Intraocular pressure measuring device and measuring method

Abstract

The invention relates to the technical field of medical instruments, in particular to an intraocular pressure measuring device and method, and the method comprises the following steps: recognizing and segmenting an interference fringe region contained in a picture based on a YOLOv11-segment model; performing binarization segmentation on the interference fringe region based on a deep convolutional neural network and extracting the binarization segmented interference fringe region; extracting a skeleton of the binarized interference fringes, marking the level of each fringe, and further calculating the center deflection of the film of the Fabry-Perot microcavity after the film is pressed; and obtaining intraocular pressure according to the center deflection. Square interference fringe areas contained in a picture are accurately recognized, segmented and extracted on the basis of a YOLOv11-seg model, skeletons of the fringes are extracted after binaryzation, the level of each fringe is marked, then the center deflection of the square thin film after being pressed is calculated, the center deflection corresponds to pressure values borne by a sensor calibrated through a calibration device in a one-to-one mode, and the precision of the square thin film is improved. A shot picture containing an interference fringe image is converted into a pressure value.

Description

Technical Field

[0001] The present invention relates to the technical field of medical devices, and in particular to an intraocular pressure measuring device and a measuring method. Background Art

[0002] Glaucoma is one of the three major causes of blindness and carries significant risks. High intraocular pressure (IOP) is considered a significant risk factor for glaucoma. Therefore, IOP is a crucial indicator for determining glaucoma treatment targets and assessing treatment efficacy and prognosis in clinical practice.

[0003] Currently, the primary method for measuring intraocular pressure is to measure the patient's immediate intraocular pressure using instruments, including applanation tonometers and air jet tonometers. Applanation tonometers are complex to measure, requiring topical anesthesia before measurement, applying sodium fluorescein to the cornea during measurement, and having the measured value affected by central corneal thickness. Compared to applanation tonometers, air jet tonometers simplify the intraocular pressure measurement process and do not require topical anesthesia or sodium fluorescein. However, air jet tonometers also have several issues: their impact airflow can cause eye discomfort, and the instrument is expensive and unportable.

[0004] In addition, there are many studies on miniature implantable intraocular pressure sensors based on different principles. The common characteristics of these studies are: 1. The sensor is separated from the detection equipment, and the sensing method is non-contact; 2. The sensor area and volume are very small, ranging from hundreds of microns to several millimeters; 3. The sensor is in contact with the eye structure and is attached to the eyeball or implanted inside the eyeball.

[0005] These implantable intraocular pressure sensors are categorized by sensing principle into three main types: electrical, microfluidic, and optical. Chen et al. designed a contact lens-based intraocular pressure sensor based on capacitance-pressure sensitivity. The frequency of the LC oscillator formed by the capacitor and inductor changes with pressure, and the reading device is a large network analyzer. Agaoglu et al. used a microfluidic chip to measure intraocular pressure. An artificial lens with an integrated microfluidic chip was implanted into the eye using cataract surgery. Fluctuations in intraocular pressure cause displacement of the liquid-air interface of the artificial lens, and monitoring this interface position provides the intraocular pressure reading. Electrical sensing is limited by circuit structure and materials, making it difficult to achieve submillimeter dimensions, and the reading device is bulky and expensive. Microfluidic sensing is limited by its stringent airtightness requirements and the indirect sensing principle of photographic reading, which presents a bottleneck for miniaturization. Optical sensors are generally smaller than electrical and microfluidic sensors. Therefore, MEMS intraocular pressure sensors based on optical sensors have become a major research direction for implantable sensors.

[0006] Current measurement schemes for MEMS intraocular pressure sensors based on optical sensing are all based on the Fabry–Pérot interferometer structure. By capturing the interference fringe pattern, the interference area is identified and extracted, and the interference fringe pattern is converted into pressure. Among them, the existing interference area identification and extraction methods usually require manual image cropping. Manual image cropping makes this method difficult to apply to real-time continuous demodulation processes, limiting its application in intraocular pressure measurement. Summary of the Invention

[0007] In order to solve the problem that the interference area identification and extraction method applied to MEMS intraocular pressure sensors in the prior art is difficult to apply to the real-time continuous demodulation process, the present invention provides an intraocular pressure measurement method. In this measurement method, the interference fringe area contained in the photo is accurately identified, segmented and extracted by introducing the YOLOv11-seg model, so that the measurement method can realize the identification and extraction of the interference area under the premise of avoiding manual image cropping, thereby solving the problem that the interference area identification and extraction method in the prior art is usually difficult to apply to the real-time continuous demodulation process due to manual image cropping.

[0008] The technical solution adopted by the present invention to solve its technical problem is:

[0009] A method for measuring intraocular pressure comprises the following steps:

[0010] S1: Sensing intraocular pressure changes through the Fabry-Perot microcavity on the intraocular pressure sensor;

[0011] S2: emitting light to the Fabry-Perot microcavity through a photographing module, and obtaining a photograph of the interference pattern through a photographing element;

[0012] S3: Identify and segment the interference fringe region contained in the photo based on the YOLOv11-segment model to obtain the interference fringe region;

[0013] S4: Binarizing and segmenting the interference fringe region based on a deep convolutional neural network and extracting the interference fringe region to obtain binary interference fringes;

[0014] S5: extracting the skeleton of the binary interference fringes, marking the order of each fringes, and then calculating the central deflection of the film of the Fabry-Perot microcavity after being compressed;

[0015] S6: Obtain intraocular pressure based on the central deflection.

[0016] Optionally, identifying and segmenting the interference fringe region contained in the photo based on a YOLOv11-segment model includes:

[0017] S31: Acquire photos of interference patterns under multiple different environmental conditions;

[0018] S32: manually labeling the interference fringe areas in the photo using the Labelme plug-in to obtain a txt format file as a label set corresponding to the training set image;

[0019] S33: Using part of the images as a training set and part of the images as a validation set, the model is trained to obtain the interference fringe area.

[0020] Optionally, the deep convolutional neural network includes a left path, a right path, an encoding path and a decoding path.

[0021] Optionally, the loss function used in the model training of the deep convolutional neural network includes a focal loss function and a multi-scale structural similarity metric.

[0022] Optionally, the focus loss function is defined as follows:

[0023] L FL (p, y) = L FL (p t )=-0.5(1-p t ) r log(p t );

[0024] in,

[0025] y∈{1,0} represents the true label value of the sample, p∈[0,1] is the output of the model, that is, the probability that the true label of the predicted sample is 1; r is the adjustment factor in the focal loss function.

[0026] Optionally, the multi-scale structural similarity measure is defined as follows:

[0027]

[0028] Where M is the number of scales, α M ,β j ,γ j l M (x,y),c j (x,y),s j The weight of (x,y), l M (x,y),c j (x,y),s j (x,y) are the similarities of x and y in terms of brightness, contrast and structure at scale M,i,j respectively.

[0029] Optionally, the deflection distribution function of the Fabry-Perot microcavity film after being compressed is as follows:

[0030]

[0031] Where (x, y) is the coordinate of the point on the diaphragm, the center of the diaphragm is the coordinate origin, w0 is the central deflection of the diaphragm, α is the side length of the diaphragm, and c1 and c2 are two empirical parameters.

[0032] Optionally, obtaining the intraocular pressure according to the central deflection includes: calibrating the central deflection of the Fabry-Perot microcavity film at different pressure values using a calibration device, and then obtaining the intraocular pressure according to the calculated central deflection.

[0033] Another object of the present invention is to provide an intraocular pressure measuring device for measuring intraocular pressure using the intraocular pressure measuring method described above.

[0034] Optionally, it includes an intraocular pressure sensor, a shooting module and a shooting element; wherein,

[0035] The intraocular pressure sensor is provided with a Fabry-Perot microcavity;

[0036] The shooting module includes a housing, an optical path component arranged inside the housing, and a light source arranged outside the housing;

[0037] The housing is provided with a shooting hole;

[0038] The light emitted by the light source is transmitted to the Fabry-Perot microcavity after passing through the optical path component, and generates an interference pattern;

[0039] The shooting element includes a lens;

[0040] The shooting hole is suitable for cooperating with the lens to obtain the interference pattern through the shooting element.

[0041] The beneficial effects of the present invention are:

[0042] The intraocular pressure measurement method provided by the present invention uses a deep learning method to accurately identify, segment and extract the square interference fringe area contained in the photo based on the YOLOv11-seg model, binarize the skeleton of the extracted fringe, identify the order of each fringe, and then calculate the central deflection of the square film after being compressed; the central deflection of the sensor corresponds one-to-one to the pressure value calibrated by the calibration device, thereby converting the captured photo containing the interference fringe image into a pressure value. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] The present invention will be further described below with reference to the accompanying drawings and examples.

[0044] Figure 1 Schematic diagram of the process of the intraocular pressure measurement method of the present invention;

[0045] Figure 2 This is a schematic diagram of a specific implementation process of converting an image into pressure in the present invention;

[0046] Figure 3 It is a schematic diagram of the training results of YOLOv11-seg in the present invention;

[0047] Figure 4 It is a schematic diagram of the structure of the deep convolutional neural network model in the present invention;

[0048] Figure 5 It is part of the dataset images used in training in the present invention;

[0049] Figure 6 is the change of stripe pattern under different pressures in the present invention;

[0050] Figure 7 This is a simplified structural diagram of the intraocular pressure measuring device of the present invention;

[0051] Figure 8 is an exploded view of the intraocular pressure sensor of the present invention;

[0052] Figure 9 is an exploded view of the shooting module of the present invention;

[0053] Figure 10 It is a structural schematic diagram of the optical path component in the present invention;

[0054] Figure 11 Schematic diagram of the effect of shooting angle on the completeness of the optical interference pattern in the present invention;

[0055] Figure 12 It is a schematic diagram of the pressurization process in the present invention.

[0056] Figure 13 It is a physical picture of the pressurizing device in the present invention.

[0057] In the figure: 1-intraocular pressure sensor; 11-sensor body; 12-bracket; 121-mounting end; 122-fixed end; 1221-wide section; 1222-gradient section; 1223-narrow section; 1224-anti-slip structure; 123-drainage groove; 1231-first groove-shaped structure; 1232-second groove-shaped structure; 2-shooting module; 21-housing; 211-shooting hole; 22-optical path component; 221-beam splitter cube; 222-plano-convex lens; 223-narrow-band filter; 23-light source; 24-clamp; 241-C-shaped element; 2411-mounting groove; 2412-through hole; 242-threaded connector; 3-shooting element; 31-lens. DETAILED DESCRIPTION

[0058] The present invention will now be described in further detail. The embodiments described below are exemplary and intended to explain the present invention, but should not be construed as limiting the present invention. All other embodiments obtained by persons of ordinary skill in the art based on the embodiments of the present invention without creative effort are intended to fall within the scope of protection of the present invention.

[0059] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0060] In order to solve the problem that the interference area recognition and extraction method used in the prior art MEMS intraocular pressure sensor is difficult to be applied to the real-time continuous demodulation process, the present invention provides an intraocular pressure measurement method, taking the shooting element as a mobile phone as an example, see Figure 1 、 Figure 2 As shown, the measurement method includes the following steps:

[0061] S1: Sensing intraocular pressure changes through the Fabry-Perot microcavity on the intraocular pressure sensor;

[0062] This step can be achieved by implanting a corresponding MEMS intraocular pressure sensor based on a light sensor into the anterior chamber of the eyeball, where the outer surface of the Fabry-Perot microcavity contacts the intraocular fluid to sense changes in intraocular pressure.

[0063] S2: The camera module emits light into the Fabry-Perot microcavity and obtains a photo of the interference pattern through the camera element;

[0064] The light emitted by the camera module is incident on the Fabry-Perot microcavity, enters the cavity of the Fabry-Perot microcavity, is reflected by multiple surfaces and interferes, and an interference pattern is obtained;

[0065] Specifically, the preferred shooting module of the present invention includes an optical path component and a light source; the light emitted by the light source is transmitted to the Fabry-Perot microcavity after passing through the optical path component, and generates an interference pattern; the shooting element includes a lens to obtain the interference pattern through the shooting element;

[0066] The light source of the present invention is preferably a white light source; the imaging element can be any existing digital camera, video camera, smart phone, etc. including a CMOS image sensor; to further reduce the difficulty of shooting, the imaging element of the present invention is preferably a mobile phone; the mobile phone is provided with a CMOS image sensor to realize the collection and imaging of the interference pattern;

[0067] During operation, the light emitted by the light source passes through the optical path component and is vertically incident on the Fabry-Perot microcavity of the intraocular pressure sensor, namely the FP resonant cavity. After entering the cavity of the Fabry-Perot microcavity, it is reflected by multiple surfaces and interferes, forming an interference pattern. The obtained interference pattern is then transmitted to the lens of the imaging element after passing through the optical path component. The imaging element takes a photo to capture the interference pattern in real time.

[0068] S3: Based on the YOLOv11-segment model, the interference fringe area in the photo is identified and segmented to obtain the interference fringe area;

[0069] Existing interference region recognition and extraction methods usually involve manual image cropping or edge extraction to extract the edges of the region for further segmentation. Manual image cropping makes this method difficult to apply to real-time continuous demodulation processes. However, the use of classic edge extraction methods will result in different imaging edge features under different experimental conditions due to different light intensities, different imaging clarity, and other different environmental factors. This requires manual adjustment of appropriate parameters for each experiment, which is also not conducive to real-time continuous measurement in multiple scenarios. In addition, the strong edge features of the interference fringes themselves further increase the instability of traditional edge detection in this task.

[0070] Based on this, the present invention uses a deep learning method and the YOLOv11-seg model to accurately identify, segment and extract the interference fringe area contained in the photo, avoiding manual parameter adjustment during the interference fringe extraction process, so that the intraocular pressure measurement method provided by the present invention can be applied to a real-time continuous demodulation process;

[0071] S4: Based on a deep convolutional neural network, the interference fringe area is segmented and extracted into binary form, avoiding interference caused by noise and uneven background brightness in the image, and obtaining binary interference fringes;

[0072] Furthermore, in order to extract the phase information carried by the interference fringes, the present invention uses a deep convolutional neural network to further binarize the interference fringes in the interference area;

[0073] S5: Extract the skeleton of the binary interference fringes, identify the order of each fringes, and then calculate the central deflection of the Fabry-Perot microcavity film after compression;

[0074] S6: Obtain intraocular pressure based on central deflection.

[0075] The intraocular pressure measurement method provided by the present invention uses a deep learning method to accurately identify, segment and extract the square interference fringe area contained in the photo based on the YOLOv11-seg model, binarize the skeleton of the extracted fringe, identify the order of each fringe, and then calculate the central deflection of the square film after being compressed; the central deflection of the sensor corresponds one-to-one to the pressure value calibrated by the calibration device, thereby converting the captured photo containing the interference fringe image into a pressure value.

[0076] Specifically, the present invention preferably identifies and segments the interference fringe area contained in the photo based on the YOLOv11-segment model, including:

[0077] S31: Acquire photos of interference patterns under multiple different environmental conditions;

[0078] In this step, a mobile phone is used as an example to capture sensor interference photos. The photos captured contain different sensors, different pressures, different angles, different light intensities, different mobile phone camera settings, and other different environmental conditions.

[0079] S32: manually labeling the interference fringe areas in the photo using the Labelme plug-in to obtain a txt format file as a label set corresponding to the training set image;

[0080] The Label file exported by the Labelme plug-in is in json format by default. Through rule conversion, the json file is converted into the txt file format required by the YOLO model as the label set corresponding to the training set images; these images and labels are rotated, scaled, translated, and other processes are performed to further expand the data set;

[0081] S33: Using part of the images as a training set and part of the images as a validation set, the model is trained to obtain the interference fringe area;

[0082] During model training, 153 images are selected as training set and 54 images are selected as validation set. The pre-trained weights of YOLOv11n-seg are loaded. Data augmentation is enabled during training. The image input size is specified as 640x640x3. The performance indicators during the training iteration are as follows: Figure 3 As shown in Figure 3, by the end of training, the mask accuracy has reached over 99%.

[0083] In the present invention, the stripe binarization task is regarded as an image segmentation problem. The model is used to segment bright stripes and dark stripes into two different classes. The model is an improvement of the network M-net applied in medical image segmentation. It is a typical deep convolutional neural network that communicates feature maps of various sizes through jump connections at different scales to achieve the effect of multi-scale feature fusion. The deep convolutional neural network contains a total of four paths, including a left path, a right path, an encoding path, and a decoding path. Among them, the left and right paths play the role of deep supervision. The entire network involves convolutional layers, maximum pooling layers, upsampling layers, batch normalization layers, ReLU layers, and Sigmoid layers. Each path in the network contains feature maps of four sizes. The encoding layer adopts a typical CNN architecture. At each level, two 3×3 Conv-BN-ReLU modules are first used to extract features. Then, 2×2 max-pooling with a step size of 2 is used to reduce the feature map size by 1 / 2, so the number of convolution kernels in the next convolution layer is set to twice that of the previous layer. The decoder path uses the same structure as the encoder. In the decoder path, the structure and convolution layer parameter settings of each level are the same as those of the symmetrical encoder path. The difference is that the decoder part uses a 2×2 up-sampling layer as the inverse operation of the maximum pooling layer to restore the scale of the feature map to the original image size step by step. Finally, the output of the right path is cascaded with the output of the decoder in the channel dimension and sent to a 1×1 Conv-Sigmoid module to obtain the probability of each pixel being classified as a positive sample. Each layer has jump connections within the encoding path and decoding path, as well as between adjacent different paths, to obtain better segmentation effects. The specific model structure is as follows: Figure 4 As shown;

[0084] Furthermore, the present invention preferably uses a loss function in the model training of a deep convolutional neural network including a focal loss function and a multi-scale structural similarity metric; the loss function used in the model training is mainly composed of two parts, namely, a focal loss function (FocalLoss, FL) and a multi-scale structural similarity metric (Multi-Scale Structural Similarity, MS-SSIM):

[0085] L FL (p, y) = L FL (p t )=-0.5(1-p t ) r log(p t );

[0086] Specifically, during the experiment, pixels at the edge of the stripes were easily misclassified, resulting in uneven stripe edges. Focal loss is an improved cross-entropy loss function used to address the severe imbalance between foreground and background classes during the training of dense detectors in object detection tasks. The preferred focal loss function of the present invention is defined as follows:

[0087] L FL (p, y) = L FL (p t )=-0.5(1-p t ) r log(p t );

[0088] in,

[0089] y∈{1,0} represents the true label value of the sample, p∈[0,1] is the output of the model, that is, the probability that the true label of the predicted sample is 1; r is the adjustment factor in the focal loss function, which is set to 2 in this invention.

[0090] Interference fringes have strong structural characteristics. The human visual system is very good at extracting structural information from image scenes. The structural similarity metric (SSIM) is a good estimate of the image quality perceived by humans. The formula for calculating the structural similarity of images x and y is:

[0091] SSIM(x,y)=[l(x,y)] α [c(x, y)] β [s(x, y)] γ ;

[0092] Where α, β, and γ are weighting factors; l(x,y), c(x,y), and s(x,y) represent the image similarity in brightness, contrast, and structure, respectively. However, SSIM is calculated at a fixed scale and is therefore only applicable to images of a specific scale. The multi-scale structural similarity metric improves this. When the viewing angle changes, MS-SSIM is more flexible than the single-scale SSIM. The multi-scale structural similarity metric in this invention is defined as follows:

[0093]

[0094] Where M is the number of scales, α M ,β j ,γ j l M (x,y),c j (x,y),s j The weight of (x,y), lM (x,y),c j (x,y),s j (x,y) are the similarities of x and y in terms of brightness, contrast and structure at scale M,i,j respectively.

[0095] The value range of MS-SSIM is [0,1], and it takes 1 if and only if the two images are exactly the same. Therefore, the loss function of the multi-scale structural similarity measure is defined as:

[0096] L MS-SSIM (p, y) = 1 - MS_SSIM(x, y).

[0097] Furthermore, the method for establishing a dataset for the deep convolutional network in the present invention is as follows: 26 interference fringe images taken in the experiment were used as the training set. These images were first binarized using the FCM clustering algorithm, and the binarized images were further improved using Adobe Photoshop. Because the number of such datasets is too small, the training set and the label set were randomly cropped, rotated, mirrored, perspective transformed, denoised, and the brightness and contrast were changed to increase the number of samples to 2,600. In addition, 500 interference images and corresponding binarized images were simulated using MATLAB 2024a, totaling 3,100 images with a size of 496×496 as the dataset for training. The equipment used for training is equipped with a 12th Gen Intel(R) Core(TM) i7-12650H CPU with 32GB RAM and an NVIDIA GeForce RTX4060Laptop GPU. During the training process, the SGD algorithm was used to optimize the network parameters. The batch size was set to 4, the initial learning rate was set to 0.01, the Nesterov momentum was set to 0.75, the learning rate was decayed by 0.00005 per iteration, and the training was stopped early if the validation set loss did not decrease after 5 epochs. The maximum number of network training cycles was set to 50, and finally 35 cycles were trained, totaling about 9 hours. The training set images used in some training are as follows: Figure 5 shown.

[0098] Furthermore, the skeleton lines of the binary interference fringe image are extracted and the order of each skeleton line is marked, and the corresponding unwinding phase is calculated. The deflection distribution function of the Fabry-Perot microcavity film after compression is as follows:

[0099]

[0100] Where (x, y) is the coordinate of the point on the diaphragm, the center of the diaphragm is the coordinate origin, w0 is the center deflection of the diaphragm, α is the side length of the diaphragm, and c1 and c2 are two empirical parameters. In the demodulation calculation process, only the center section of the diaphragm needs to be used for fitting, that is, y is 0, and the deflection shape function of the square diaphragm changes to the center section deflection distribution function of the square diaphragm:

[0101]

[0102] The present invention extracts the pixel coordinates of the central horizontal line skeleton line of the binary interference fringe and the corresponding unwrapping phase value, fits the deflection distribution of the central section of the diaphragm, and then obtains the deflection value of the center position of the interference fringe image, that is, the center of the square diaphragm.

[0103] Furthermore, for the central deflection of the sensor diaphragm demodulated in the photo, the present invention associates it with the pressure value of the narrow chamber where the sensor is located through a calibration experiment; that is, obtaining the intraocular pressure based on the central deflection includes: using a calibration device to calibrate the central deflection of the Fabry-Perot microcavity film at different pressure values, and then obtaining the intraocular pressure based on the calculated central deflection.

[0104] For details, see Figure 12 As shown, the cubic-shaped intraocular pressure sensor is fixed by an I-shaped bracket made of silicone. The intraocular pressure sensor is mounted on the I-shaped bracket using UV-curable adhesive, and the I-shaped bracket is further adhered to the pressure chamber. Figure 13 As shown, the pressurizing device in the experiment uses a syringe. At the beginning of the experiment, the pressure chamber and the bottom of the measuring cylinder need to be fixed to the same horizontal plane to ensure that the pressure value in the pressure chamber is the liquid level displayed in the measuring cylinder. In the experiment, the liquid level of pure water in the measuring cylinder is controlled by pushing and pulling the syringe, thereby controlling the hydraulic pressure applied to the intraocular pressure sensor in the pressure chamber.

[0105] During the experiment, the syringe was pushed and pulled under control, and three rounds of pressurization reciprocating experiments were performed at room temperature, with the pressures remaining at 5cmH2O, 15cmH2O, 25cmH2O, 35cmH2O, 45cmH2O, and 55cmH2O (i.e., 3.68mmHg, 11.03mmHg, 18.39mmHg, 25.74mmHg, 33.1mmHg, and 40.46mmHg) for 3 minutes to ensure that the intraocular pressure sensor was in a stable hydraulic environment. The fringe images recorded during the experiment are shown in Figure 2. Figure 6 As shown, it can be observed that the stripe density increases with the increase of the hydraulic pressure value.

[0106] The images recorded in the test were demodulated, and the center deflection curves under different pressures were drawn. The six curves were subjected to linear regression analysis, and the coefficient of determination R of the three-wheel lift and lower pressure curves was obtained. 2All the six curves are higher than 0.99, and the six curves have high repeatability. The final pressure sensitivity obtained by fitting the data of three rounds of pressure increase and decrease is 20.92nm / mmHg.

[0107] Another object of the present invention is to provide an intraocular pressure measuring device, which measures intraocular pressure using the intraocular pressure measurement method described above.

[0108] The intraocular pressure measurement device provided by the present invention uses a deep learning method to accurately identify, segment and extract the square interference fringe area contained in the photo based on the YOLOv11-seg model during the intraocular pressure measurement process, and then the skeleton of the extracted fringe is binarized, the order of each fringe is identified, and the central deflection of the square film after being compressed is calculated; the central deflection of the sensor corresponds one-to-one to the pressure value calibrated by the calibration device, thereby converting the captured photo containing the interference fringe image into a pressure value.

[0109] For details, see Figure 7 As shown, the intraocular pressure measuring device of the present invention includes an intraocular pressure sensor 1, a shooting module 2 and a shooting element 3; it should be noted that the present invention Figure 7 In order to clearly show the structure of the intraocular pressure measurement device, the size of the intraocular pressure sensor is artificially enlarged; wherein, the intraocular pressure sensor 1 is a MEMS sensor, see Figure 8 As shown, the intraocular pressure sensor 1 is provided with a Fabry-Perot microcavity; see Figure 9 As shown, the shooting module 2 includes a shell 21, which is preferably made of polylactic acid (PLA) material and prepared by an extrusion 3D printing process; an optical path component 22 is arranged inside the shell 21, and a light source 23 is arranged outside the shell 21; a shooting hole 211 is provided on the shell 21; the light emitted by the light source 23 is transmitted to the Fabry-Perot microcavity after passing through the optical path component 22, and generates an interference pattern; the shooting element 3 includes a lens 31; the shooting hole 211 is suitable for cooperating with the lens 31 to obtain the interference pattern through the shooting element 3.

[0110] In the present invention, the light source 23 is preferably a white light source.

[0111] During operation, the light emitted by the light source 23 passes through the optical path component 22 and is vertically incident on the Fabry-Perot microcavity, i.e., the FP resonant cavity, of the intraocular pressure sensor 1. After entering the cavity of the Fabry-Perot microcavity, it is reflected by multiple surfaces and interferes to obtain an interference pattern. The obtained interference pattern is transmitted to the lens 31 of the shooting element 3 after passing through the optical path component 22. The interference pattern can be captured in real time by taking a picture with the shooting element 3, and the real-time intraocular pressure can be obtained based on the interference pattern captured in real time.

[0112] Specifically, the method for obtaining intraocular pressure based on the interference pattern may be the method described above.

[0113] During use, the intraocular pressure sensor 1 is implanted in the anterior chamber of the eyeball, and the outer surface of the Fabry-Perot microcavity contacts the intraocular fluid to sense changes in intraocular pressure; when the intraocular pressure rises, the Fabry-Perot microcavity deforms, causing the optical path of the reflected light to change, the generated interference pattern to change, and the interference fringes to bend. The interference fringe area contained in the interference pattern is accurately identified, segmented and extracted, the skeleton of the fringes is extracted after binarization, the order of each fringe is identified, and then the central deflection of the Fabry-Perot microcavity film after being compressed is calculated; the central deflection of the Fabry-Perot microcavity film corresponds one-to-one to the pressure value of the Fabry-Perot microcavity film calibrated by the calibration device, so that the photograph containing the interference fringe image is converted into a pressure value, and the intraocular pressure can be obtained according to the change of the interference pattern.

[0114] The intraocular pressure measurement device provided by the present invention can complete the detection of intraocular pressure with the help of the existing shooting element 3 by introducing a shooting module 2 compatible with the existing shooting element 3; and during the detection process, the position of the shooting module 2 and the shooting element 3 can be flexibly adjusted according to the position of the Fabry-Perot microcavity in the intraocular pressure sensor 1, so that the incident light beam meets the requirement of vertical incidence, thereby greatly reducing the difficulty of angle adjustment during the intraocular pressure detection process, making it more portable, efficient and easy to use.

[0115] To measure intraocular pressure, see Figure 10 As shown, the optical path component 22 of the present invention preferably includes a spectroscopic cube 221 and a plano-convex lens 222 sequentially arranged in the shooting hole 211, wherein the operating wavelength range of the spectroscopic cube 221 is 450-650nm. When the light is incident at an incident angle of 45°, the incident light can be split into two beams with a ratio of approximately 50% transmission (T) and 50% reflection (R), with a tolerance of ±5% (T / R=50%:50%±5%). After the light emitted by the light source 23 is reflected by the spectroscopic cube 221 to adjust the light path direction, it is converged by the plano-convex lens 222 to the target plane for generating the optical interference pattern, that is, the position of the Fabry-Perot microcavity; the reflected light of the interference pattern is then transmitted to the lens 31 of the shooting element 3 after passing through the plano-convex lens 222 and the spectroscopic cube 221, to achieve the collection and imaging of the interference pattern; the design wavelength of the plano-convex lens 222 is preferably 350-700nm, and the focal length is 20mm.

[0116] Furthermore, the preferred optical path component 22 of the present invention also includes a narrowband filter 223 arranged between the light source 23 and the spectroscopic cube 221, wherein the narrowband filter 223 is a 633nm narrowband filter, so that the light emitted by the light source 23 is filtered by the narrowband filter 223 to obtain monochromatic light with a central wavelength of 633nm and a bandwidth of ±10nm.

[0117] The shooting element 3 in the present invention can be any existing digital camera, video camera, smart phone, etc. including a CMOS image sensor; in order to further reduce the difficulty of shooting, the present invention preferably uses a mobile phone as the shooting element 3; the mobile phone is provided with a CMOS image sensor to realize the collection and imaging of interference patterns.

[0118] When the existing optical sensing-based pressure sensor measures intraocular pressure, the incident light needs to meet equal-inclination interference; for equal-inclination interference, the main requirement is that the incident angle and the reflection angle (or refraction angle) of the two interfering light beams are equal when reflected or refracted. In other words, the incident light beam must meet the requirement of vertical incidence to ensure that the interference fringes are fully formed. When the micro pressure sensor is implanted in the pressure detection environment, it is impossible to ensure whether it is level due to its small size. When using a desktop microscope, the microscope can generally only maintain a vertical downward angle. If you want to obtain a complete optical interference pattern, you can only adjust the spatial angular position of the object to be measured by feeling, which is extremely difficult, especially when the environment to be measured is an object whose spatial angular position cannot be adjusted, such as an intraocular pressure sensor implanted in the eye. Based on this, the present invention proposes an external shooting module 2 that can be adapted to any smartphone. Compared with adjusting the spatial angular position of the uncertain object to be measured, adjusting the angle of the handheld mobile phone is obviously much easier. Furthermore, we can judge at which angle to tilt the mobile phone based on the image captured in real time by the mobile phone camera. The optical interference pattern captured by a mobile phone camera in real time is a square area. When the incident angle of light is not perpendicular to the interference plane, the square area is incomplete, showing a partially bright and partially dark state. Figure 11 As shown, we can imagine the square interference area as a sealed "box" filled with water, and the bright part as a "bubble" within the sealed space. The "box" should tilt in the direction the "bubble" is located in the square area until the "bubble" moves to the center of the square area. The phone is the "box," and when the "bubble" moves to the center of the square area, the incident angle of the light is perpendicular to the interference plane, and the complete optical interference pattern is captured.

[0119] To facilitate connection with a mobile phone, the preferred embodiment of the present invention includes a camera module 2 further comprising a clamp 24 ; one end of the clamp 24 is connected to the camera element 3 , ie, the mobile phone, and the other end is connected to the housing 21 .

[0120] The preferred clamp 24 of the present invention is made of polylactic acid (PLA) material and is prepared by an extrusion 3D printing process; and the clamp 24 is further preferably a C-type clamp structure, including a C-shaped element 241 and a threaded connector 242; the C-shaped element 241 is connected to the shell 21; the C-shaped element 241 is connected to the shooting element 3 through the threaded connector 242; during use, the mobile phone is placed in the C-shaped element 241, and the connection with the mobile phone is achieved by tightening the threaded connector 242; the preferred opening and closing range of the C-shaped element 241 of the present invention is 8-20mm, which can adapt to the thickness of most smartphones on the market.

[0121] Furthermore, the present invention preferably connects the shell 21 and the clamp 24 by means of a snap fit. Specifically, the C-shaped element 241 of the clamp 24 is preferably provided with a mounting groove 2411 that matches the shell 21, and a concave point is provided in the mounting groove 2411, and a convex point that matches the concave point is provided on the outer side of the shell 21. The convex point and the concave point cooperate with each other, so that the two parts can be easily assembled or disassembled.

[0122] In addition, a through hole 2412 adapted to the shooting hole 211 is provided in the installation groove 2411 to prevent the C-shaped element 241 from affecting light transmission.

[0123] The present invention provides an external shooting module 2 that is compatible with any smartphone. The shell 21 of the external shooting module 2 is made of environmentally friendly polylactic acid (PLA) material and is prepared through a precisely controlled extrusion 3D printing process to ensure consistency in structural strength and quality. The shooting module 2 integrates an optimized customized optical path design and high-performance optical components, and combines real-time image capture to achieve stable shooting of high-quality images. Compared with the first-generation desktop microscope-style image capture method, this module not only ensures image clarity and optical imaging quality, but also significantly reduces the complexity of user operation and the impact of jitter during shooting, making the device more portable and more efficient to operate, while improving the experience and applicability.

[0124] The intraocular pressure sensor 1 in the present invention can be any existing intraocular pressure sensor provided with a Fabry-Perot microcavity; since the intraocular pressure monitoring device provided by the present invention detects intraocular pressure based on optical sensing, as mentioned above, during the detection process, the incident light needs to satisfy equal-inclination interference, and equal-inclination interference requires that the incident light beam meets the requirement of vertical incidence; therefore, in order to ensure the clarity of the detected image, the position of the intraocular pressure sensor 1 is required to remain fixed and not move during the detection process; in order to avoid the movement of the intraocular pressure sensor 1 during the detection process, the present invention preferably includes a sensor body 11 and a bracket 12 connected to the sensor body 11, so that the intraocular pressure sensor body 11 can be fixed by the bracket 12, thereby reducing the difficulty of detection and improving the clarity of the detected image.

[0125] Existing brackets for fixing intraocular implants are mostly cylindrical structures. Due to the intraocular pressure measurement device provided by the present invention, if the position of the intraocular pressure sensor body 11 moves slightly during the detection process, the incident light beam will not be incident vertically, and the shooting angle of the shooting element 3 needs to be readjusted. Therefore, to ensure the stability of the position of the intraocular pressure sensor body 11 during the detection process, the present invention preferably adopts a plate-like structure for the bracket 12 to increase the contact area between the bracket 12 and the inside of the eye and prevent the intraocular pressure sensor body 11 from moving.

[0126] Specifically, the preferred bracket 12 of the present invention includes a mounting end 121 and a fixed end 122 connected to the mounting end 121; wherein the mounting end 121 is used to be connected to the intraocular pressure sensor body 11, and the intraocular pressure sensor body 11 and the mounting end 121 in the present invention can be connected by high-temperature bonding, compatible material bonding, etc.; the size of the mounting end 121 is determined according to the size of the intraocular pressure sensor body 11.

[0127] Since the eyeball has a certain curvature, in order to improve the fit between the bracket 12 and the eyeball, improve the stability of the intraocular pressure sensor body 11, and improve the patient's comfort, the present invention preferably has the fixed end 122 as an arc-shaped structure, the curvature of which is determined according to the curvature of the eyeball.

[0128] In order to ensure the stability of the intraocular pressure sensor body 11 while improving comfort, the present invention preferably has the fixed end 122 include a wide section 1221, a gradient section 1222 and a narrow section 1223 connected in sequence; wherein the widths of the wide section 1221, the gradient section 1222 and the narrow section 1223 decrease in sequence.

[0129] It should be noted that, in the bracket 12 of the present invention, the length direction in which the wide section 1221 , the gradient section 1222 and the narrow section 1223 are distributed is the width direction, and the direction perpendicular to the length direction on the plane of the plate-like structure of the bracket 12 is the width direction.

[0130] Specifically, the present invention preferably has the wide section 1221 as a plate-like structure with a rectangular cross-section to ensure the contact area between the bracket 12 and the eyeball; the width of the gradient section 1222 preferably decreases successively, and the width of one end connected to the wide section 1221 is the same as the width of the wide section 1221, and the width of one end connected to the narrow section 1223 is the same as the width of the narrow section 1223.

[0131] In order to take into account both comfort and the stability of the position of the intraocular pressure sensor body 11, the present invention preferably has a length ratio of the wide section 1221, the gradient section 1222 and the narrow section 1223 of (1.4-1.7): (1.1-1.4): (0.7-1).

[0132] In order to further improve the stability of the intraocular pressure sensor body 11 after implantation, the present invention preferably provides an anti-slip structure 1224 on the outside of the narrow section 1223 , and specifically preferably the anti-slip structure 1224 is a protruding structure extending outward along the narrow section 1223 .

[0133] In order to take into account the stability and comfort of the position of the intraocular pressure sensor body 11, the present invention further prefers that the number of protrusion structures is at least two groups, each group includes two protrusions of the same size, symmetrically arranged on both sides of the narrow section 1223; and the size of the protrusion gradually decreases in the direction away from the gradient section 1222.

[0134] The present invention further preferably provides a drainage groove 123 on the fixed end 122, so that the intraocular pressure measurement device provided by the present invention has a certain drainage effect while having the intraocular pressure measurement function.

[0135] Existing intraocular pressure sensors usually only have the function of detecting intraocular pressure and cannot achieve drainage. When the intraocular pressure is high, it is necessary to introduce a corresponding drainage device to achieve a therapeutic effect. Based on this, the present invention preferably provides a drainage groove 123 on the bracket 12 to facilitate the diffusion of aqueous humor to the surrounding tissues of the eye through the drainage groove 23, so that the intraocular pressure sensor can have both intraocular pressure detection and drainage functions, so that intraocular pressure detection and drainage can be achieved through a single implant without increasing the number of implants, increasing surgical damage, or increasing patient pain.

[0136] The drainage groove 123 includes a groove-like structure distributed longitudinally along the bracket 12, which passes through the mounting end 121 and the fixed end 122 in sequence, and is recorded as a first groove-like structure 1231; in order to further improve the drainage effect, the drainage groove 123 also includes a second groove-like structure 1232 obliquely distributed on the wide section 1221, and the second groove-like structure 1232 is connected to the first groove-like structure 1231.

[0137] The intraocular pressure sensor provided by the present invention can be implanted into the eye by injection, which significantly reduces surgical trauma. The intraocular pressure sensor provided by the present invention can be implanted into the eye by minimally invasive injection through a breakthrough design. The system does not require electronic components and electromagnetic energy supply to achieve all-weather continuous monitoring of intraocular pressure, and the intraocular pressure measurement accuracy reaches ±1mmHg.

[0138] The intraocular pressure sensor in the present invention can establish a functional relationship between the deflection of the Fabry-Perot microcavity in the central area of the sensor and the intraocular pressure based on the change in the spacing of the sensor interference fringes caused by changes in intraocular pressure. Through the advanced YOLO target detection instance segmentation and M-net deep convolutional neural network algorithm integrated in the mobile phone APP, it automatically focuses, identifies and crops the sensor interference fringes area and performs real-time intraocular pressure demodulation, enabling patients to self-measure their intraocular pressure with their mobile phones at home. It does not rely on signal transmission of electromagnetic energy supply and effectively avoids signal loss caused by external factors.

[0139] The intraocular pressure sensor provided by the present invention has stronger compatibility and universality, and can be compatible and used in conjunction with any existing ophthalmic implant device, which will help further explore the feasibility of new clinical technologies that integrate glaucoma monitoring and diagnosis and treatment.

[0140] With the above-described preferred embodiments of the present invention as a guide, and with reference to the above description, relevant personnel are fully capable of making various changes and modifications without departing from the technical scope of this invention. The technical scope of this invention is not limited to the contents of the specification and must be determined according to the scope of the claims.

Claims

1. A method for measuring intraocular pressure, characterized in that: The steps include: S1: Sensing intraocular pressure changes through the Fabry-Perot microcavity on the intraocular pressure sensor; S2: emitting light to the Fabry-Perot microcavity through a photographing module, and obtaining a photograph of the interference pattern through a photographing element; S3: Identify and segment the interference fringe region contained in the photo based on the YOLOv11-segment model to obtain the interference fringe region; S4: Binarizing and segmenting the interference fringe region based on a deep convolutional neural network and extracting the interference fringe region to obtain binary interference fringes; S5: extracting the skeleton of the binary interference fringes, marking the order of each fringes, and then calculating the central deflection of the film of the Fabry-Perot microcavity after being compressed; S6: Obtain intraocular pressure based on the central deflection.

2. The intraocular pressure measurement method according to claim 1, wherein Based on the YOLOv11-segment model, the interference fringe area contained in the photo is identified and segmented, including: S31: Acquire photos of interference patterns under multiple different environmental conditions; S32: manually labeling the interference fringe areas in the photo using the Labelme plug-in to obtain a txt format file as a label set corresponding to the training set image; S33: Using part of the images as a training set and part of the images as a validation set, the model is trained to obtain the interference fringe area.

3. The intraocular pressure measurement method according to claim 1, wherein The deep convolutional neural network includes a left path, a right path, an encoding path and a decoding path.

4. The intraocular pressure measurement method according to claim 3, wherein: The loss functions used in the model training of the deep convolutional neural network include a focal loss function and a multi-scale structural similarity metric.

5. The intraocular pressure measurement method according to claim 4, wherein: The focal loss function is defined as follows: L FL (p,y)=L FL (pt)=-0.5(1-p t ) r log(p t ); in, y∈{1,0} represents the true label value of the sample, p∈[0,1] is the output of the model, that is, the probability that the true label of the predicted sample is 1; r is the adjustment factor in the focal loss function.

6. The intraocular pressure measurement method according to claim 4, wherein: The multi-scale structural similarity measure is defined as follows: Where M is the number of scales, α M ,β j ,γ j l M (x,y),c j (x,y),s j The weight of (x,y), l M (x,y),c j (x,y),s j (x,y) are the similarities of x and y in terms of brightness, contrast and structure at scale M,i,j respectively.

7. The intraocular pressure measurement method according to claim 1, wherein The deflection distribution function of the Fabry-Perot microcavity film after being compressed is as follows: Where (x, y) is the coordinate of the point on the diaphragm, the center of the diaphragm is the coordinate origin, w0 is the central deflection of the diaphragm, α is the side length of the diaphragm, and c1 and c2 are two empirical parameters.

8. The intraocular pressure measuring method according to claim 1, wherein: Acquiring intraocular pressure according to the central deflection includes: using a calibration device to calibrate the central deflection of the Fabry-Perot microcavity film at different pressure values, and then acquiring the intraocular pressure according to the calculated central deflection.

9. An intraocular pressure measuring device, characterized in that: The intraocular pressure is measured by the intraocular pressure measurement method according to any one of claims 1 to 8.

10. The intraocular pressure measuring device according to claim 9, wherein It comprises an intraocular pressure sensor (1), a shooting module (2) and a shooting element (3); wherein, The intraocular pressure sensor (1) is provided with a Fabry-Perot microcavity; The shooting module (2) comprises a housing (21), an optical path component (22) arranged inside the housing (21), and a light source (23) arranged outside the housing (21); The housing (21) is provided with a shooting hole (211); The light emitted by the light source (23) passes through the optical path component (22) and is then transmitted to the Fabry-Perot microcavity, thereby generating an interference pattern; The photographing element (3) includes a lens (31); The shooting hole (211) is suitable for cooperating with the lens (31) to obtain the interference pattern through the shooting element (3).

Images