Real-time dynamic tonometer

Abstract

The present application relates to the technical field of tonometer, in particular to a real-time dynamic tonometer, comprising a signal acquisition module, a signal processing module, an algorithm model module and a display module, the signal acquisition module and the signal processing module are in wireless communication through Bluetooth, the display module is used for displaying real-time intraocular pressure signal data, and the signal acquisition module comprises a plurality of flexible patches.In the present application, the flexible patch is provided with a three-axis acceleration sensor and a surface electromyography sensor, the three-axis acceleration sensor and the surface electromyography sensor are used for detecting eyelid tremor signals, the signal processing module is used for processing the tremor signals, the algorithm model module is used for calculating the intraocular pressure value, the measurement can be realized without contacting the eye membrane, the discomfort during measurement can be reduced, the tonometer is not affected by the corneal biomechanical characteristics, is suitable for patients with corneal diseases, and the user can perform measurement by himself.

Description

Technical Field

[0001] This invention relates to the field of tonometer technology, specifically to a real-time dynamic tonometer. Background Technology

[0002] Intraocular pressure measurement is an important indicator for assessing eye health and plays a key role in the diagnosis and treatment of eye diseases such as glaucoma. Based on different measurement principles and operating methods, modern tonometers can be mainly divided into two categories: contact and non-contact, each of which includes a variety of specific technical forms.

[0003] Contact tonometers measure intraocular pressure by directly or indirectly contacting the cornea. These devices typically require physical contact with the patient's cornea during measurement. Based on different working principles, they can be further subdivided into applanation, indentation, and rebound types. The advantage of contact tonometers is that the measurement results are relatively accurate and reliable. Non-contact tonometers, on the other hand, measure intraocular pressure indirectly through airflow or other non-contact methods, avoiding direct contact with the cornea. The biggest advantage of these devices is that they are easy to operate, require no anesthesia, and have no risk of infection, making them particularly suitable for large-scale screening and daily monitoring.

[0004] However, both methods are instantaneous measurements and cannot capture 24-hour intraocular pressure fluctuations. Furthermore, both are affected by corneal biomechanical characteristics, such as hardness and thickness, and require comprehensive evaluation in conjunction with other examinations. Additionally, some patients with corneal diseases cannot use them, such as those with severe dry eye or active inflammation. Contact tonometers are not suitable for these patients, while non-contact tonometers are not suitable for those with uneven corneas or severe ocular surface diseases. Summary of the Invention

[0005] The purpose of this invention is to provide a real-time dynamic tonometer to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] The real-time dynamic tonometer includes a signal acquisition module, a signal processing module, an algorithm model module, and a display module. The signal acquisition module and the signal processing module communicate wirelessly via Bluetooth. The display module is used to display real-time intraocular pressure signal data. The signal acquisition module includes several flexible patches, each equipped with a triaxial accelerometer and a surface electromyography sensor. The flexible patches are equipped with a positioning mechanism, and the flexible patches are equipped with eyeglass frames. A Bluetooth transmitter and a power supply are also located on the flexible patches.

[0008] The triaxial accelerometer can detect eyelid tremors, and the surface electromyography (EMG) sensor can collect the electrical activity of the orbicularis oculi muscle. The triaxial accelerometer and the EMG sensor can transmit the collected signal data to the signal processing module via Bluetooth communication. The signal processing module can extract the intraocular pressure tremor characteristics and transmit the intraocular pressure tremor signal data to the algorithm model module. The algorithm model module can output an estimated intraocular pressure value based on the calibration relationship between the tremor signal and intraocular pressure. The algorithm model module can transmit the intraocular pressure value to the display module for display.

[0009] Furthermore, the flexible patch is provided with a sponge block, which is adhered to the flexible patch;

[0010] The positioning mechanism includes a positioning frame, several plug-in sockets, several locking slots, several return springs, and several sliding slots;

[0011] The positioning frame is set on the sponge block;

[0012] Several plug-in sockets are set on the positioning frame;

[0013] Several slots are respectively provided on the socket;

[0014] Several return springs are fixedly connected to the plug socket;

[0015] Several sliding grooves are formed on the positioning frame, the plug-in seat is slidably connected to the adjacent sliding groove, and the return spring is fixedly connected to the adjacent sliding groove.

[0016] Furthermore, the connector is fixedly connected with a stop, and the positioning mechanism is equipped with an elastic band, which is fixedly connected with Velcro 1 and Velcro 2.

[0017] Furthermore, when the signal processing module processes the signal data, it uses a bandpass filter from 0.5Hz to 35Hz to eliminate baseline drift and slow body movements through low-frequency cutoff and suppress high-frequency electronic noise through high-frequency cutoff.

[0018] Furthermore, during the signal processing process, the signal processing module performs PCA analysis on the triaxial acceleration data to extract the principal axis direction with the greatest vibration energy, and uses the data in this direction as the main flutter signal for analysis.

[0019] Preferably, the signal processing module can use amplitude thresholding and pattern recognition algorithms to automatically identify and remove invalid data segments such as large-amplitude active blinking and sudden head turning.

[0020] Preferably, the signal processing module can extract frequency domain features, time domain features, and nonlinear features from the processed signal through feature engineering, thereby capturing key information related to intraocular pressure in multiple dimensions. The frequency domain features include the dominant frequency, spectral energy distribution, and spectral entropy features; the time domain features include the root mean square and zero crossover rate features; and the nonlinear features are mainly the entropy features of the signal.

[0021] Preferably, in the process of establishing the calibration relationship between tremor frequency and intraocular pressure in the algorithm model module, standard data collection of patient intraocular pressure is carried out. While the signal acquisition module collects eyelid tremor data, a flattening tonometer is used to simultaneously measure and record the real, calibrated intraocular pressure value, forming a paired dataset of features and real intraocular pressure. Based on the importance ranking of the tree model, the feature subset with the most relevant intraocular pressure and the least redundancy is selected from all the features extracted above.

[0022] Furthermore, in the process of establishing a paired dataset of features and real intraocular pressure in the algorithm model module, multiple linear regression is used as the starting model, and a more complex gradient boosting tree is used to capture the potential nonlinear relationship between features and intraocular pressure. By dividing the dataset into training set and test set, the model is trained with training set data to optimize its parameters.

[0023] Preferably, the algorithm model module builds a model based on large sample training data. The model is adjusted by combining the baseline value and the user's physiological parameters in the first use, and the deviation is dynamically corrected in the long term. The general model is fine-tuned according to personal data.

[0024] Compared with the prior art, the beneficial effects of the present invention are:

[0025] The device incorporates a triaxial accelerometer and a surface electromyography (EMG) sensor in a flexible patch. By attaching the patch to the eyelid, these sensors detect eyelid tremor signals. A signal processing module processes these signals and extracts key information related to intraocular pressure (IOP). An algorithmic model module then calculates IOP based on a multivariate regression model establishing a calibration relationship between the tremor signal and IOP. Finally, a display module shows the IOP value in real time. This method allows for measurement without contact with the cornea, reducing the risk of infection and discomfort. It is unaffected by corneal biomechanical properties and suitable for patients with certain corneal diseases. No professional operation is required; users can perform measurements themselves. It is suitable for home monitoring, large-scale population screening, and continuous postoperative monitoring. Users can close their eyes and rest while wearing the patch, and it continuously records IOP changes, such as diurnal fluctuations, facilitating early detection of nighttime IOP peaks.

[0026] The connector has a slotted groove. Under the action of the return spring, the slot on the positioning frame can be locked onto the eyeglass frame. The positioning mechanism then connects the flexible patch and the sponge block to the eyeglass frame. This way, after wearing the eyeglass frame, the flexible patch can be aligned with the patient's eyelid, making it relatively convenient to wear. The position of the positioning mechanism on the eyeglass frame can be adjusted according to the patient's interpupillary distance to accommodate patients with different interpupillary distances. Alternatively, an elastic band can be passed between the stop and the positioning frame and tied to the head to position the flexible patch, making it easier for the patient to rest on their side at night. Attached Figure Description

[0027] Figure 1 This is an overall block diagram of the real-time dynamic tonometer of the present invention;

[0028] Figure 2 This is a schematic diagram of the flexible patch structure in this invention;

[0029] Figure 3 This is a schematic diagram of the positioning mechanism structure in this invention;

[0030] Figure 4 This is a schematic diagram of the internal structure of the positioning mechanism in this invention;

[0031] Figure 5 This is a schematic diagram of the elastic band structure in this invention;

[0032] Figure 6 This is a schematic diagram of the signal receiving master controller in this invention.

[0033] In the diagram: 10, flexible patch; 20, signal receiving controller; 30, eyeglass frame; 40, positioning mechanism; 41, positioning frame; 42, connector; 43, slot; 44, stop block; 45, return spring; 46, sliding groove; 50, sponge block; 60, elastic band; 61, Velcro 1; 62, Velcro 2. Detailed Implementation

[0034] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0035] Please see Figure 1-6In this embodiment of the invention, the real-time dynamic tonometer includes a signal acquisition module, a signal processing module, an algorithm model module, and a display module. The signal acquisition module and the signal processing module communicate wirelessly via Bluetooth. The display module is used to display real-time intraocular pressure signal data. The signal acquisition module includes several flexible patches 10. Each flexible patch 10 is equipped with a triaxial accelerometer and a surface electromyography (EMG) sensor. Each flexible patch 10 is equipped with a positioning mechanism 40. Each of the flexible patches 10 is equipped with an eyeglass frame 30. Each flexible patch 10 is equipped with a Bluetooth transmitter and a power supply. The Bluetooth transmitter is used to send detection signals, and the power supply is used to power the triaxial accelerometer, the EMG sensor, and the Bluetooth transmitter. Each flexible patch 10 is equipped with a sponge block 50, which is attached to the flexible patch 10.

[0036] The triaxial accelerometer can detect eyelid tremors, and the surface electromyography (EMG) sensor can collect the electrical activity of the orbicularis oculi muscle. The triaxial accelerometer and the EMG sensor can transmit the collected signal data to the signal processing module via Bluetooth communication. The signal processing module can extract the intraocular pressure tremor features and transmit the intraocular pressure tremor signal data to the algorithm model module. The algorithm model module can build a model based on a large sample training data and output an estimated intraocular pressure value based on the calibration relationship between the tremor signal and intraocular pressure. The algorithm model module can transmit the intraocular pressure value to the display module for display.

[0037] Specifically, the signal processing module, algorithm model module, and display module can all be integrated into the signal receiving main controller 20. A Bluetooth receiver is set in the signal receiving main controller 20, and the signal acquisition module transmits the signal data to the signal processing module in the signal receiving main controller 20 through Bluetooth communication. The signal processing module performs signal processing and analysis, and the algorithm model module runs the model. The display module displays real-time and historical intraocular pressure trend graphs, provides abnormal value alarms, and generates health reports. The signal receiving main controller 20 can transmit data to the cloud platform through the Internet of Things. The cloud platform stores long-term user data and continuously optimizes and iterates the core algorithm model using larger-scale group data. At the same time, it provides doctors with a remote data access interface so that doctors can view the monitoring data in a timely manner.

[0038] The triaxial accelerometer employs a high-sensitivity, low-noise MEMS sensor to acquire raw triaxial acceleration and angular velocity data at a sampling rate of at least 100Hz, ensuring the capture of tremor signals from 5 to 30 Hz. When the flexible patch 10 is attached to the surface of the eyelid skin, the triaxial accelerometer records the amplitude (acceleration value), frequency, and waveform (such as sine wave and pulse wave) of the vibration to reflect the mechanical characteristics of the vibration. The surface electromyography (sEMG) sensor can simultaneously acquire the electrical activity of the orbicularis oculi muscle, obtain the amplitude and frequency of the electromyographic signal, correlate it with the muscle contraction intensity, exclude non-muscle-related vibrations such as external touch, and achieve a comprehensive characterization of eyelid vibration, providing a rich data foundation for accurate measurement of intraocular pressure.

[0039] By collecting signals through sensors and combining signal processing algorithms and machine learning models, the accuracy of measurements can be improved, thereby ensuring that the error range meets clinical needs as much as possible. The measurement process does not require contact with the cornea, which helps reduce the risk of infection and discomfort during measurement. It is also unaffected by the biomechanical properties of the cornea and is suitable for some patients with corneal diseases. No professional operation is required; users can wear it themselves for measurement. It is suitable for home monitoring, large-scale population screening, and continuous postoperative monitoring. Users can close their eyes and rest after wearing it, and it can continuously record changes in intraocular pressure, such as diurnal fluctuations, which is helpful for the early detection of nighttime intraocular pressure peaks.

[0040] Ten minutes before data collection, avoid strenuous exercise and frequent blinking to reduce the impact of eyelid muscle fatigue on the signal. During the measurement process, users should remain seated with their eyes closed to avoid eye movement and facial muscle activity.

[0041] Example 1

[0042] like Figure 2-4 As shown, in this embodiment, the positioning mechanism 40 includes a positioning frame 41, a plurality of plug-in seats 42, a plurality of locking slots 43, a plurality of reset springs 45, and a plurality of sliding slots 46.

[0043] The positioning frame 41 is set on the sponge block 50, a number of plug-in seats 42 are set on the positioning frame 41, a number of locking slots 43 are respectively opened on the plug-in seats 42, a number of return springs 45 are respectively fixedly connected to the plug-in seats 42, and a number of sliding slots 46 are all opened on the positioning frame 41. The plug-in seats 42 are slidably connected to the adjacent sliding slots 46, and the return springs 45 are fixedly connected to the adjacent sliding slots 46. The plug-in seats 42 can be connected to the positioning frame 41 through the return springs 45.

[0044] In practice, the two connectors 42 can be moved back to back so that the locking slot 43 corresponds to the eyeglass frame 30. After the connectors 42 are released, the locking slot 43 on the positioning frame 41 can be locked onto the eyeglass frame 30 under the action of the return spring 45. Thus, the flexible patch 10 and the sponge block 50 are connected to the eyeglass frame 30 through the positioning mechanism 40. In this way, after wearing the eyeglass frame 30, the flexible patch 10 can be aligned with the patient's eyelid, making it relatively convenient to wear. Moreover, the position of the positioning mechanism 40 on the eyeglass frame 30 can be adjusted according to the patient's interocular distance to accommodate patients with different interocular distances.

[0045] like Figure 1 As shown, in this embodiment, when the signal processing module processes the signal data, it uses a bandpass filter from 0.5Hz to 35Hz to eliminate baseline drift and slow body movements through low-frequency cutoff and suppress high-frequency electronic noise through high-frequency cutoff. During the signal data processing process, the signal processing module performs PCA analysis on the triaxial acceleration data to extract the principal axis direction with the greatest vibration energy. The data in this direction is used as the main vibration signal for analysis to reduce the influence of the randomness of the sensor orientation. The signal processing module can use amplitude thresholding and pattern recognition algorithms to automatically identify and remove invalid data segments such as large-amplitude active blinking and sudden head rotation.

[0046] In practice, a bandpass filter ranging from 0.5Hz to 35Hz is used to retain the main frequency range of eyelid vibration (blink frequency of approximately 0.5-2Hz) and filter low frequencies to eliminate baseline drift, slow body movements, and respiratory interference noise of 0.2-0.5Hz. At the same time, high-frequency electromagnetic interference noise greater than 100Hz is filtered out. The signal processing module can introduce independent component analysis (ICA) to separate the electromyographic signal of the orbicularis oculi muscle from the mixed signal and the electrical signals of other muscles (such as the temporalis muscle), eliminating interference from non-target muscles. By using bandpass filtering and independent component analysis techniques, interference signals such as breathing and other facial muscle activities are effectively filtered out, and the orbicularis oculi muscle signal related to intraocular pressure is separated more accurately, improving the purity and reliability of the signal.

[0047] like Figure 1 As shown in this embodiment, the signal processing module can extract frequency domain features, time domain features, and nonlinear features from the processed signal through feature engineering, thereby capturing key information related to intraocular pressure in multiple dimensions. The frequency domain features include the dominant frequency, spectral energy distribution, and spectral entropy features; the time domain features include the root mean square and zero crossover rate features; and the nonlinear features are mainly the entropy features of the signal.

[0048] In practice, by performing a fast Fourier transform on the effective signal segment, the peak frequency of the power spectral density in the range of five to thirty hertz is found, the main frequency is obtained, and by calculating the weighted average frequency of the spectrum within the tremor frequency band, the concentration point of tremor energy is reflected, and thus the spectral energy distribution is obtained. The spectral entropy can characterize the regularity and complexity of the tremor frequency distribution and is related to the stability of intraocular pressure.

[0049] The root mean square value of the time-domain characteristics can characterize the overall energy or amplitude of the tremor. The number of times the signal crosses the zero point per unit time is called the zero crossover rate, which is a simple time-domain estimate of the frequency.

[0050] Sample entropy, or approximate entropy, can characterize the complexity and unpredictability of quantified signal sequences and is used to capture subtle changes in neural control patterns caused by changes in intraocular pressure.

[0051] like Figure 1 As shown in this embodiment, during the process of establishing the calibration relationship between tremor frequency and intraocular pressure in the algorithm model module, standard data collection of patient intraocular pressure is performed. While the signal acquisition module collects eyelid tremor data, a flattening tonometer is used to simultaneously measure and record the real, calibrated intraocular pressure value, forming a paired dataset of features and real intraocular pressure. Based on the importance ranking of the tree model, the subset of features most relevant to intraocular pressure and with the least redundancy is selected from all the features extracted above. In the process of establishing the paired dataset of features and real intraocular pressure in the algorithm model module, multiple linear regression is used as the starting model, and a more complex gradient boosting tree is used to capture the potential nonlinear relationship between features and intraocular pressure. By dividing the dataset into a training set and a test set, the model is trained with the training set data to optimize its parameters and establish a mapping relationship between eyelid tremor features and intraocular pressure values.

[0052] In practice, the Goldmann tonometer or dynamic profilometry is used to measure the real, calibrated gold standard intraocular pressure value. Multiple linear regression, support vector regression or random forest regression can be used as the starting model, and gradient boosting tree or shallow neural network can be used as the advanced model. The dataset is divided into training set and test set. The model is trained using the training set data. The goal is to minimize the root mean square error between the predicted intraocular pressure and the real intraocular pressure.

[0053] On a test set independent of the training set, the consistency between the predicted values ​​of the algorithm model module and the Goldmann measurement values ​​is evaluated. Statistical indicators such as intragroup correlation coefficient and 95% consistency limit of Bland-Altman plot are calculated to verify accuracy. Furthermore, the product's performance under different body positions (sitting, lying down), different activity states (resting, slight activity), and different user groups (different ages, different corneal thicknesses) can be tested to conduct robustness testing.

[0054] Example 2

[0055] Based on Example 1, such as Figure 5 As shown, in this embodiment, the plug-in socket 42 is fixedly connected to the stop block 44, and the positioning mechanism 40 is equipped with an elastic band 60, which is fixedly connected to Velcro 61 and Velcro 62.

[0056] In practice, the positioning mechanism 40 can be removed from the eyeglass frame 30, and the elastic band 60 can be passed between the stop block 44 and the positioning frame 41. By binding the elastic band 60 to the head, the Velcro 61 and Velcro 62 can be bonded to each other, and the flexible patch 10 can be positioned by the elastic band 60, which makes it easier for the patient to rest on their side at night.

[0057] like Figure 1 As shown, in this embodiment, the algorithm model module builds a model based on a large sample training data. The model is initially adjusted by combining the benchmark value and the user's physiological parameters. In the long term, the deviation is dynamically corrected, and the general model is fine-tuned based on individual data.

[0058] In practice, a multivariate regression model or deep learning model is built based on a large sample training data, and personalized calibration is performed. The model is adjusted by combining the gold standard intraocular pressure value and the user's physiological parameters in the first use, and the deviation is dynamically corrected in the long term, which helps to improve the universality of the model and make it suitable for different groups of people.

[0059] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

[0060] Furthermore, it should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This narrative style is merely for clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

Claims

1. A real-time dynamic tonometer, comprising a signal acquisition module, a signal processing module, an algorithm model module, and a display module, wherein the signal acquisition module and the signal processing module communicate wirelessly via Bluetooth, and the display module is used to display real-time intraocular pressure signal data, characterized in that, The signal acquisition module includes several flexible patches (10), each flexible patch (10) is equipped with a triaxial accelerometer and a surface electromyography sensor, each flexible patch (10) is equipped with a positioning mechanism (40), each flexible patch (10) is equipped with an eyeglass frame (30), and each flexible patch (10) is attached to the surface of the eyelid skin. A triaxial accelerometer records the amplitude, frequency, and waveform of vibrations to reflect the mechanical characteristics of eyelid vibrations. A surface electromyography (EMG) sensor simultaneously collects the electrical activity of the orbicularis oculi muscle, acquiring the amplitude and frequency of EMG signals, correlating them with muscle contraction intensity, and excluding non-muscle-related vibrations. The triaxial accelerometer and EMG sensor transmit the collected signal data to a signal processing module via Bluetooth. The signal processing module extracts intraocular pressure tremor characteristics and transmits the tremor signal data to an algorithm model module. During signal data processing, the signal processing module performs PCA analysis on the triaxial acceleration data to extract the principal axis direction with the greatest vibration energy, using the data in this direction as the primary tremor signal for analysis. The algorithm model module outputs an estimated intraocular pressure value based on the calibration relationship between the tremor signal and intraocular pressure, and transmits the intraocular pressure value to a display module for display.

2. The real-time dynamic tonometer according to claim 1, characterized in that, The flexible patch (10) is provided with a sponge block (50); The positioning mechanism (40) includes: The positioning frame (41) is set on the sponge block (50); Several plug-in sockets (42) are set on the positioning frame (41); Several slots (43) are respectively opened on the socket (42); Several return springs (45) are fixedly connected to the plug socket (42); Several sliding grooves (46) are provided on the positioning frame (41), the plug-in seat (42) is slidably connected to the adjacent sliding groove (46), and the reset spring (45) is fixedly connected to the adjacent sliding groove (46).

3. The real-time dynamic tonometer according to claim 2, characterized in that, The plug-in socket (42) is fixedly connected to a stop block (44), and the positioning mechanism (40) is equipped with an elastic band (60). The elastic band (60) is fixedly connected to Velcro 1 (61) and Velcro 2 (62).

4. The real-time dynamic tonometer according to any one of claims 1-3, characterized in that, When processing signal data, the signal processing module uses a bandpass filter from 0.5Hz to 35Hz to eliminate baseline drift and slow body movements through low-frequency cutoff and suppress high-frequency electronic noise through high-frequency cutoff.

5. The real-time dynamic tonometer according to claim 4, characterized in that, The signal processing module can use amplitude thresholding and pattern recognition algorithms to automatically identify and remove invalid data segments such as large-amplitude active blinking and sudden head rotation.

6. The real-time dynamic tonometer according to claim 4, characterized in that, The signal processing module can extract frequency domain features, time domain features, and nonlinear features from the processed signal through feature engineering, thereby capturing key information related to intraocular pressure in multiple dimensions. The frequency domain features include the dominant frequency, spectral energy distribution, and spectral entropy features; the time domain features include the root mean square and zero crossover rate features; and the nonlinear features are the entropy features of the signal.

7. The real-time dynamic tonometer according to any one of claims 1-3, characterized in that, In the process of establishing the calibration relationship between tremor frequency and intraocular pressure in the algorithm model module, standard data collection of patient intraocular pressure is carried out. While the signal acquisition module collects eyelid tremor data, a flattening tonometer is used to simultaneously measure and record the real, calibrated intraocular pressure value, forming a paired dataset of features and real intraocular pressure. Based on the importance ranking of the tree model, the subset of features that are most relevant to intraocular pressure and have the least redundancy is selected from all the extracted features.

8. The real-time dynamic tonometer according to claim 7, characterized in that, In the process of establishing the paired dataset of features and true intraocular pressure, the algorithm model module uses multiple linear regression as the starting model and gradient boosting tree to capture the potential nonlinear relationship between features and intraocular pressure. By dividing the dataset into training set and test set, the model is trained and the parameters are optimized using the training set data.

9. The real-time dynamic tonometer according to claim 7, characterized in that, The algorithm model module is built based on large sample training data. The model is initially adjusted by combining the benchmark value and the user's physiological parameters. In the long term, the deviation is dynamically corrected and the model is fine-tuned according to individual data.

Images