Contact type intraocular pressure measurement method, device and equipment based on data learning and storage medium

Abstract

The invention discloses a contact type intraocular pressure measurement method, device and equipment based on data learning and a storage medium, and belongs to the technical field of intraocular pressure measurement. The method comprises the following steps: preparing intraocular pressure measurement based on personal data of a user to obtain prepared intraocular pressure measurement equipment; the intraocular pressure measuring probe abuts against the eyelids of the user for measurement, and intraocular pressure data of the user are obtained; and performing intraocular pressure calculation based on the user intraocular pressure data and the prepared intraocular pressure measurement equipment to obtain a user intraocular pressure value. The intraocular pressure measurement is prepared according to the personal data of the user, then intraocular pressure measurement is started, finally, intraocular pressure calculation is performed based on the obtained intraocular pressure data of the user, the intraocular pressure value of the user is obtained, a patient does not need to open eyes for measurement in the whole intraocular pressure measurement process, the contact infection risk is avoided, and the intraocular pressure measurement cost is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of intraocular pressure measurement, and in particular to a contact-type intraocular pressure measurement method, device, equipment and storage medium based on data learning. Background Art

[0002] Intraocular pressure (IOP) is the pressure exerted by the contents of the eye on the wall of the eyeball and is an important reference for diagnosing eye diseases such as glaucoma. Currently, there are three main types of tonometers commonly used in clinical practice: applanation, indentation, and non-contact. Applanation tonometers, such as the Goldmann tonometer, are based on the Imbert-Fick law and measure IOP by flattening the cornea to a certain area and reading the external pressure value. They are considered to be the most accurate type of tonometer. However, this method requires local anesthesia for the patient and there is a risk of cross-infection between patients. Non-contact tonometers, such as the jet tonometer, deform the cornea by spraying air onto the surface, and calculate the IOP based on the airflow pressure, thus achieving non-contact measurement. However, there are large errors in the estimation of airflow velocity and applanation area, and the measurement accuracy is not high.

[0003] Existing contact-based intraocular pressure measurement methods based on data learning all have certain drawbacks. Professional intraocular pressure equipment is expensive and requires professional training, resulting in a high learning curve. Portable intraocular pressure monitors are inaccurate and still require a high learning curve. Most intraocular pressure monitors require the eye to be open for measurement, which carries a certain risk of contact infection. Some also require anesthesia, and disposable supplies can add additional costs. Therefore, there is an urgent need for an intraocular pressure measurement method that does not require eye opening, anesthesia, or disposable supplies, and offers high accuracy and ease of use.

[0004] The above content is only used to assist in understanding the technical solution of the present invention and does not constitute an admission that the above content is prior art. Summary of the Invention

[0005] The main purpose of the present invention is to provide a contact intraocular pressure measurement method, device, equipment and storage medium based on data learning, aiming to solve the technical problems of inaccurate intraocular pressure measurement accuracy and contact infection risk in the existing technology.

[0006] To achieve the above object, the present invention provides a contact-type intraocular pressure measurement method based on data learning, the method comprising the following steps:

[0007] Prepare intraocular pressure measurement based on user personal data and obtain the prepared intraocular pressure measurement device;

[0008] Place the intraocular pressure measuring probe against the user's eyelid to measure and obtain the user's intraocular pressure data;

[0009] The intraocular pressure is calculated based on the user's intraocular pressure data and the prepared intraocular pressure measurement device to obtain the user's intraocular pressure value.

[0010] Optionally, preparing for intraocular pressure measurement based on the user's personal data to obtain a prepared intraocular pressure measurement device specifically includes:

[0011] Parse the intraocular pressure measurement instruction and obtain the user's personal data based on the instruction parsing result;

[0012] If the instruction parsing result is a historical user intraocular pressure measurement instruction, calling the user's personal data based on the intraocular pressure measurement instruction to prepare for intraocular pressure measurement and obtain a prepared intraocular pressure measurement device;

[0013] If the instruction parsing result is a new user intraocular pressure measurement instruction, the user's personal data is entered based on the intraocular pressure measurement instruction to prepare for intraocular pressure measurement, and a prepared intraocular pressure measurement device is obtained.

[0014] Optionally, performing intraocular pressure calculation based on the user intraocular pressure data and the prepared intraocular pressure measurement device to obtain the user intraocular pressure value specifically includes:

[0015] Determining an intraocular pressure calculation model based on the prepared intraocular pressure measurement device;

[0016] If the intraocular pressure calculation model is an intraocular pressure linear calculation model, inputting the user's intraocular pressure data into the intraocular pressure linear calculation model to perform intraocular pressure calculation to obtain the user's intraocular pressure value;

[0017] If the intraocular pressure calculation model is an intraocular pressure nonlinear calculation model, the user's intraocular pressure data is input into the intraocular pressure nonlinear calculation model to perform intraocular pressure calculation to obtain the user's intraocular pressure value.

[0018] Optionally, if the intraocular pressure calculation model is an intraocular pressure linear calculation model, the user's intraocular pressure data is input into the intraocular pressure linear calculation model to perform intraocular pressure calculation to obtain the user's intraocular pressure value. The construction process of the intraocular pressure linear calculation model specifically includes:

[0019] Obtain historical intraocular pressure measurement data set;

[0020] Performing intraocular pressure linear calculation model coefficient fitting based on the historical intraocular pressure measurement data set to obtain linear model coefficient results;

[0021] A model is constructed based on the linear model coefficient results to obtain an intraocular pressure linear calculation model.

[0022] Optionally, if the intraocular pressure calculation model is a nonlinear intraocular pressure calculation model, the user intraocular pressure data is input into the nonlinear intraocular pressure calculation model to perform intraocular pressure calculation to obtain the user intraocular pressure value. The construction process of the nonlinear intraocular pressure calculation model specifically includes:

[0023] Perform data preprocessing on the intraocular pressure training sample data set;

[0024] Inputting the processed intraocular pressure training sample data set into the intraocular pressure nonlinear calculation model to be trained for machine learning to obtain a trained intraocular pressure nonlinear calculation model;

[0025] Performing a model test on the trained intraocular pressure nonlinear calculation model, and outputting a corresponding intraocular pressure nonlinear calculation model if the model test is qualified;

[0026] If the model fails the test, the unqualified intraocular pressure nonlinear calculation model will be retrained by the machine until the model passes the test and the corresponding intraocular pressure nonlinear calculation model is output.

[0027] Optionally, the processed intraocular pressure training sample data set is input into the intraocular pressure nonlinear calculation model to be trained for machine learning to obtain the trained intraocular pressure nonlinear calculation model, specifically including:

[0028] Performing feature engineering selection based on the processed intraocular pressure training sample data set to obtain a filtered feature data set;

[0029] Performing model selection based on the filtered feature data set to obtain a corresponding intraocular pressure nonlinear calculation model to be trained;

[0030] The filtered feature data set is input into the intraocular pressure nonlinear calculation model to be trained for machine learning to obtain the trained intraocular pressure nonlinear calculation model.

[0031] Optionally, performing model selection based on the filtered feature data set to obtain a corresponding nonlinear intraocular pressure calculation model to be trained specifically includes:

[0032] Performing label recognition based on the filtered feature data set to obtain corresponding label recognition results;

[0033] If the label recognition result is a first-category label recognition result, the regression model is selected as the intraocular pressure nonlinear calculation model to be trained;

[0034] If the label recognition result is a second type of label recognition result, selecting a deep learning model as the intraocular pressure nonlinear calculation model to be trained;

[0035] If the label recognition result is a third-category label recognition result, the classification model is selected as the intraocular pressure nonlinear calculation model to be trained.

[0036] In addition, to achieve the above-mentioned purpose, the present invention further provides an intraocular pressure measuring device, comprising:

[0037] Measurement preparation module: prepares intraocular pressure measurement based on user personal data and obtains the prepared intraocular pressure measurement device;

[0038] Intraocular pressure measurement module: Place the intraocular pressure measurement probe against the user's eyelid to measure and obtain the user's intraocular pressure data;

[0039] An intraocular pressure calculation module calculates intraocular pressure based on the user's intraocular pressure data and the prepared intraocular pressure measurement device to obtain the user's intraocular pressure value.

[0040] In addition, to achieve the above-mentioned purpose, the present invention also proposes a contact-type intraocular pressure measurement device based on data learning, and the contact-type intraocular pressure measurement device based on data learning includes: a memory, a processor, and an intraocular pressure measurement program stored on the memory and runnable on the processor, and the intraocular pressure measurement program is configured to implement the steps of the contact-type intraocular pressure measurement method based on data learning as described above.

[0041] In addition, to achieve the above-mentioned purpose, the present invention also proposes a computer-readable storage medium storing a computer program, wherein the storage medium stores an intraocular pressure measurement program, and when the intraocular pressure measurement program is executed by a processor, the steps of the contact intraocular pressure measurement method based on data learning as described above are implemented.

[0042] The present invention prepares intraocular pressure measurement based on user personal data to obtain a prepared intraocular pressure measurement device; places an intraocular pressure measurement probe against the user's eyelids for measurement to obtain the user's intraocular pressure data; and calculates intraocular pressure based on the user's intraocular pressure data and the prepared intraocular pressure measurement device to obtain the user's intraocular pressure value. The present invention first prepares intraocular pressure measurement based on the user's personal data, then starts intraocular pressure measurement, and finally calculates intraocular pressure based on the obtained user's intraocular pressure data to obtain the user's intraocular pressure value. The entire intraocular pressure measurement process does not require the patient to open their eyes for measurement, reducing the patient's fear of intraocular pressure measurement and the discomfort of the intraocular pressure measurement process, avoiding the risk of contact infection, and reducing the cost of intraocular pressure measurement. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 Schematic diagram of the structure of an intraocular pressure measurement device in the hardware operating environment involved in an embodiment of the present invention;

[0044] Figure 2 This is a flow chart of a first embodiment of a contact-type intraocular pressure measurement method based on data learning according to the present invention;

[0045] Figure 3 This is a flow chart of a second embodiment of a contact-type intraocular pressure measurement method based on data learning according to the present invention;

[0046] Figure 4 1 is a flow chart of a third embodiment of a contact-type intraocular pressure measurement method based on data learning according to the present invention;

[0047] Figure 5 This is a structural block diagram of the first embodiment of the intraocular pressure measurement device of the present invention.

[0048] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings and in conjunction with the embodiments. DETAILED DESCRIPTION

[0049] 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.

[0050] Reference Figure 1 , Figure 1 This is a schematic diagram of the structure of an intraocular pressure measurement device in the hardware operating environment involved in the embodiment of the present invention.

[0051] like Figure 1 As shown, the intraocular pressure measurement device may include: a processor 1001, such as a central processing unit (CPU), a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. Among them, the communication bus 1002 is used to realize the connection and communication between these components. The user interface 1003 may include a display screen (Display), an input unit such as a keyboard (Keyboard), and the user interface 1003 may optionally include a standard wired interface and a wireless interface. The network interface 1004 may optionally include a standard wired interface and a wireless interface (such as a wireless fidelity (Wireless-Fidelity, Wi-Fi) interface). The memory 1005 may be a high-speed random access memory (Random Access Memory, RAM) or a stable non-volatile memory (Non-Volatile Memory, NVM), such as a disk storage. The memory 1005 may optionally be a storage device independent of the aforementioned processor 1001.

[0052] Those skilled in the art will understand that Figure 1 The structure shown in the figure does not constitute a limitation on the intraocular pressure measurement device, and may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently.

[0053] like Figure 1As shown, the memory 1005 as a storage medium may include an operating system, a network communication module, a user interface module, and an intraocular pressure measurement program.

[0054] exist Figure 1 In the intraocular pressure measurement device shown, the network interface 1004 is mainly used for data communication with the network server; the user interface 1003 is mainly used for data interaction with the user; the processor 1001 and the memory 1005 in the intraocular pressure measurement device of the present invention can be set in the intraocular pressure measurement device, and the intraocular pressure measurement device calls the intraocular pressure measurement program stored in the memory 1005 through the processor 1001, and executes the contact intraocular pressure measurement method based on data learning provided by an embodiment of the present invention.

[0055] The embodiment of the present invention provides a contact type intraocular pressure measurement method based on data learning, referring to Figure 2 , Figure 2 This is a flow chart of a first embodiment of a contact-type intraocular pressure measurement method based on data learning according to the present invention.

[0056] In this embodiment, the contact-type intraocular pressure measurement method based on data learning includes the following steps:

[0057] Step S10: preparing for intraocular pressure measurement based on the user's personal data, and obtaining a prepared intraocular pressure measurement device;

[0058] It should be noted that, in a specific implementation, the user's personal data includes the user's personal eye structure feature information and physiological status information, wherein the eye structure feature information can be obtained through hospital-related tests, which may include morphological feature information (which can be obtained through corneal topography examination and optical or ultrasonic instrument detection, which may specifically include corneal thickness, anterior chamber depth, etc.) and biomechanical property information (which can be measured by eye response analyzer and visual corneal biomechanical analyzer, which may specifically include corneal Young's modulus, viscoelastic coefficient, etc.); physiological status information can be obtained through hospital-related tests and user historical medical records, which may include hemodynamic indicators (which may come from the patient's examination data stored in the hospital, which may specifically include basal blood pressure, blood sugar, heart rate) and historical diagnosis and treatment data (which may come from the user's diagnosis and treatment records in this hospital and other hospitals, which may specifically include previous intraocular pressure measurement values and eye medication records).

[0059] It should also be noted that, in a specific implementation, the process of preparing for intraocular pressure measurement based on the user's personal data is essentially a personalized configuration based on the user's identity, so that the prepared intraocular pressure measurement device can accurately measure the user's intraocular pressure.

[0060] Step S20: placing the intraocular pressure measuring probe against the user's eyelid to measure the intraocular pressure data of the user;

[0061] It should be noted that, in a specific implementation, the intraocular pressure measurement probe includes a pressure sensor. Therefore, when the intraocular pressure measurement probe is placed against the user's eyelid for measurement, the pressure applied to the probe can be accurately obtained, that is, the user's intraocular pressure data set includes the pressure data applied to the probe.

[0062] Step S30: Calculate the intraocular pressure based on the user's intraocular pressure data and the prepared intraocular pressure measurement device to obtain the user's intraocular pressure value.

[0063] It can be understood that in the specific implementation, the process of calculating intraocular pressure based on the user's intraocular pressure data and the prepared intraocular pressure measurement equipment is essentially an intraocular pressure measurement model implemented based on the user's personal information and the measurement curve, where the measurement curve is the dynamic pressure-displacement response curve when the intraocular pressure measurement instrument is in contact with the eye.

[0064] This embodiment prepares intraocular pressure measurement based on the user's personal data to obtain a prepared intraocular pressure measurement device; places the intraocular pressure measurement probe against the user's eyelids for measurement to obtain the user's intraocular pressure data; and calculates intraocular pressure based on the user's intraocular pressure data and the prepared intraocular pressure measurement device to obtain the user's intraocular pressure value. This embodiment first prepares intraocular pressure measurement based on the user's personal data, then begins intraocular pressure measurement. Finally, it calculates intraocular pressure based on the obtained user's intraocular pressure data to obtain the user's intraocular pressure value. The entire intraocular pressure measurement process does not require the patient to open their eyes for measurement, reducing the patient's fear of intraocular pressure measurement and the discomfort of the intraocular pressure measurement process, avoiding the risk of contact infection, and reducing the cost of intraocular pressure measurement.

[0065] refer to Figure 3 , Figure 3 This is a flow chart of a second embodiment of a contact-type intraocular pressure measurement method based on data learning according to the present invention.

[0066] Based on the above first embodiment, in this embodiment, step S10 specifically includes:

[0067] Step S11: parsing the intraocular pressure measurement instruction and obtaining the user's personal data according to the instruction parsing result;

[0068] It should be noted that, in a specific implementation, the intraocular pressure measurement instruction is a working instruction of the intraocular pressure measurement device, which can be used to call or enter the user's personal data and measure the user's intraocular pressure.

[0069] Step S12: If the instruction parsing result is a historical user intraocular pressure measurement instruction, the user's personal data is called based on the intraocular pressure measurement instruction to prepare for intraocular pressure measurement and obtain the prepared intraocular pressure measurement device;

[0070] It should be noted that in the specific implementation, when the instruction parsing result is a historical user intraocular pressure measurement instruction, it indicates that the user is an old user of this intraocular pressure measurement device (the user's personal data has been stored in this intraocular pressure measurement device), that is, when performing this intraocular pressure measurement, it is only necessary to call the user's personal data from this intraocular pressure measurement device.

[0071] Step S13: If the instruction parsing result is a new user intraocular pressure measurement instruction, the user's personal data is input based on the intraocular pressure measurement instruction to prepare for intraocular pressure measurement, and the prepared intraocular pressure measurement device is obtained.

[0072] It should be noted that, in the specific implementation, when the instruction parsing result is a new user intraocular pressure measurement instruction, it indicates that the user is a new user of this intraocular pressure measurement device (the user's personal data is not stored in this intraocular pressure measurement device), that is, when performing this intraocular pressure measurement, the user's personal data needs to be entered into this intraocular pressure measurement device to establish the user's personal account.

[0073] This embodiment first obtains the user's personal data according to the instruction parsing result, and then calls or enters the user's personal data based on the instruction parsing result to obtain the corresponding prepared intraocular pressure measurement device, which provides a basis for subsequent intraocular pressure measurement.

[0074] refer to Figure 4 , Figure 4 This is a flow chart of a third embodiment of a contact-type intraocular pressure measurement method based on data learning according to the present invention.

[0075] Based on the above first embodiment, in this embodiment, step S30 specifically includes:

[0076] Step S31: determining an intraocular pressure calculation model based on the prepared intraocular pressure measurement device;

[0077] It should be noted that in the specific implementation, the prepared intraocular pressure measurement device needs to calculate the user's intraocular pressure through the built-in intraocular pressure calculation algorithm model, and due to the different intraocular pressure calculation algorithm models, the intraocular pressure calculation process is also different. Therefore, it is necessary to identify and confirm the intraocular pressure calculation model in the prepared intraocular pressure measurement device, thereby providing a basis for subsequent intraocular pressure calculation.

[0078] Step S32: If the intraocular pressure calculation model is an intraocular pressure linear calculation model, input the user's intraocular pressure data into the intraocular pressure linear calculation model to perform intraocular pressure calculation to obtain the user's intraocular pressure value;

[0079] It should be noted that, in a specific implementation, since the user's intraocular pressure value and the probe pressure are in a linear relationship, the intraocular pressure can be calculated using the constructed intraocular pressure linear calculation model and the user's intraocular pressure data to obtain the user's intraocular pressure value.

[0080] Step S33: If the intraocular pressure calculation model is a nonlinear intraocular pressure calculation model, the user's intraocular pressure data is input into the nonlinear intraocular pressure calculation model to perform intraocular pressure calculation to obtain the user's intraocular pressure value.

[0081] It should be noted that in the specific implementation, since the user's intraocular pressure and the depth of the sensor in the probe at different advancement positions have a nonlinear relationship, the intraocular pressure can be calculated through the constructed intraocular pressure nonlinear calculation model and the user's intraocular pressure data to obtain the user's intraocular pressure value.

[0082] This embodiment first determines the intraocular pressure calculation model based on the prepared intraocular pressure measurement equipment, and then inputs the user's intraocular pressure data into the corresponding calculation model to calculate the intraocular pressure, obtain the user's intraocular pressure value, and realize the measurement of the user's intraocular pressure.

[0083] Furthermore, if the intraocular pressure calculation model is an intraocular pressure linear calculation model, the user intraocular pressure data is input into the intraocular pressure linear calculation model to perform intraocular pressure calculation to obtain the user intraocular pressure value. The construction process of the intraocular pressure linear calculation model specifically includes: obtaining a set of historical intraocular pressure measurement data; fitting the intraocular pressure linear calculation model coefficients based on the historical intraocular pressure measurement data set to obtain the linear model coefficient results; and constructing a model based on the linear model coefficient results to obtain the intraocular pressure linear calculation model.

[0084] It should be noted that, in a specific implementation, the process of constructing the intraocular pressure linear calculation model may be that the user first inputs a certain number of intraocular pressure values measured using other more accurate tonometers (such as the gold standard GAT), and then uses this product for multiple measurements to obtain the TW intraocular pressure value of this product. Based on the TW intraocular pressure value and GAT intraocular pressure value of this product, the user's personal data slopes a and b can be fitted. After calibrating a and b, the user can measure again to directly obtain the user's intraocular pressure linear calculation formula through data fitting. The specific intraocular pressure linear calculation formula is P=aIOP+b, where p is the probe pressure and IOP is the user's intraocular pressure value.

[0085] Furthermore, if the intraocular pressure calculation model is an intraocular pressure nonlinear calculation model, the user intraocular pressure data is input into the intraocular pressure nonlinear calculation model for intraocular pressure calculation to obtain the user intraocular pressure value. The construction process of the intraocular pressure nonlinear calculation model specifically includes: data preprocessing of the intraocular pressure training sample data set; inputting the processed intraocular pressure training sample data set into the intraocular pressure nonlinear calculation model to be trained for machine learning to obtain the trained intraocular pressure nonlinear calculation model; performing model detection on the trained intraocular pressure nonlinear calculation model, and if the model detection is qualified, outputting the corresponding intraocular pressure nonlinear calculation model; if the model detection is unqualified, re-machine training the unqualified intraocular pressure nonlinear calculation model until its model detection is qualified and the corresponding intraocular pressure nonlinear calculation model is output.

[0086] It should be noted that in the specific implementation, the intraocular pressure training sample data needs to be preprocessed before model training. The process first uses a filtering smoothing curve to remove noise, and then extracts the characteristics of the pressure curve, such as time domain characteristics (peak value, slope), frequency domain characteristics (frequency domain energy distribution) and dynamic characteristics (pressure change rate, steady-state maintenance time). Finally, the data is labeled, and the intraocular pressure value is used as a continuous label, and whether the intraocular pressure is normal or not is used as a discrete label.

[0087] It should also be noted that in the specific implementation, the process of constructing the intraocular pressure nonlinear calculation model can be to first obtain the curve obtained when the probe sensor advances the same distance through multiple measurements, as well as the intraocular pressure values measured using other more accurate tonometers (such as the gold standard GAT), and use them as the data set to train the model (machine learning model or deep learning model), and then continuously update the model parameters through the optimizer, and finally obtain the ideal probe pressure curve and intraocular pressure model (i.e., the intraocular pressure nonlinear calculation model).

[0088] Furthermore, the processed intraocular pressure training sample data set is input into the intraocular pressure nonlinear calculation model to be trained for machine learning to obtain the trained intraocular pressure nonlinear calculation model, which specifically includes: performing feature engineering selection based on the processed intraocular pressure training sample data set to obtain a screened feature data set; performing model selection based on the screened feature data set to obtain the corresponding intraocular pressure nonlinear calculation model to be trained; and inputting the screened feature data set into the intraocular pressure nonlinear calculation model to be trained for machine learning to obtain the trained intraocular pressure nonlinear calculation model.

[0089] It should be noted that, in the specific implementation, feature engineering selection (for screening significant related features and processing high-dimensional features) is performed based on the processed intraocular pressure training sample data set. The process can be to first analyze the Pearson / Spearman correlation coefficient (Pearson correlation coefficient and Spearman rank correlation coefficient) of the data features and the intraocular pressure to screen significant related features. If there are high-dimensional features, it is necessary to use PCA (a linear dimensionality reduction method) or t-SNE (a nonlinear dimensionality reduction method) to process the high-dimensional features, or automatically select key features for dimensionality reduction through L1 regularization (LASSO), and then combine the advice of ophthalmologists to decide whether to retain biomechanically related features (such as corneal parameters, etc.), and finally obtain the screened feature data set.

[0090] It should also be noted that in the specific implementation, after model selection is completed, the training set, validation set, and test set can be divided, and then hyperparameter tuning (grid search, Bayesian optimization) can be performed. If the data is unbalanced, it is necessary to address the data imbalance through sampling, weighted loss functions, etc. Among them, data imbalance specifically refers to a large difference in the number of samples in different intraocular pressure ranges, such as a ratio of the number of samples in different intraocular pressure ranges greater than 5:1 or 10:1.

[0091] Furthermore, model selection is performed based on the filtered feature data set to obtain a corresponding intraocular pressure nonlinear calculation model to be trained, specifically including: performing label recognition based on the filtered feature data set to obtain a corresponding label recognition result; if the label recognition result is a first-category label recognition result, a regression model is selected as the intraocular pressure nonlinear calculation model to be trained; if the label recognition result is a second-category label recognition result, a deep learning model is selected as the intraocular pressure nonlinear calculation model to be trained; if the label recognition result is a third-category label recognition result, a classification model is selected as the intraocular pressure nonlinear calculation model to be trained.

[0092] It should be noted that in the specific implementation, the label recognition results are distinguished based on the label definition. If the intraocular pressure value is used as the label and the data volume is small (when the data volume is less than 10,000, the label recognition result is the first type of label recognition result), it is necessary to select a regression model as the intraocular pressure nonlinear calculation model to be trained, and the model can be linear regression, support vector regression (SVR), random forest regression or XGBoost model, etc.; if the intraocular pressure value is used as the label and the data volume is sufficient (when the data volume is not less than 10,000, the label recognition result is the second type of label recognition result), a deep learning model can be used as the intraocular pressure nonlinear calculation model to be trained, and the model can be a 1D CNN (processing original time series) or LSTM (capturing temporal dependency) model, etc.; if whether the intraocular pressure is normal is used as the label (when the label recognition result is the third type of label recognition result), the model needs to select a classification model as the intraocular pressure nonlinear calculation model to be trained, and the model can be a logistic regression, random forest classifier or gradient boosting tree, etc.

[0093] In addition, an embodiment of the present invention also proposes a computer-readable storage medium storing a computer program, wherein the storage medium stores an intraocular pressure measurement program, and when the intraocular pressure measurement program is executed by a processor, the steps of the contact intraocular pressure measurement method based on data learning as described above are implemented.

[0094] Since the storage medium adopts all the technical solutions of all the above embodiments, it has at least all the beneficial effects brought about by the technical solutions of the above embodiments, which will not be described one by one here.

[0095] Reference Figure 5 , Figure 5 This is a structural block diagram of the first embodiment of the contact-type intraocular pressure measurement device based on data learning of the present invention.

[0096] like Figure 5 As shown, the contact-type intraocular pressure measurement device based on data learning proposed in an embodiment of the present invention includes:

[0097] Measurement preparation module 10: performs intraocular pressure measurement preparation based on the user's personal data and obtains the prepared intraocular pressure measurement device;

[0098] IOP measurement module 20: Places an IOP measurement probe against the user's eyelids to measure and obtain the user's IOP data;

[0099] The intraocular pressure calculation module 30 calculates the intraocular pressure based on the user's intraocular pressure data and the prepared intraocular pressure measurement device to obtain the user's intraocular pressure value.

[0100] This embodiment prepares intraocular pressure measurement based on the user's personal data to obtain a prepared intraocular pressure measurement device; places the intraocular pressure measurement probe against the user's eyelids for measurement to obtain the user's intraocular pressure data; and calculates intraocular pressure based on the user's intraocular pressure data and the prepared intraocular pressure measurement device to obtain the user's intraocular pressure value. This embodiment first prepares intraocular pressure measurement based on the user's personal data, then begins intraocular pressure measurement. Finally, it calculates intraocular pressure based on the obtained user's intraocular pressure data to obtain the user's intraocular pressure value. The entire intraocular pressure measurement process does not require the patient to open their eyes for measurement, reducing the patient's fear of intraocular pressure measurement and the discomfort of the intraocular pressure measurement process, avoiding the risk of contact infection, and reducing the cost of intraocular pressure measurement.

[0101] It should be understood that the above is only an example and does not constitute any limitation to the technical solution of the present invention. In specific applications, those skilled in the art can make settings as needed, and the present invention does not impose any limitation on this.

[0102] It should be noted that the workflow described above is merely illustrative and does not limit the scope of protection of the present invention. In practical applications, technicians in this field can select part or all of it according to actual needs to achieve the purpose of the embodiment scheme, and no limitation is made here.

[0103] In addition, for technical details not fully described in this embodiment, please refer to the data learning-based contact intraocular pressure measurement method provided in any embodiment of the present invention, and will not be repeated here.

[0104] In addition, it should be noted that, in this document, the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, article, or system comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or system. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or system comprising the element.

[0105] The serial numbers of the above embodiments of the present invention are for description only and do not represent the advantages or disadvantages of the embodiments.

[0106] Through the description of the above embodiments, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus the necessary general hardware platform, or of course by hardware, but in many cases the former is a better embodiment. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. The computer software product is stored in a storage medium (such as a read-only memory (ROM) / RAM, a magnetic disk, or an optical disk), and includes a number of instructions for enabling a terminal device (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present invention.

[0107] The above are only preferred embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention description and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A contact intraocular pressure measurement method based on data learning, characterized in that: include: Prepare intraocular pressure measurement based on user personal data and obtain the prepared intraocular pressure measurement device; Place the intraocular pressure measuring probe against the user's eyelid to measure and obtain the user's intraocular pressure data; The intraocular pressure is calculated based on the user's intraocular pressure data and the prepared intraocular pressure measurement device to obtain the user's intraocular pressure value.

2. The contact-type intraocular pressure measurement method based on data learning according to claim 1, characterized in that: Prepare intraocular pressure measurement based on user personal data and obtain the prepared intraocular pressure measurement device, specifically including: Parse the intraocular pressure measurement instruction and obtain the user's personal data based on the instruction parsing result; If the instruction parsing result is a historical user intraocular pressure measurement instruction, calling the user's personal data based on the intraocular pressure measurement instruction to prepare for intraocular pressure measurement and obtain a prepared intraocular pressure measurement device; If the instruction parsing result is a new user intraocular pressure measurement instruction, the user's personal data is entered based on the intraocular pressure measurement instruction to prepare for intraocular pressure measurement, and a prepared intraocular pressure measurement device is obtained.

3. The contact-type intraocular pressure measurement method based on data learning according to claim 1 or 2, characterized in that: Calculating the intraocular pressure based on the user's intraocular pressure data and the prepared intraocular pressure measurement device to obtain the user's intraocular pressure value specifically includes: Determining an intraocular pressure calculation model based on the prepared intraocular pressure measurement device; If the intraocular pressure calculation model is an intraocular pressure linear calculation model, inputting the user's intraocular pressure data into the intraocular pressure linear calculation model to perform intraocular pressure calculation to obtain the user's intraocular pressure value; If the intraocular pressure calculation model is an intraocular pressure nonlinear calculation model, the user's intraocular pressure data is input into the intraocular pressure nonlinear calculation model to perform intraocular pressure calculation to obtain the user's intraocular pressure value.

4. The contact-type intraocular pressure measurement method based on data learning according to claim 3, characterized in that: If the intraocular pressure calculation model is an intraocular pressure linear calculation model, the user's intraocular pressure data is input into the intraocular pressure linear calculation model to perform intraocular pressure calculation to obtain the user's intraocular pressure value. The construction process of the intraocular pressure linear calculation model specifically includes: Obtain historical intraocular pressure measurement data set; Performing intraocular pressure linear calculation model coefficient fitting based on the historical intraocular pressure measurement data set to obtain linear model coefficient results; A model is constructed based on the linear model coefficient results to obtain an intraocular pressure linear calculation model.

5. The contact-type intraocular pressure measurement method based on data learning according to claim 3, characterized in that: If the intraocular pressure calculation model is a nonlinear intraocular pressure calculation model, the user's intraocular pressure data is input into the nonlinear intraocular pressure calculation model to perform intraocular pressure calculation to obtain the user's intraocular pressure value. The construction process of the nonlinear intraocular pressure calculation model specifically includes: Perform data preprocessing on the intraocular pressure training sample data set; Inputting the processed intraocular pressure training sample data set into the intraocular pressure nonlinear calculation model to be trained for machine learning to obtain a trained intraocular pressure nonlinear calculation model; Performing a model test on the trained intraocular pressure nonlinear calculation model, and outputting a corresponding intraocular pressure nonlinear calculation model if the model test is qualified; If the model fails the test, the unqualified intraocular pressure nonlinear calculation model will be retrained by the machine until the model passes the test and the corresponding intraocular pressure nonlinear calculation model is output.

6. The contact-type intraocular pressure measurement method based on data learning according to claim 5, characterized in that: The processed intraocular pressure training sample data set is input into the intraocular pressure nonlinear calculation model to be trained for machine learning to obtain the trained intraocular pressure nonlinear calculation model, specifically including: Performing feature engineering selection based on the processed intraocular pressure training sample data set to obtain a filtered feature data set; Performing model selection based on the filtered feature data set to obtain a corresponding intraocular pressure nonlinear calculation model to be trained; The filtered feature data set is input into the intraocular pressure nonlinear calculation model to be trained for machine learning to obtain the trained intraocular pressure nonlinear calculation model.

7. The contact-type intraocular pressure measurement method based on data learning according to claim 6, characterized in that: Model selection is performed based on the filtered feature data set to obtain a corresponding nonlinear intraocular pressure calculation model to be trained, specifically including: Performing label recognition based on the filtered feature data set to obtain corresponding label recognition results; If the label recognition result is a first-category label recognition result, the regression model is selected as the intraocular pressure nonlinear calculation model to be trained; If the label recognition result is a second type of label recognition result, selecting a deep learning model as the intraocular pressure nonlinear calculation model to be trained; If the label recognition result is a third-category label recognition result, the classification model is selected as the intraocular pressure nonlinear calculation model to be trained.

8. An intraocular pressure measuring device, characterized in that: The intraocular pressure measuring device comprises: Measurement preparation module: prepares intraocular pressure measurement based on user personal data and obtains the prepared intraocular pressure measurement device; Intraocular pressure measurement module: Place the intraocular pressure measurement probe against the user's eyelid to measure and obtain the user's intraocular pressure data; An intraocular pressure calculation module calculates intraocular pressure based on the user's intraocular pressure data and the prepared intraocular pressure measurement device to obtain the user's intraocular pressure value.

9. A contact-type intraocular pressure measurement device based on data learning, characterized in that: The data learning-based contact intraocular pressure measurement device includes: a memory, a processor, and an intraocular pressure measurement program stored in the memory and executable on the processor, wherein the intraocular pressure measurement program is configured to implement the data learning-based contact intraocular pressure measurement method according to any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the contact intraocular pressure measurement method based on data learning according to any one of claims 1 to 7 can be implemented.

Images