos;

[0032] Input the image of the patient with enophthalmos into the evaluation model for the volume of enophthalmos lesions or the relative volume model for enophthalmos lesions described in any one of the above to obtain the volume of enophthalmos lesions of the patient.

[0033] A method for calculating the filling amount for the treatment of enophthalmos, characterized in that the method includes:

[0034] Obtain the enophthalmos disease variable of the patient based on the method for calculating the disease variable in the eyeball;

[0035] Determine the volume of the filling material during filling treatment based on the volume of the enophthalmos lesion and the properties of the filling material.

[0036] Furthermore, the method further includes: determining the volume of the filling material during filling treatment based on the change in volume of the enophthalmos.

[0037] The third aspect of the present application discloses a computer device, the device includes: a memory and a processor; the memory is used to store program instructions; the processor is used to call the program instructions, and when the program instructions are executed, it is used to execute any step of the method for constructing the evaluation model for the enophthalmos disease variable described above, or the step of the method for calculating the disease variable in the eyeball or the method for calculating the filling amount for the treatment of enophthalmos.

[0038] A fourth aspect of the present application discloses a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, any step of the method for constructing the evaluation model of the enophthalmos variable described above, or the step of the calculation method of the enophthalmos variable or the calculation method of the filling amount for enophthalmos treatment is realized.

[0039] A fifth aspect of the present application discloses a computer program product, including a computer program. When the computer program is executed by a processor, any step of the method for constructing the evaluation model of the volume change amount of enophthalmos described above, or the step of the calculation method of the volume of intraocular lesions or the calculation method of the filling amount for enophthalmos treatment is realized.

[0040] The present application has the following beneficial effects:

[0041] (1) In the past volume analysis of enophthalmos, the influence of the volumes of various orbital cavity tissues on enophthalmos was not considered. Through the fine annotation of the tissues affecting enophthalmos, the present application realizes the accurate quantification of the volume of orbital cavity tissues, providing an accurate measure of the filling amount for filling treatment surgery.

[0042] (2) By using 3D reconstruction technology to reconstruct the orbital cavity tissues based on fine annotation, the quantification of the volume of orbital cavity tissues is more accurate compared to image-delineated ROI.

[0043] (3) In the past treatment of enophthalmos, only the change in orbital volume was often concerned, and enophthalmos caused by the change in orbital volume was corrected by filling. The present application comprehensively considers the change in orbital volume and the change in the orbital cavity tissues within the orbit, measures the change in the relative volume of the orbit through two-way changes, and determines the filling amount during filling treatment surgery based on the change in relative volume, with more accurate quantification and better correction effect after surgery. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative efforts.

[0045] Figure 1 is a schematic flowchart of the method provided in the first aspect of the embodiment of the present invention;

[0046] Figure 2 is a schematic diagram of the program product provided in the second aspect of the embodiment of the present invention;

[0047] Figure 3 is a schematic diagram of the computer device provided in the embodiment of the present invention;

[0048] Figure 4 It is a schematic diagram of the architecture of an exemplary computing device provided by an embodiment of the present invention;

[0049] Figure 5 It is a schematic diagram of a storage medium provided by an embodiment of the present invention;

[0050] Figure 6A It is one of the schematic diagrams of the steps of a 3D annotation provided by an embodiment of the present invention;

[0051] Figure 6B It is the second schematic diagram of the steps of a 3D annotation provided by an embodiment of the present invention;

[0052] Figure 6C It is the third schematic diagram of the steps of a 3D annotation provided by an embodiment of the present invention;

[0053] Figure 6D It is a schematic diagram of the result of a 3D annotation provided by an embodiment of the present invention;

[0054] Figure 6E It is a schematic diagram of a reconstructed 3D image with annotations provided by an embodiment of the present invention;

[0055] Figure 7 It is a schematic diagram of annotating orbital cavity tissues provided by an embodiment of the present invention;

[0056] Figure 8 It is a schematic diagram of the sunken eyeball shown in an image provided by an embodiment of the present invention;

[0057] Figure 9 It is a schematic diagram of volume measurement and calculation provided by an embodiment of the present invention;

[0058] Figure 10 It is a schematic diagram of the healthy orbital cavity and the diseased orbital cavity after reconstruction provided by an embodiment of the present invention. Detailed implementation manners

[0059] In order to enable those skilled in the art to better understand the solution of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0060] In some of the processes described in the specification, claims, and above-mentioned drawings of the present invention, a plurality of operations appear in a specific order. However, it should be clearly understood that these operations may not be executed in the order in which they appear herein or may be executed in parallel. The serial numbers of the operations, such as S101, S102, etc., are only used to distinguish different operations, and the serial numbers themselves do not represent any execution order. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel. It should be noted that the descriptions such as "first" and "second" in this article are used to distinguish different messages, devices, modules, etc., and do not represent a sequence, nor do they limit that "first" and "second" are of different types.

[0061] The technical solutions in the embodiments of the present invention will be clearly and completely described below 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 the embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative efforts belong to the scope of protection of the present invention.

[0062] Figure 1 It is a schematic flowchart of a method for constructing an enophthalmos variable estimation model provided by an embodiment of the present invention. Specifically, the method includes the following steps:

[0063] S101: Obtain an image set of enophthalmos patients, where the images in the image set include the orbits of the healthy side and the affected side.

[0064] S102: Label the image set to obtain the boundaries of the orbital cavity tissue regions on the healthy side and the affected side. The orbital cavity tissue includes the eyeball, orbital fat, and any one or more of the following: extraocular muscles, optic nerve, cornea, lacrimal gland.

[0065] S103: Based on the boundaries of the orbital cavity tissue regions, extract the volumes of the orbital cavity tissues on the healthy side and the affected side respectively, and calculate the volume change amount of the orbital cavity tissues on the affected side and the healthy side.

[0066] S104: Use the image set as the training data set, and use the volume change amount of the orbital cavity tissue as the prediction target to input into the neural network model for training to obtain the constructed enophthalmos lesion volume evaluation model.

[0067] At present, there is no report on the application of accurately depicting and quantifying the volume of orbital soft tissues based on CT / MRI data in the treatment of enophthalmos deformity at home and abroad. The relevant applications of soft tissue quantification are limited to non-traumatic diseases, such as patients with thyroid-related ophthalmopathy. The characteristic of thyroid-associated orbitopathy (TAO) is exophthalmos caused by the increase in the volume of orbital fat and extraocular muscles.

[0068] Through clinical research on enophthalmos diseases, it is found in this application that for enophthalmos deformity, its core causes not only include the change in orbital volume caused by bony changes, but also include the change in the volume of soft tissues in the orbital cavity caused by trauma. Therefore, in order to further accurately evaluate the filling volume in filling treatment, the following important work is carried out in this application.

[0069] In some embodiments, previously, we innovatively manually delineated more than 20 important tissues in the orbital cavity (such as the eyeball, optic nerve, extraocular muscles, etc.) ( Figure 7 as shown), and have completed the annotation of the soft tissues in the orbital cavity of 26 cases (26 eyes) with enophthalmos to calculate the volume to be filled in the orbital cavity. Manual delineation for each case and each eye takes about 2 hours, which is time-consuming. However, through manual delineation, we have achieved accurate measurement of the volume of soft tissues in the orbital cavity, and the volume to be filled in the orbital cavity obtained is more in line with clinical practice.

[0070] The important tissues in the orbital cavity related to enophthalmos include one or more of the following: extraocular muscles (6 on each of the left and right sides, a total of 12), corneas (1 on each of the left and right sides, a total of 2), optic nerves (1 on each of the left and right sides, a total of 2), lacrimal glands (1 on each of the left and right sides, a total of 2), eyeballs (1 on each of the left and right sides, a total of 2), orbital fats (1 on each of the left and right sides, a total of 2), a total of 22 kinds.

[0071] Among them, the extraocular muscles include the medial rectus muscle, lateral rectus muscle, superior rectus muscle, inferior rectus muscle, superior oblique muscle, and inferior oblique muscle; there are 6 on each side, and a total of 12 on the left and right sides;

[0072] In some embodiments, the annotation includes: the eyeball, orbital fat.

[0073] In some embodiments, the annotation includes: the eyeball, orbital fat, cornea, optic nerve.

[0074] In some embodiments, the annotation includes: the eyeball, orbital fat, cornea, optic nerve, lacrimal gland, eyeball.

[0075] In some embodiments, the annotation includes: cornea, optic nerve, lacrimal gland, eyeball, orbital fat.

[0076] In some embodiments, the annotation includes: cornea, optic nerve, lacrimal gland, eyeball, orbital fat, extraocular muscle.

[0077] In some embodiments, the annotation method is as Figure 7 shown, and the orbital cavity tissues of the left eye and the right eye are annotated separately. There are 22 annotation objects in total, as Figure 7 and Table 1 show:

[0078] Table 1 Orbital cavity tissues

[0079] Orbital cavity tissue of the left eye Orbital cavity tissue of the right eye Left eye lens / eyeball Right eye lens / eyeball Left optic nerve Right optic nerve Left lacrimal gland Right lacrimal gland Left orbital fat Right orbital fat Left cornea Right cornea Left superior rectus muscle Right superior rectus muscle Left inferior rectus muscle Right inferior rectus muscle Left medial rectus muscle Right medial rectus muscle Left lateral rectus muscle Right lateral rectus muscle Left superior oblique muscle Right superior oblique muscle Left inferior oblique muscle Right inferior oblique muscle

[0080] As Figure 6A 、6B As shown in 6C, an annotation example is presented where we perform annotations on CT images in 3D reconstruction software. The annotation window is the upper-left window, showing the transverse view of the head image. The upper-right window is the 3D window, displaying the reconstructed 3D image and orbital cavity tissue after annotation. The lower-left window shows the coronal view of the image;

[0081] After the annotation is completed, as Figure 6D shown, the 3D image corresponding to the CT image is obtained through 3D reconstruction. Meanwhile, the annotation is reflected in the reconstructed 3D image as: the annotated 3D orbital cavity tissue, and by zooming in, the annotated reconstructed 3D image can be seen ( Figure 6E ).

[0082] In some embodiments, the volume to be filled in the orbital cavity is calculated based on the annotated reconstructed 3D image. The calculation of the volume to be filled only measures the difference in soft tissue volume between the affected side and the healthy side, and the calculation method is expressed as:

[0083] ΔV content = V affected-content - V healthy-content

[0084] where ΔV content represents the change in orbital cavity tissue volume, which is also the volume to be filled by us. V affected represents the volume of orbital cavity tissue on the affected side, and V healthy represents the volume of orbital cavity tissue on the healthy side;

[0085] In some embodiments, the calculation of the volume to be filled in the orbital cavity measures the difference in soft tissue volume between the affected side and the healthy side on the one hand, and also considers the difference in orbital volume caused by bony changes. At this time, the volume to be filled in the orbital cavity is the relative volume change, expressed as:

[0086] RVD = ΔV orbital - ΔV content

[0087] where RVD represents the relative volume change, and ΔV orbital represents the change in orbital volume; ΔV content represents the change in orbital cavity tissue volume, where

[0088] ΔV orbital The calculation method is expressed as:

[0089] ΔV orbital = V affected - V healthy

[0090] where ΔV orbital represents the change in orbital volume, and V affected represents the orbital volume on the affected side, Vhealthy represents the healthy orbital volume;

[0091] Due to the slow speed of the above manual tracing, in order to improve the efficiency of tracing and extracting the volume of orbital soft tissues, we use the image data of the completed manual tracing samples as the training set data, and the results of manual tracing as the output of the neural network, and train to obtain an automatic segmentation model for orbital soft tissues.

[0092] Then, we increase the sample size. First, we initially trace the orbital soft tissues through the automatic segmentation model of orbital soft tissues to improve efficiency, and then manually correct the tracing of orbital soft tissues. After reducing the tracing error, more volumes to be filled are extracted.

[0093] In some embodiments, in order to automatically extract the volume to be filled, we use the image data set as the training set data, and the change amount of the orbital tissue volume extracted as the label or training target of the deep learning model. After iterative training, a trained model for the change amount of orbital tissue volume is obtained. The model for evaluating the amount of enophthalmos disease variable we mentioned here is this trained model for the change amount of orbital tissue volume. This model for the change amount of orbital tissue volume can output the change amount of orbital tissue volume according to the images of newly input enophthalmos patients, so as to help doctors determine the filling amount during filling treatment.

[0094] In some embodiments, the volume to be filled during treatment is the change amount of orbital tissue volume.

[0095] In some embodiments, we use the image data set as the training set data, and the change amount of the orbital volume extracted as the label or training target of the deep learning model. After iterative training, a trained model for the change amount of orbital volume is obtained. The model for evaluating the amount of enophthalmos disease variable we mentioned here is this trained model for the change amount of orbital volume. This model for the change amount of orbital volume can output the change amount of orbital volume according to the images of newly input enophthalmos patients, so as to help doctors determine the filling amount during filling treatment.

[0096] In some embodiments, the volume to be filled during treatment is the diseased volume of enophthalmos, that is, the change amount of orbital volume.

[0097] Furthermore, doctors can respectively obtain the change amount of orbital tissue volume and the change amount of orbital volume of new patients to be tested according to the model for the change amount of orbital tissue volume and the model for the change amount of orbital volume, so as to obtain the relative change amount of orbital volume. Doctors determine the filling amount according to the relative change amount of orbital volume during filling treatment.

[0098] In some embodiments, the volume to be filled during treatment is the diseased volume of enophthalmos, that is, the relative change amount of orbital volume.

[0099] In some embodiments, we use the image dataset as the training set data, and use the relative orbital volume change obtained by extracting the training set data as the label or training target of the deep learning model. After iterative training, a trained relative orbital volume change model is obtained. The evaluation model for the enophthalmos variable we mentioned here is this trained relative orbital volume change model. This relative orbital volume change model can output the relative orbital volume change according to the newly input images of enophthalmos patients, so as to help doctors determine the filling amount during filling treatment.

[0100] In some embodiments, the intended filling amount during treatment is the lesion volume of enophthalmos, that is, the relative volume change.

[0101] In some embodiments, due to the different tolerances of patients to fillers and the differences in the treatment effects of fillers, the actual filling amount used for treatment is obtained after correcting according to the lesion volume output by the model during actual treatment.

[0102] In this application, 3D digital technology is used to quantitatively analyze the volume of orbital soft tissues of patients with post-traumatic enophthalmos deformity, so as to achieve precise correction of enophthalmos deformity.

[0103] This study is based on the cutting-edge technology of the combination of medicine and engineering, and innovatively applies an improved manual tracing and segmentation calculation method to quantitatively evaluate the pre-filled volume of the orbital cavity of patients with post-traumatic enophthalmos deformity, and then trains the model to precisely correct the enophthalmos deformity caused by complex traumas such as training injuries or war traumas. There are no relevant reports at home and abroad.

[0104] For patients with complex post-traumatic enophthalmos deformity with both orbital volume deficiency and increased orbital cavity volume, precise quantitative filling is achieved, and the treatment effect is predicted at the same time.

[0105] In some embodiments, the research methods include:

[0106] 1. Collecting cases: 80 patients (80 eyes) with post-traumatic enophthalmos deformity who needed autologous granular fat filling and were admitted to the Department of Orbital Diseases, Ophthalmology Department, the Third Medical Center of Chinese PLA General Hospital were included. Among them, for the experimental group, the expected filling volume was accurately quantified by digital manual tracing to obtain the 3D quantification, and for the control group, the expected filling volume was calculated using the classical formula.

[0107] 2. Project implementation:

[0108] Improve preoperative examinations (orbital CT, MRI, vision examination, exophthalmometry, physical examination),

[0109] Preliminarily determine the location and range of ocular adhesions, eye position, eye movement and strabismus degree, and evaluate the visual acuity and visual function of the affected eye.

[0110] 3. Tracer method: Before and after the operation, based on CT-based Dicom data, 24 orbital cavity tissue structures such as extraocular muscles, eyeballs, and optic nerves are manually traced and segmented using VECTRA M3 3D Slicer.

[0111] Preoperatively, quantitatively evaluate the volume of the soft tissue filling object in the orbital cavity.

[0112] 4. Reduce manual tracing errors: Three doctors with more than 10 years of clinical experience in orbital disease surgery perform manual tracing and take the average value of the volume (previously trained).

[0113] In some embodiments, a training set and the volume corresponding to the marked orbital cavity tissue structures are obtained through the above research, so as to calculate the change amount of the volume of the intraorbital tissue on the affected side;

[0114] The training set images are input into a convolutional neural network to predict the change amount of the volume of the intraorbital tissue on the affected side. The predicted value is compared with the measured and calculated value. With the aim of reducing the error between the predicted value and the measured and calculated value, for example, the loss function uses the MSE function, and continuously iterates and optimizes until a certain number of times or the error is less than the expected value, then stops iterating to obtain an eyeball enophthalmos filling volume evaluation model. For new eyeball enophthalmos patients, the images of their eyes are input into the eyeball enophthalmos filling volume evaluation model, and the predicted change amount of the volume of the intraorbital tissue on the affected side is output; in the filling treatment of patients with orbital enophthalmos, the predicted change amount of the volume of the intraorbital tissue on the affected side is used as a reference value for the surgical filling volume for treatment.

[0115] Figure 3 It is a schematic diagram of a computer device provided by an embodiment of the present invention, as Figure 3 shown, the device 2000 may include: one or more processors 2010, and one or more memories 2020; wherein, computer-readable code is stored in the memory, and when the computer-readable code is run by the one or more processors, the above-mentioned method can be executed.

[0116] The processor in this embodiment may be an integrated circuit chip with signal processing capabilities. The above-mentioned processor may be a general-purpose processor, 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, discrete hardware components. It can implement or execute the various methods, operations and logic block diagrams disclosed in the embodiments of the present disclosure. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc., and may be of the X86 architecture or the ARM architecture.

[0117] In general, the various example embodiments of the present disclosure may be implemented in hardware or in dedicated circuits, software, firmware, logic, or any combination thereof. Certain aspects may be implemented in hardware, while other aspects may be implemented in firmware or software that can be executed by a controller, a microprocessor, or other computing device. When aspects of the embodiments of the present disclosure are illustrated or described as block diagrams, flowcharts, or using some other graphical representation, it will be understood that the blocks, devices, systems, techniques, or methods described herein may be implemented as non-limiting examples in hardware, software, firmware, dedicated circuits or logic, general hardware or controllers or other computing devices, or some combination thereof.

[0118] For example, the method or apparatus according to an embodiment of the present disclosure may also be implemented by means of Figure 4 the architecture of the computing device 3000 shown. As Figure 4 shown, the computing device 3000 may include a bus 3010, one or more CPUs 3020, a read-only memory (ROM) 3030, a random access memory (RAM) 3040, a communication port 3050 connected to a network, an input / output component 3060, a hard disk 3070, etc. The storage devices in the computing device 3000, such as the ROM 3030 or the hard disk 3070, may store various data or files used for the processing and / or communication of the methods provided by the present disclosure and the program instructions executed by the CPU. The computing device 3000 may also include a user interface 3080. Of course, Figure 4 the architecture shown is only exemplary, and when implementing different devices, one or more components shown in the Figure 4 computing device may be omitted according to actual needs.

[0119] An embodiment of the present invention also provides a computer-readable storage medium, such as Figure 5As shown, it is a schematic diagram of a storage medium 4000 provided by an embodiment of the present invention. Computer-readable instructions 4010 are stored on the computer storage medium 4020. When the computer-readable instructions 4010 are run by a processor, the method according to the embodiments of the present disclosure described with reference to the above figures can be executed. The computer-readable storage medium in the embodiments of the present disclosure may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. The non-volatile memory may be 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), or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example but not limitation, many forms of RAM are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM), and direct memory bus random access memory (DR RAM). It should be noted that the memories of the methods described herein are intended to include but are not limited to these and any other suitable types of memories. It should be noted that the memories of the methods described herein are intended to include but are not limited to these and any other suitable types of memories.

[0120] The embodiments of the present disclosure also provide a computer program product or a computer program, which when executed by a processor implements the steps of the above method, such as Figure 2 As shown, the computer program product or the computer program includes:

[0121] An acquisition module 201: configured to acquire an image set of a patient with enophthalmos, and the images in the image set include the orbits of the healthy side and the affected side;

[0122] A labeling module 202: configured to label the image set to obtain the boundaries of the orbital cavity tissue regions of the healthy side and the affected side, and the orbital cavity tissue includes the eyeball, orbital fat, and any one or more of the following: extraocular muscles, optic nerve, cornea, lacrimal gland;

[0123] A disease variable calculation module 203: configured to respectively extract the volumes of the orbital cavity tissues of the healthy side and the affected side based on the boundaries of the orbital cavity tissue regions, and calculate the volume change amounts of the orbital cavity tissues of the affected side and the healthy side;

[0124] A model training module 204: configured to use the image set as a training data set and the volume change amount of the orbital cavity tissue as a prediction target to input into a neural network model for training to obtain a constructed enophthalmos lesion volume evaluation model.

[0125] It should be noted that the flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of systems, methods, and computer program products according to various embodiments of the present disclosure. 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 some alternative implementations, the functions marked in the blocks may occur in a different order from that marked in the accompanying drawings. For example, two consecutively represented 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 block diagram and / or flowchart, as well as the combination of blocks in the block diagram and / or flowchart, can be implemented by a dedicated hardware-based system for performing the specified functions or operations, or can be implemented by a combination of dedicated hardware and computer instructions.

[0126] Generally speaking, the various exemplary embodiments of the present disclosure can be implemented in hardware or dedicated circuits, software, firmware, logic, or any combination thereof. Some aspects can be implemented in hardware, while other aspects can be implemented in firmware or software that can be executed by a controller, a microprocessor, or other computing devices. When aspects of the embodiments of the present disclosure are illustrated or described as block diagrams, flowcharts, or using some other graphical representation, it will be understood that the blocks, devices, systems, techniques, or methods described herein can be implemented as non-limiting examples in hardware, software, firmware, dedicated circuits or logic, general hardware or controllers or other computing devices, or some combination thereof.

[0127] Those skilled in the art can clearly understand that for the convenience and conciseness of description, the specific working processes of the systems, devices, and units described above can refer to the corresponding processes in the foregoing method embodiments, and will not be elaborated herein.

[0128] In the several embodiments provided in the present 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 illustrative. For example, the division of the units is only a logical function division, and there may be other division methods in actual implementation. For example, 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 couplings or direct couplings or communication connections shown or discussed with each other can be indirect couplings or communication connections through some interfaces, devices, or units, and can be in electrical, mechanical, or other forms.

[0129] The unit described as a separation component may or may not be physically separated. The component shown as a unit may or may not be a physical unit, that is, it may be located in one place or may be distributed over multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0130] In addition, each functional unit in various embodiments of the present invention may be integrated in a processing unit, may exist separately as individual physical units, or two or more units may be integrated in one unit. The above integrated unit may be implemented in the form of hardware or in the form of a software functional unit.

[0131] The exemplary embodiments of the present disclosure described in detail above are merely illustrative and not restrictive. Those skilled in the art should understand that various modifications and combinations can be made to these embodiments or their features without departing from the principles and spirit of the present disclosure, and such modifications should fall within the scope of the present disclosure.

Claims

1. A method for constructing an enophthalmos pathology assessment model, characterized in that: The method comprises: S1: acquiring an image set of patients with enophthalmos, wherein the images in the image set include a healthy eye and an affected eye; S2: annotating the image set to obtain the orbital tissue area boundaries of the healthy side and the affected side, wherein the orbital tissue includes the eyeball, orbital fat and any one or more of the following: extraocular muscles, optic nerves, corneas, and lacrimal glands; S3: extracting the orbital tissue volumes of the healthy side and the affected side based on the orbital tissue area boundaries, and calculating the orbital tissue volume changes of the affected side and the healthy side; S4: using the image set as a training data set, and inputting the orbital tissue volume changes as a prediction target into a neural network model, and obtaining an enophthalmos lesion volume assessment model after training.

2. The method for constructing an enophthalmos pathology assessment model according to claim 1, characterized in that: The extraocular muscles include one or more of the following: medial rectus muscle, lateral rectus muscle, superior rectus muscle, inferior rectus muscle, superior oblique muscle, and inferior oblique muscle.

3. The method for constructing an enophthalmos pathology assessment model according to claim 1, characterized in that: The method further includes: performing three-dimensional reconstruction based on the image set and the annotated orbital tissue region boundaries to obtain three-dimensional boundaries of the orbital tissue; extracting the orbital tissue volumes of the healthy side and the affected side based on the three-dimensional boundaries of the orbital tissue, and calculating the changes in the orbital tissue volumes of the affected side and the healthy side.

4. The method for constructing an enophthalmos pathology assessment model according to claim 1, characterized in that: The method further comprises: marking the orbital boundary during marking, extracting the orbital volume and orbital tissue volume of the healthy side and the affected side respectively based on the orbital boundary and the orbital tissue area boundary, and calculating the orbital volume change and orbital tissue volume change of the affected side and the healthy side; calculating the relative volume change between the affected side and the healthy side based on the orbital volume change and orbital tissue volume change; using the image set as a training data set, and inputting the relative volume change as a prediction target into a neural network model to obtain a relative volume assessment model for enophthalmos lesions after training.

5. The method for constructing an enophthalmos pathology assessment model according to claim 1, characterized in that: The orbital cavity tissue volume change is the difference between the orbital cavity tissue volumes of the healthy side and the affected side; Optionally, the orbital tissue volume change is expressed as: in, represents the volume change of orbital tissue, represents the volume of orbital tissue on the affected side. represents the volume of orbital cavity tissue on the healthy side; Optionally, the relative volume change is expressed as: Where RVD represents the relative volume change, Indicates the change in orbital volume; It indicates the volume change of orbital tissue; Optionally, the orbital volume change is expressed as: in, represents the change in orbital volume, represents the orbital volume of the affected side. Represents the orbital volume of the healthy side.

6. A method for calculating the amount of enophthalmos lesions, characterized in that: The method comprises: acquiring an image of an enophthalmos patient; inputting the image of the enophthalmos patient into the enophthalmos lesion volume assessment model described in any one of claims 1 to 3 or the enophthalmos lesion relative volume assessment model described in any one of claims 4 to 5 to obtain the enophthalmos lesion volume of the patient.

7. A method for calculating the filling amount for the treatment of enophthalmos, characterized in that: The method comprises: obtaining the amount of enophthalmos of the patient according to the method of claim 6; and determining the filler volume during filling treatment based on the amount of enophthalmos and the properties of the filler.

8. A computer device, characterized in that: The device comprises: a memory and a processor; the memory is used to store a computer program; the processor executes the computer program to implement the steps of the method according to any one of claims 1 to 7.

9. A computer-readable storage medium, characterized in that: A computer program is stored thereon, and when the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method described in any one of claims 1 to 7 are implemented.

Images