Orthokeratology lens fitting method, orthokeratology lens fitting system, portable shooting equipment and electronic equipment

Abstract

The invention discloses an orthokeratology lens fitting method and system, portable shooting equipment and electronic equipment. According to the method, the lens parameters are dynamically optimized in real time, and the problem of adaptation deviation caused by cornea morphological change in traditional fitting is directly solved. A continuous adjustment mechanism of lens radian and pressure distribution can accurately match biomechanical characteristics of the cornea of a patient, and local compression or insufficient correction caused by long-term wearing is avoided. The adaptive capacity remarkably improves the fitting degree of the lens and the cornea, the vision correction effect is more stable, meanwhile, the eye discomfort risk caused by fixed parameters is reduced, the intelligent dynamic parameter optimization engine combines professional judgment of doctors with an automatic algorithm, and the intelligent dynamic parameter optimization engine is more efficient. Therefore, the fitting process is converted into a continuous tracking dynamic closed loop from a single static operation. A doctor can check a parameter adjustment track in real time through the cloud platform and quickly confirm an optimization scheme, so that the time cost of manual repeated measurement and calculation is reduced.

Description

Technical Field

[0001] The present invention belongs to the technical field of orthokeratology lens fitting, specifically to a method, system, portable shooting equipment and electronic equipment for orthokeratology lens fitting. Background Art

[0002] Orthokeratology lenses, also known as OK lenses, are rigid, gas-permeable contact lenses with an inverse geometric design. Wearing them at night, they leverage the pressure of the tear film when the eye is closed to apply targeted mechanical pressure to the central cornea, flattening the curvature of the cornea. This temporarily reduces myopia and improves uncorrected vision. It's a reversible, non-surgical, physical correction method. Orthokeratology lenses are commonly used to control the progression of myopia in adolescents. They should be fitted under the guidance of a professional ophthalmologist and undergo regular checkups to ensure safety and effectiveness. The materials used are breathable and biocompatible to maintain corneal health. The effects of orthokeratology lenses are temporary; myopia generally recovers after discontinuation, so long-term wear is essential. Because they act directly on the cornea, they require high hygiene and lens care. Improper use can increase the risk of corneal infection. As a method for preventing and controlling myopia, orthokeratology lenses should be combined with healthy eye habits and regular eye exams to achieve optimal control.

[0003] However, current solutions use paper records or local databases to store patient information, forcing doctors to manually integrate data and lacking an efficient, real-time doctor-patient communication platform. These solutions result in untimely information updates and isolated data, impacting doctor decision-making efficiency and patient experience, while also making it difficult to ensure fitting accuracy and data reliability. Summary of the Invention

[0004] The purpose of the present invention is to provide a method, system, portable photographing device and electronic device for fitting orthokeratology lenses in order to solve the above-mentioned problems.

[0005] The technical solution adopted by the present invention is as follows: a method for fitting orthokeratology lenses, the method comprising the following steps: S1: Patients use a portable corneal imaging device to capture high-resolution eye images. The device has a built-in convex lens assembly and optimized light source to ensure image clarity and stability. S2: Eye images are uploaded to the cloud platform in real time. The system automatically associates the patient ID and integrates historical fitting records, follow-up data, and wearing feedback information. S3: The medical image analysis module standardizes eye images and extracts corneal curvature, thickness, and morphological features through a deep learning algorithm to generate a three-dimensional corneal model. S4: Based on the 3D corneal model and the patient's personalized data, the cloud platform generates initial orthokeratology lens design parameters and displays key indicators through a visual interface; S5: Doctors use an intelligent dynamic parameter optimization engine to dynamically adjust the lens curvature and pressure distribution parameters based on real-time feedback of wearing comfort data and corneal morphology trends. S6: The final fitting parameters are synchronized to the doctor-patient interaction platform. Patients can receive remote guidance and regularly upload wearing data. The system automatically triggers follow-up reminders and cloud data updates.

[0006] In a preferred embodiment, in step S1, the patient's eye image is captured by a portable corneal photography device. The device adopts a modular design and includes an adjustable bracket, a clamping base and a lens barrel structure with a light-shielding step. The bracket adjustment accuracy is ±0.1 mm, which is adapted to different facial contours. The built-in annular cold light source system provides uniform illumination, the light source color temperature is 5500K, and the brightness supports 10 levels of adjustment to ensure that the eye image is free of glare interference. The core imaging component is a high-resolution optical convex lens group with a fixed focal length of 35 mm. It is equipped with a 5-megapixel CMOS sensor and the captured image resolution reaches 2560×1920 pixels. During shooting, the patient wears a cow horn eye mask to fix the head. The device automatically calibrates the eye distance through an infrared positioning sensor. The error is controlled within ±0.3 mm. The single acquisition time does not exceed 3 seconds, and the imaging data is stored in real time in a local encrypted cache area.

[0007] In a preferred embodiment, in step S2, the collected eye image is encrypted and transmitted to the cloud platform via the 5G or Wi-Fi 6 protocol, and the transmission delay is less than 50 milliseconds. The cloud system builds a distributed database index based on the patient ID. The ID is uniquely generated by a combination of 18 digits and letters. The associated fields include the fitting timestamp, corneal morphology hash value and device serial number. The data integration engine uses a time series database to store historical fitting records, supports 100,000 concurrent queries per second, and automatically matches the follow-up data, wearing feedback logs and re-examination plans of the same patient. The system realizes real-time data synchronization through a microservice architecture. The loading delay of the doctor-side interface does not exceed 0.5 seconds. The key indicators are dynamically displayed in heat maps and line charts, and support sliding and filtering the data range by timeline.

[0008] In a preferred embodiment, in step S3, the eye image is first standardized and pre-processed, noise is removed using a non-local mean filtering algorithm, and the image contrast is enhanced to 150% using a dynamic range expansion technique. The deep learning model uses a U-Net++ architecture, the input image size is normalized to 512x512 pixels, and the output layer includes the corneal curvature, vertex thickness, and peripheral shape contour semantic segmentation results. In the feature extraction stage, the curvature radius gradient is calculated by a residual convolutional network, with an accuracy of 0.01 millimeters and a thickness measurement error of less than ±2 microns. The three-dimensional corneal model reconstruction is based on a finite element mesh partitioning algorithm, with a grid density of 200 nodes per square millimeter, and a curvature continuity error of less than 0.05%. The generated standard STL model file is synchronized in real time to the cloud design platform.

[0009] In a preferred embodiment, in step S4, the cloud platform generates initial design parameters based on the three-dimensional corneal model and patient-specific data. The parameter calculation engine integrates a biomechanical simulation module, uses an Ogden hyperelastic material model to simulate corneal deformation, sets the iteration step size to 0.01 seconds, and the convergence threshold to 0.001 millimeters. The initial parameters include the lens base curve radius, optical zone diameter, and peripheral arc pressure distribution, with numerical precision retained to three decimal places. The visualization interface displays the cornea and lens contact stress cloud map using a three-dimensional rendering engine, supports doctors to drag and adjust the pressure distribution weight, and refreshes in real time at a frequency of 60 frames per second. The key indicator panel synchronously displays the curvature matching degree, predicted corrected visual acuity, and safety factor, with data update delay less than 100 milliseconds.

[0010] In a preferred embodiment, in step S5, the intelligent dynamic parameter optimization engine is used to realize real-time dynamic adjustment of the lens parameters. The engine first integrates multiple data streams, including the patient's corneal shape change trend after wearing, real-time comfort feedback, visual acuity correction effect, and historical fitting parameters. The corneal shape data is continuously collected by a portable shooting device and transmitted to the analysis module through the cloud platform, and key features such as curvature radius and local thickness change rate are extracted. At the same time, the patient's comfort feedback is recorded into the system through a mobile application or a smart wearable device, forming a structured data label. The engine uses a time series analysis model to dynamically correlate and trend the above data, identify the micro-deformation law and adaptability threshold of the cornea during wearing; In this process, the intelligent dynamic parameter optimization engine adopts a double-layer optimization architecture. The first layer builds an initial adjustment range of lens parameters based on historical data and real-time feedback, and the second layer generates dynamic correction factors combined with a deep learning prediction model. When the system detects that the local corneal pressure distribution deviates from the preset safe interval, the engine triggers an adaptive adjustment mechanism to optimize the curvature gradient and pressure distribution weight of the lens arc through iterative calculation. The corrected parameters are immediately synchronized to the cloud, and doctors can view the parameter adjustment trajectory through the visualization interface and manually confirm or fine-tune the optimization scheme recommended by the system to ensure that the lens design and the biomechanical properties of the patient's cornea are highly matched.

[0011] The calculation formula of the dynamic curvature gradient correction model in the dynamic adjustment process of the lens arc is: ; Where ΔK t represents the curvature gradient correction amount at time t, C m is the real-time monitoring value of corneal curvature, F ci and F si represent the current pressure value and safety threshold of the i-th region, respectively, and α, β, γ are dynamic weight coefficients. This formula combines the instantaneous change rate of corneal shape and multi-region pressure deviation, balances the contribution of historical data and current state through an exponential decay function, and realizes the gradual optimization of curvature parameters.

[0012] The dynamic balance of lens pressure distribution is driven by the following objective function calculation formula: ; Where P d is the ideal pressure distribution, P o is the current actual distribution, D j represents the deviation of the j-th comfort index, σ j is the statistical standard deviation, and λ1 and λ2 are weight coefficients. This formula minimizes the weighted combination of pressure distribution error and comfort deviation to ensure that the lens maximizes wearing comfort while correcting vision, reflecting the dual optimization of biomechanical adaptation and patient subjective experience.

[0013] In a preferred embodiment, the final fitting parameters in step S6 are pushed to the doctor-patient interaction platform through the HTTPS two-way authentication protocol, and the patient mobile terminal application adopts AES-256 encryption to receive the lens design file and wearing guide. The system automatically generates daily wearing data collection tasks, and records the wearing time and lens cleanliness through the Bluetooth 5.0 protocol connection intelligent mirror box, and the data upload interval is 24 hours. The re-examination reminding module analyzes the corneal shape change curve based on the dynamic time warping algorithm, if the continuous three monitoring data deviates from the preset safety threshold by more than 5%, immediately trigger SMS and application push reminder. The cloud database adopts multi-redundant storage, the data persistence availability reaches 99.999%, and the historical version supports on-demand backtracking comparison.

[0014] In summary, due to the adoption of the technical scheme, the application has the following beneficial effects: 1、In the application, by real-time dynamic optimization of lens parameters, the problem of fitting deviation caused by corneal shape change in traditional fitting is directly solved. The continuous adjustment mechanism of lens curvature and pressure distribution can accurately match the biomechanical properties of the patient's cornea, avoiding local compression or insufficient correction caused by long-term wearing. This adaptive ability significantly improves the fit of the lens and the cornea, ensures more stable vision correction effect, and reduces the risk of eye discomfort caused by fixed parameters.

[0015] 2、In the application, the intelligent dynamic parameter optimization engine combines the doctor's professional judgment with automatic algorithms, so that the fitting process changes from a single static operation to a continuous tracking dynamic closed loop. Doctors can view the parameter adjustment trajectory in real time through the cloud platform, quickly confirm the optimization scheme, and reduce the time cost of manual repeated calculation. Patients can also obtain personalized wearing suggestions in a timely manner through regular data upload and remote guidance, avoid the decline of correction effect caused by communication delay or information asymmetry, and form a more efficient whole-process management mode.

[0016] 3、In the application, medical specifications and patient subjective feelings are deeply integrated, lens design not only meets the clinical safety threshold, but also takes into account the actual needs such as wearing comfort. The dynamic adjustment mechanism can quickly respond to corneal micro-deformation and comfort feedback, avoiding the poor adaptability problem caused by parameter lag in traditional fitting. Patients do not need to go to the hospital for re-examination frequently, and can complete parameter iteration through remote data synchronization, which not only saves time cost, but also ensures that the correction effect is continuously optimized with individual changes, improving the reliability and user trust of long-term wearing. BRIEF DESCRIPTION OF DRAWINGS

[0017] Figure 1 The flow principle diagram of the application. DETAILED DESCRIPTION

[0018] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. Example

[0019] Reference Figure 1 , a method for fitting orthokeratology lenses, the method comprising the following steps: S1: Patients use a portable corneal imaging device to capture high-resolution eye images. The device has a built-in convex lens assembly and optimized light source to ensure image clarity and stability. S2: Eye images are uploaded to the cloud platform in real time. The system automatically associates the patient ID and integrates historical fitting records, follow-up data, and wearing feedback information. S3: The medical image analysis module standardizes eye images and extracts corneal curvature, thickness, and morphological features through a deep learning algorithm to generate a three-dimensional corneal model. S4: Based on the 3D corneal model and the patient's personalized data, the cloud platform generates initial orthokeratology lens design parameters and displays key indicators through a visual interface; S5: Doctors use an intelligent dynamic parameter optimization engine to dynamically adjust the lens curvature and pressure distribution parameters based on real-time feedback of wearing comfort data and corneal morphology trends. S6: The final fitting parameters are synchronized to the doctor-patient interaction platform. Patients can receive remote guidance and regularly upload wearing data. The system automatically triggers follow-up reminders and cloud data updates.

[0020] In step S1, the patient's eye image is acquired through a portable corneal photography device. The device adopts a modular design and includes an adjustable bracket, a clamping base, and a lens barrel structure with a light-shielding step. The bracket adjustment accuracy is ±0.1 mm, which is suitable for different facial contours. The built-in annular cold light source system provides uniform illumination. The light source color temperature is 5500K and the brightness supports 10 levels of adjustment to ensure that the eye image is free of glare interference. The core imaging component is a high-resolution optical convex lens group with a fixed focal length of 35 mm. With a 5-megapixel CMOS sensor, the image resolution reaches 2560×1920 pixels. During shooting, the patient wears a horn eye mask to fix the head. The device automatically calibrates the eye distance through an infrared positioning sensor. The error is controlled within ±0.3 mm. The single acquisition time does not exceed 3 seconds. The imaging data is stored in real time in a local encrypted cache area.

[0021] In step S2, the collected eye images are encrypted and transmitted to the cloud platform via the 5G or Wi-Fi 6 protocol, with a transmission delay of less than 50 milliseconds. The cloud system builds a distributed database index based on the patient ID. The ID is uniquely generated by a combination of 18 digits and letters. The associated fields include the fitting timestamp, corneal morphology hash value, and device serial number. The data integration engine uses a time series database to store historical fitting records, supports 100,000 concurrent queries per second, and automatically matches the follow-up data, wearing feedback logs, and re-examination plans of the same patient. The system achieves real-time data synchronization through a microservice architecture. The loading delay of the doctor-side interface does not exceed 0.5 seconds. Key indicators are dynamically displayed in heat maps and line charts, and support sliding and filtering data ranges by timeline.

[0022] In step S3, the eye image is first standardized and preprocessed, and the non-local mean filtering algorithm is used to remove noise. The dynamic range extension technology is used to increase the image contrast to 150%. The deep learning model uses the U-Net++ architecture. The input image size is normalized to 512×512 pixels. The output layer contains the semantic segmentation results of corneal curvature, vertex thickness, and peripheral morphological contours. During the feature extraction stage, the curvature radius gradient is calculated through a residual convolutional network with an accuracy of 0.01 mm and a thickness measurement error of less than ±2 microns. The three-dimensional corneal model is reconstructed based on the finite element meshing algorithm with a mesh density of 200 nodes per square millimeter and a curvature continuity error of less than 0.05%. The generated standard STL model file is synchronized to the cloud design platform in real time.

[0023] In step S4, the cloud platform generates initial design parameters based on the three-dimensional corneal model and the patient's personalized data. The parameter calculation engine integrates a biomechanical simulation module and uses the Ogden hyperelastic material model to simulate corneal deformation. The iteration step is set to 0.01 seconds and the convergence threshold is 0.001 mm. The initial parameters include the lens base curve radius, the optical zone diameter and the peripheral arc pressure distribution, and the numerical accuracy is retained to three decimal places. The visual interface uses a three-dimensional rendering engine to display the contact stress cloud map of the cornea and the lens, supporting doctors to drag and adjust the pressure distribution weight, and the real-time refresh rate is 60 frames per second. The key indicator panel simultaneously displays the curvature matching, predicted corrected visual acuity and safety factor, and the data update delay is less than 100 milliseconds.

[0024] In step S5, real-time dynamic adjustment of lens parameters is achieved through an intelligent dynamic parameter optimization engine. The engine first integrates multi-source data streams, including the trend of corneal morphology changes after the patient wears the lens, real-time comfort feedback, vision correction effects, and historical fitting parameters. Corneal morphology data is continuously collected by a portable camera and transmitted to the analysis module through a cloud platform to extract key features such as curvature radius and local thickness change rate. At the same time, the patient's comfort feedback is entered into the system in real time through a mobile application or smart wearable device to form a structured data tag. Based on a time series analysis model, the engine dynamically correlates and predicts trends of the above data to identify the micro-deformation pattern and adaptability threshold of the cornea during the wearing process.

[0025] During this process, the intelligent dynamic parameter optimization engine adopts a two-layer optimization architecture. The first layer builds the initial adjustment range of the lens parameters based on historical data and real-time feedback, and the second layer generates dynamic correction factors in combination with deep learning prediction models. When the system detects that the local pressure distribution of the cornea deviates from the preset safety range, the engine triggers the adaptive adjustment mechanism to optimize the curvature gradient and pressure distribution weight of the lens curvature through iterative calculation. The corrected parameters are immediately synchronized to the cloud. Doctors can view the parameter adjustment trajectory through the visual interface and manually confirm or fine-tune according to the optimization plan recommended by the system to ensure that the lens design is highly matched with the biomechanical properties of the patient's cornea.

[0026] The calculation formula of the dynamic curvature gradient correction model during the dynamic adjustment of the lens curvature is: ; Where ΔK t Indicates the curvature gradient correction at time t, C m is the real-time monitoring value of corneal curvature, F ci With F si Represent the current pressure value and safety threshold of the i-th region, respectively. α, β, and γ are dynamic weight coefficients. This formula combines the instantaneous rate of change of corneal morphology with the multi-region pressure deviation. It balances the contribution of historical data and the current state through an exponential decay function to achieve progressive optimization of the curvature parameters.

[0027] The dynamic balance of lens pressure distribution is driven by the following objective function and the calculation formula is: ; Among them, P d is the ideal pressure distribution, P o is the current actual distribution, D j represents the deviation of the j-th comfort index, σ jis the statistical standard deviation, and λ1 and λ2 are weighting coefficients. This formula minimizes the weighted combination of pressure distribution error and comfort deviation, ensuring that the lens maximizes wearing comfort while correcting vision, reflecting the dual optimization of biomechanical adaptation and patient subjective experience.

[0028] In step S6, the final fitting parameters are pushed to the doctor-patient interaction platform through the HTTPS two-way authentication protocol, and the patient's mobile application uses AES-256 encryption to receive the lens design files and wearing instructions. The system automatically generates daily wearing data collection tasks, connects to the smart mirror box via the Bluetooth 5.0 protocol to record the wearing time and lens cleanliness, and the data upload interval is 24 hours. The follow-up reminder module analyzes the corneal morphology change curve based on the dynamic time warping algorithm. If the monitoring data deviates from the preset safety threshold by more than 5% for three consecutive times, an SMS and in-app push reminder will be triggered immediately. The cloud database uses multi-copy redundant storage, and the data persistence availability reaches 99.999%. Historical versions support on-demand backtracking and comparison.

[0029] A corneal reshaping lens fitting system, which runs the corneal reshaping lens fitting method as described above when in use.

[0030] A portable camera device for fitting orthokeratology lenses includes a processor and a storage device; the processor loads and executes instructions and data in the storage device to implement the above-mentioned orthokeratology lens fitting method.

[0031] An electronic device for fitting orthokeratology lenses includes: a processor and a storage device; the processor loads and executes instructions and data in the storage device to implement the above-mentioned orthokeratology lens fitting method.

[0032] This invention directly addresses the problem of fit deviation caused by corneal morphology changes in traditional lens fitting by dynamically optimizing lens parameters in real time. The continuous adjustment of the lens' curvature and pressure distribution precisely matches the biomechanical properties of the patient's cornea, avoiding localized compression or undercorrection caused by long-term wear. This adaptive capability significantly improves the fit of the lens to the cornea, ensuring more stable vision correction while reducing the risk of eye discomfort caused by fixed parameters.

[0033] In this invention, an intelligent dynamic parameter optimization engine combines the physician's professional judgment with automated algorithms, transforming the fitting process from a single, static operation into a dynamic, closed-loop, continuous tracking process. Physicians can view parameter adjustment trajectories in real time through a cloud-based platform, quickly confirming optimization solutions and reducing the time and effort of manual recalculation. Patients can also receive timely, personalized fitting recommendations through regular data uploads and remote guidance, avoiding compromised correction outcomes due to communication delays or information asymmetry, and creating a more efficient, end-to-end management model.

[0034] This invention deeply integrates medical standards with patient experience, resulting in lens designs that not only meet clinical safety thresholds but also address practical needs such as wearing comfort. The dynamic adjustment mechanism rapidly responds to corneal micro-deformation and comfort feedback, avoiding the poor adaptability associated with parameter lag in traditional fitting. Patients no longer need frequent hospital visits; parameter iteration can be completed through remote data synchronization, saving time and costs while ensuring continuous optimization of correction effects as individuals change, enhancing long-term wear reliability and user trust.

[0035] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprises" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device that includes a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the presence of other identical elements in the process, method, article or device that includes the element.

[0036] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A method for fitting orthokeratology lenses, characterized by: The method comprises the following steps: S1: The patient uses a portable corneal imaging device to capture high-resolution eye images. The device has a built-in convex lens assembly and optimized light source to ensure image clarity and stability. S2: Eye images are uploaded to the cloud platform in real time. The system automatically associates the patient ID and integrates historical fitting records, follow-up data, and wearing feedback information. S3: The medical image analysis module standardizes eye images and extracts corneal curvature, thickness, and morphological features using a deep learning algorithm to generate a three-dimensional corneal model. S4: Based on the 3D corneal model and the patient's personalized data, the cloud platform generates initial orthokeratology lens design parameters and displays key indicators through a visual interface; S5: Doctors use an intelligent dynamic parameter optimization engine to dynamically adjust the lens curvature and pressure distribution parameters based on real-time feedback of wearing comfort data and corneal morphology trends. S6: The final fitting parameters are synchronized to the doctor-patient interaction platform. Patients can receive remote guidance and regularly upload wearing data. The system automatically triggers follow-up reminders and cloud data updates.

2. The orthokeratology lens fitting method according to claim 1, wherein: In step S1, the patient's eye image is captured by a portable corneal photography device; the device adopts a modular design, including an adjustable bracket, a clamping base and a lens barrel structure with a light-shielding step. The bracket adjustment accuracy is ±0.1 mm, which is suitable for different facial contours; the built-in annular cold light source system provides uniform lighting, the light source color temperature is 5500K, and the brightness supports 10 levels of adjustment to ensure that the eye image is free of glare interference; the core imaging component is a high-resolution optical convex lens group with a fixed focal length of 35 mm, combined with a 5-megapixel CMOS sensor, and the captured image resolution reaches 2560×1920 pixels; when shooting, the patient wears a cow horn eye mask to fix the head, and the device automatically calibrates the eye distance through an infrared positioning sensor, and the error is controlled within ±0.3 mm. The single acquisition time does not exceed 3 seconds, and the imaging data is stored in real time in a local encrypted cache area.

3. The orthokeratology lens fitting method according to claim 1, wherein: In step S2, the collected eye image is encrypted and transmitted to the cloud platform via the 5G or Wi-Fi 6 protocol, with a transmission delay of less than 50 milliseconds; The cloud-based system builds a distributed database index based on the patient ID. The ID is uniquely generated by a combination of 18 digits and letters. Associated fields include fitting timestamp, corneal morphology hash value and device serial number. The data integration engine uses a time-series database to store historical fitting records, supports 100,000 concurrent queries per second, and automatically matches the follow-up data, wearing feedback logs and re-visit plans of the same patient. The system achieves real-time data synchronization through a microservice architecture. The loading delay of the doctor-side interface does not exceed 0.5 seconds. Key indicators are dynamically displayed in heat maps and line charts, and support is provided for sliding and filtering data ranges by timeline.

4. The orthokeratology lens fitting method according to claim 1, wherein: In step S3, the eye image is first standardized and preprocessed, and noise is removed using a non-local mean filtering algorithm. The dynamic range extension technology is used to increase the image contrast to 150%. The deep learning model adopts the U-Net++ architecture, the input image size is normalized to 512×512 pixels, and the output layer contains the semantic segmentation results of corneal curvature, vertex thickness, and peripheral morphological contours. The feature extraction stage calculates the curvature radius gradient through a residual convolutional network with an accuracy of 0.01 mm and a thickness measurement error of less than ±2 microns. The three-dimensional corneal model is reconstructed based on a finite element meshing algorithm with a mesh density of 200 nodes per square millimeter and a curvature continuity error of less than 0.05%. The generated standard STL model file is synchronized to the cloud design platform in real time.

5. The orthokeratology lens fitting method according to claim 1, wherein: In step S4, the cloud platform generates initial design parameters based on the three-dimensional corneal model and the patient's personalized data; the parameter calculation engine integrates a biomechanical simulation module, uses the Ogden hyperelastic material model to simulate corneal deformation, sets the iteration step to 0.01 seconds, and the convergence threshold to 0.001 mm; the initial parameters include the lens base curve radius, the optical zone diameter, and the peripheral arc pressure distribution, with numerical accuracy retained to three decimal places; the visualization interface uses a three-dimensional rendering engine to display the contact stress cloud map of the cornea and the lens, supports doctors to drag and adjust the pressure distribution weight, and has a real-time refresh rate of 60 frames per second; the key indicator panel synchronously displays the curvature matching, predicted corrected visual acuity, and safety factor, with a data update delay of less than 100 milliseconds.

6. The orthokeratology lens fitting method according to claim 1, wherein: In step S5, real-time dynamic adjustment of lens parameters is achieved through an intelligent dynamic parameter optimization engine. The engine first integrates multi-source data streams, including corneal morphology change trends after patient wear, real-time comfort feedback, vision correction effects, and historical fitting parameters. Corneal morphology data is continuously collected by a portable camera and transmitted to an analysis module via a cloud platform to extract key features such as curvature radius and local thickness change rate. At the same time, patient comfort feedback is entered into the system in real time via a mobile application or smart wearable device to form structured data tags. The engine dynamically correlates and predicts trends of the above data based on a time series analysis model to identify micro-deformation patterns and adaptability thresholds of the cornea during wear. During this process, the intelligent dynamic parameter optimization engine adopts a two-layer optimization architecture; The first layer establishes the initial adjustment range of lens parameters based on historical data and real-time feedback, while the second layer generates dynamic correction factors in conjunction with a deep learning prediction model. When the system detects that the local pressure distribution of the cornea deviates from the preset safety range, the engine triggers an adaptive adjustment mechanism, optimizing the curvature gradient and pressure distribution weight of the lens arc through iterative calculations. The corrected parameters are immediately synchronized to the cloud, allowing doctors to view the parameter adjustment trajectory through a visual interface and manually confirm or fine-tune the optimization plan recommended by the system to ensure that the lens design is highly compatible with the biomechanical properties of the patient's cornea. The calculation formula of the dynamic curvature gradient correction model during the dynamic adjustment of the lens curvature is: ； Where ΔK t Indicates the curvature gradient correction at time t, C m is the real-time monitoring value of corneal curvature, F ci With F si Represent the current pressure value and safety threshold of the i-th region, respectively. α, β, and γ are dynamic weight coefficients. This formula combines the instantaneous rate of change of corneal morphology with the multi-region pressure deviation, and balances the contribution of historical data and current state through an exponential decay function to achieve progressive optimization of curvature parameters. The dynamic balance of lens pressure distribution is driven by the following objective function and the calculation formula is: ； Among them, P d is the ideal pressure distribution, P o is the current actual distribution, D j represents the deviation of the j-th comfort index, σ j is the statistical standard deviation, λ1 and λ2 are weight coefficients; this formula ensures that the lens maximizes wearing comfort while correcting vision by minimizing the weighted combination of pressure distribution error and comfort deviation, reflecting the dual optimization of biomechanical adaptation and patient subjective experience.

7. The orthokeratology lens fitting method according to claim 1, wherein: In step S6, the final fitting parameters are pushed to the doctor-patient interaction platform via the HTTPS two-way authentication protocol. The patient's mobile application uses AES-256 encryption to receive the lens design files and wearing instructions. The system automatically generates a daily wear data collection task, connects to the smart lens box via the Bluetooth 5.0 protocol to record the wearing time and lens cleanliness, and the data upload interval is 24 hours. The follow-up reminder module analyzes the corneal morphology change curve based on the dynamic time warping algorithm. If the monitoring data deviates from the preset safety threshold by more than 5% for three consecutive times, it will immediately trigger a text message and in-app push reminder. The cloud database uses multi-copy redundant storage, with data persistence availability reaching 99.999%, and historical versions support on-demand backtracking and comparison.

8. Orthokeratology lens fitting system, characterized by: When in use, the system runs the orthokeratology lens fitting method as described in any one of claims 1 to 7.

9. A portable camera for orthokeratology lens fitting, characterized by: It comprises a processor and a storage device; the processor loads and executes instructions and data in the storage device to implement the orthokeratology lens fitting method described in any one of claims 1 to 7.

10. An electronic device for fitting orthokeratology lenses, characterized by: include: A processor and a storage device; the processor loads and executes instructions and data in the storage device to implement the orthokeratology lens fitting method described in any one of claims 1 to 7.

Images