Method for real-time closed-loop control of formation of subretinal injection blisters

Abstract

The invention discloses a method for real-time closed-loop control of formation of subretinal injection vesicles, and relates to the technical field of ophthalmology medical treatment, and the method comprises the following steps: S100, system initialization and preparation; s200, OCT image acquisition and speckle dynamic monitoring are carried out; s300, adjusting the injection pressure in real time; and S400, injection is completed. The safety and accuracy of subretinal injection are remarkably improved, overextension of vesicles and separation of retina can be effectively prevented, different retina pathologies and individual differences can be adapted, and wide clinical applicability is achieved.

Description

Technical Field

[0001] This invention relates to the field of ophthalmic medical technology, and in particular to a method for real-time closed-loop control of subretinal injection blister formation. Background Art

[0002] Current subretinal injection techniques require precise control of injection pressure when blisters form in order to balance treatment effectiveness and retinal safety.

[0003] In existing subretinal injection techniques, blister formation typically relies on the surgeon manually adjusting the injection pressure based on intraoperative observations of retinal dynamics, such as blister expansion patterns and retinal bulge morphology. This method has the following main drawbacks: The delay in manual adjustment: Surgeons need to visually assess tissue response to adjust pressure, and this visual observation inherently has a delay, which may lead to untimely pressure adjustment. For example, when the retina separates from the RPE, if the injection pressure is not reduced in time, it may lead to excessive retinal expansion or even retinal detachment, thereby increasing surgical risks.

[0004] Reliance on preoperative assumptions: Existing methods typically require setting pressure thresholds based on preoperative assumptions about retinal pathology, which cannot adapt to the dynamic changes in the retina during surgery. Different patients' retinal pathological conditions (such as age, disease progression, etc.) lead to differences in the adhesion between the retina and the RPE, thus requiring personalized pressure control strategies.

[0005] Lack of adaptive feedback: Current technologies lack real-time feedback on tissue biomechanical properties, making it difficult to adapt to the differences in the mechanical properties of the retina among different patients. For example, the mechanical properties of the retina may change due to local fibrosis or stiffness of the RPE-photoreceptor interface, which can pose challenges to manually adjusting injection pressure.

[0006] Analysis of existing technologies for adjusting subretinal injection bubble pressure: The invention patent "Method and Apparatus for Subretinal Injection" (Publication No. CN116801847A) discloses an apparatus for subretinal injection into the subretinal space between the retina and the retinal pigment epithelium of the eye. The apparatus includes an injection needle having a proximal end and a distal end, the distal end being configured to be inserted into the subretinal space at a location on the surface of the retina. The apparatus includes a multi-lumen tube having a distal end connected to the proximal end of the injection needle and a proximal end connected to a fluid control unit. The apparatus includes a stabilizer configured to fix the injection needle at this location on the surface of the retina. The fluid control unit has multiple fluid reservoirs containing non-therapeutic solutions, therapeutic solutions, and working fluids that can be injected into the eye via the individual lumens of the multi-lumen tube. This patent relies primarily on surgeons manually adjusting injection pressure based on intraoperative observations of retinal dynamics (such as blister expansion patterns and retinal bulge morphology), which inherently introduces a delay and may lead to excessive retinal expansion or even retinal detachment. Furthermore, this patent does not involve a real-time feedback mechanism based on OCT speckle dynamics, making it impossible to monitor microstructural changes during the separation of the retina from the RPE in real time, thus hindering precise pressure control. Finally, this patent requires setting pressure thresholds based on preoperative assumptions about retinal pathology, making it unable to adapt to dynamic changes in the retina during surgery.

[0007] The utility model patent "Injection System for Subretinal Drug Injection" (Publication No. CN218922966U) discloses an injection system for subretinal drug injection, comprising an injection device including an injection needle and a syringe for containing the drug. The syringe is provided with graduation marks. The injection system also includes an auxiliary lighting device integrated into the injection device, allowing local illumination for the graduation marks on the syringe. The injection system according to this utility model solves the problem of cumbersome dose reading during injection operations in a dark room in the prior art, improving the convenience of dose reading during subretinal drug injection and enhancing the safety of the injection operation. However, although this patent proposes an auxiliary lighting device to improve the convenience of dose reading, it still relies on manual operation and cannot solve the problem of delayed injection pressure control; this patent does not involve a real-time feedback mechanism based on tissue biomechanical properties, making it difficult to adapt to the differences in the mechanical properties of the retina of different patients; this patent mainly focuses on the problem of dose reading and fails to solve the problem of retinal tearing due to excessive pressure or repeated injection due to insufficient pressure during injection.

[0008] The patent "Retinal Vascular Injector and Injection Method for Ophthalmic Surgical Robots" (Publication No. CN110368184A) discloses a retinal vascular injector and its injection method. This invention aims to address the problems of low surgical and positioning efficiency in existing injector devices, as well as the inability to directly insert into the human eye, leading to injection failure or secondary injury. The invention includes an actuator base, a needle assembly module, a feeding module, and a rotation module. The needle assembly module is mounted on the feeding module and is driven by the feeding module to achieve feeding. The feeding module is mounted at the front of the rotation module, and the needle assembly module and feeding module rotate under the drive of the rotation module. The rotation module is mounted on the actuator base. This invention is intended for use in ophthalmic surgical robots. However, this patent primarily focuses on the design of the retinal vascular injector and does not address the real-time control mechanism of the injection pressure; the patent requires injection operations based on preoperative planning and lacks real-time response capability to dynamic changes during surgery; the patent still relies on manual adjustment of the injection pressure, which carries the risk of operational delays and human error.

[0009] The patent "An OCT Image Speckle Suppression Method" (Publication No. CN111145280A) discloses an OCT image speckle suppression method, comprising: an image registration step, which registers a first low-resolution image and a plurality of subsequent consecutive second low-resolution images to calculate motion parameters of the second low-resolution images relative to the first resolution image; and an image reconstruction step, which calculates the coordinates of the low-resolution images in the high-resolution image based on the calculated motion parameters, and reconstructs the high-resolution image by using adaptive normalized convolution to calculate pixel differences; wherein the motion parameters include a horizontal offset a, a vertical offset b, and a rotation angle θ, and the first low-resolution image and the second low-resolution images have the same resolution. By benefiting from real measurements of multiple low-resolution images to increase image entropy, this method not only reduces speckle noise from OCT images but also enhances structural characteristics. However, this patent mainly focuses on suppressing speckle noise in OCT images, without addressing the real-time monitoring and utilization of speckle dynamic changes; this patent does not address the real-time control of injection pressure during subretinal injection, and cannot solve the problem of retinal damage caused by excessive pressure during injection; this patent mainly reduces speckle noise through image processing algorithms, but does not utilize speckle dynamic changes as a real-time feedback signal.

[0010] In summary, existing technologies rely mainly on manual adjustment of injection pressure during subretinal injection, lacking a real-time feedback mechanism and making it difficult to adapt to dynamic changes in the retina during surgery.

[0011] Therefore, those skilled in the art are dedicated to developing a method for real-time closed-loop control of subretinal injection blister formation. Summary of the Invention

[0012] In view of the above-mentioned deficiencies of the prior art, the technical problem to be solved by the present invention is how to solve the problems of manual adjustment delay, reliance on preoperative assumptions and lack of adaptive feedback in the prior art, and improve the safety and accuracy of subretinal injection.

[0013] The applicant argues that existing patents primarily rely on manual adjustment of injection pressure during subretinal injection, lacking a real-time feedback mechanism and failing to adapt to dynamic changes in the retina during surgery. The applicant utilizes OCT speckle dynamics as a real-time monitoring indicator of tissue biomechanical changes during subretinal injection. Speckle decorrelation or speckle displacement reflects microstructural changes during retinal-RPE separation, providing a reliable basis for automatic injection pressure adjustment. Real-time monitoring of retinal-RPE separation using OCT speckle dynamics (such as speckle decorrelation or speckle displacement) automatically triggers pressure adjustment, avoiding the delays of manual adjustment. OCT speckle dynamics can detect retinal-RPE separation with sub-second delays, achieving rapid pressure feedback control. By monitoring tissue biomechanical changes in real time, the applicant can adapt to different retinal pathologies and individual differences without relying on preoperative assumptions. The system dynamically adjusts injection pressure based on real-time monitored speckle dynamic signals, adapting to the retinal biomechanical characteristics of different patients. Precise control of injection pressure avoids retinal tears and macular hole formation due to excessive pressure or repeated retinal incision due to insufficient pressure, thereby improving the safety and effectiveness of treatment.

[0014] In one embodiment of the present invention, a method for real-time closed-loop control of subretinal injection blister formation is provided, comprising the following steps: S100. System initialization and preparation: calibrate the OCT system, install the injection equipment, and locate the target area for subretinal injection. S200, OCT image acquisition and speckle dynamic monitoring, inject into the target area of ​​subretinal injection, during injection, real-time acquisition of OCT B-scan images, calculation of speckle dynamics, plotting of speckle dynamic curves, and generation of real-time speckle dynamic signals. S300: Real-time injection pressure adjustment. When the real-time speckle dynamic signal reaches the preset threshold, the injection pressure is reduced. When the injection pressure is reduced to the injection pressure that maintains the stable expansion of the blister, the injection pressure is maintained. S400, Injection complete. When the real-time speckle dynamic signal meets the injection completion condition, if the rise or fall is less than the set fluctuation range threshold within 3 seconds, the injection is complete. Stop the injection, shut down the injection equipment and OCT system, otherwise return to step S300.

[0015] Optionally, in the method for real-time closed-loop control of subretinal injection blister formation in the above embodiments, step S100 includes: S110, OCT system calibration: Calibrate the OCT system to ensure imaging accuracy and resolution; S120. Prepare the injection equipment, install the injection equipment, and set the initial injection pressure. S130. Target area localization: Determine the target area for subretinal injection using OCT imaging, and locate the injection point in the target area using OCT B-scan images.

[0016] Optionally, in the method for real-time closed-loop control of subretinal injection blister formation in the above embodiments, calibrating the OCT system includes: light source intensity calibration, scanning path calibration, and image contrast calibration.

[0017] Furthermore, in the method for real-time closed-loop control of subretinal injection blister formation in the above embodiments, the injection device includes an injection needle or an injection pump.

[0018] Furthermore, in the method for real-time closed-loop control of subretinal injection blister formation in the above embodiments, the initial injection pressure is set according to a clinical safety threshold to overcome the adhesion between the retina and the RPE.

[0019] Furthermore, in the method for real-time closed-loop control of subretinal injection blister formation in the above embodiments, the clinical safety threshold ranges from 4 psi to 8 psi.

[0020] Preferably, in the method for real-time closed-loop control of subretinal injection blister formation in the above embodiments, the initial injection pressure is 6 psi.

[0021] Optionally, in the method for real-time closed-loop control of subretinal injection blister formation in any of the above embodiments, step S200 includes: S210, Injection begins. Begin injecting the target area under the retina with the initial injection pressure. S220, OCT image acquisition: During the injection process, OCT B-scan images are acquired in real time according to the acquisition frequency; S230, Dynamic calculation of speckle pattern, calculating speckle decorrelation value or speckle displacement: S240. Real-time feedback signal generation: Based on the speckle decorrelation value or speckle displacement, a speckle dynamic curve is plotted to generate a real-time speckle dynamic signal.

[0022] Preferably, in the method for real-time closed-loop control of subretinal injection blister formation in the above embodiments, the acquisition frequency is 80 kHz.

[0023] Furthermore, in the method for real-time closed-loop control of subretinal injection blister formation in the above embodiments, step S230 calculates the speckle decorrelation value to quantify the temporal changes of speckle and reflect the changes in tissue microstructure, as shown in the following formula: ; in, Indicates the location and time t The speckle decorrelation value, Indicates the location and time t OCT signal strength, For time intervals.

[0024] Furthermore, in the method for real-time closed-loop control of subretinal injection blister formation in the above embodiments, the speckle displacement in step S230 is calculated using the optical flow method, which reflects the mechanical deformation of the tissue, and the formula is as follows: ; in, The gradient of the OCT image. =( x, y) is the speckle displacement vector. This is the time derivative.

[0025] Optionally, in the method for real-time closed-loop control of subretinal injection blister formation in the above embodiments, step S240 further includes smoothing the speckle dynamic curve to reduce noise interference.

[0026] Optionally, in the method for real-time closed-loop control of subretinal injection blister formation in any of the above embodiments, step S300 includes: S310, Injection pressure adjustment: When the real-time speckle dynamic signal reaches the preset threshold, the injection pressure is reduced. S320, Injection pressure maintenance: When the injection pressure decreases to the level required to maintain stable blister expansion, maintain the injection pressure.

[0027] Furthermore, in the method for real-time closed-loop control of subretinal injection blister formation in the above embodiments, the preset threshold is set based on the speckle decorrelation value or the statistical characteristics of speckle displacement.

[0028] Furthermore, in the method for real-time closed-loop control of subretinal injection blister formation in the above embodiments, the preset threshold is set as the speckle decorrelation value or the mean of speckle displacement plus twice the standard deviation.

[0029] Furthermore, in the method for real-time closed-loop control of subretinal injection blister formation in the above embodiments, the injection pressure reduction is achieved through flow control of the injection device. When the injection pressure is too high, the flow rate and flow of the injection pump are reduced to ensure stable expansion of the blister.

[0030] Preferably, in the real-time closed-loop control method for subretinal injection blister formation in the above embodiments, the injection pressure to maintain stable blister expansion is 4 psi.

[0031] Optionally, in the method for real-time closed-loop control of subretinal blister formation in any of the above embodiments, step S400 includes: S410. Injection complete. When the real-time speckle dynamic signal meets the injection completion condition, it indicates that the separation between the retina and RPE has been completed and no new separation has occurred. It is determined that the injection process has achieved the expected effect and the injection is complete. Proceed to step S420. Otherwise, return to step S300. S420, System shutdown, stop injection, shut down injection equipment and OCT system.

[0032] Optionally, in the method for real-time closed-loop control of subretinal injection blister formation in any of the above embodiments, the injection completion condition is that the real-time speckle dynamic signal rises or falls less than a set fluctuation range threshold within 3 seconds.

[0033] Optionally, in the method for real-time closed-loop control of subretinal injection blister formation in any of the above embodiments, the set fluctuation range threshold is 0.02.

[0034] This invention utilizes real-time closed-loop control of subretinal injection blister formation based on OCT speckle dynamics. By monitoring the dynamic changes of speckle patterns in OCT images in real time, it automatically adjusts the injection pressure, significantly improving the safety and accuracy of subretinal injection. It not only effectively prevents excessive blister expansion and retinal detachment but also adapts to different retinal pathologies and individual differences, exhibiting broad clinical applicability. Through automated control and real-time feedback mechanisms, this invention reduces human error, optimizes the surgical procedure, and improves surgical efficiency and repeatability. The technological innovation and clinical applicability of this invention have significant application value in the field of ophthalmic surgery.

[0035] The following will further explain the concept, specific structure, and technical effects of the present invention in conjunction with the accompanying drawings, so as to fully understand the purpose, features, and effects of the present invention. Attached Figure Description

[0036] Figure 1 This is a flowchart of an exemplary embodiment of a method for real-time closed-loop control of subretinal injection blister formation; Figure 2This is a schematic diagram of the dynamic process of retinal layer separation during the liquid injection process of the method for real-time closed-loop control of subretinal injection blister formation, an exemplary embodiment. Figure 3 This is a schematic diagram of speckle decorrelation quantification of micro-tissue deformation in a method for real-time closed-loop control of subretinal injection blister formation, an exemplary embodiment. Figure 4 This is a schematic diagram illustrating the dynamic process of adjusting injection pressure to control real-time speckle formation in a method for real-time closed-loop control of subretinal injection blister formation, an exemplary embodiment. Figure 5 This is a schematic diagram showing the gradual expansion of subretinal vesicles during injection using an open-loop control method. Figure 6 This is a schematic diagram illustrating the changes of subretinal vesicles during the injection process in a method for real-time closed-loop control of subretinal vesicle formation according to an exemplary embodiment. Figure 7 This is a comparison chart of speckle dynamic signal test results between the real-time closed-loop control method for subretinal injection blister formation and the open-loop control method, which are exemplary embodiments. Detailed Implementation

[0037] The following description, with reference to the accompanying drawings, illustrates several preferred embodiments of the present invention to make its technical content clearer and easier to understand. The present invention can be embodied in many different forms, and the scope of protection of the present invention is not limited to the embodiments mentioned herein.

[0038] In the accompanying drawings, components with the same structure are indicated by the same numerical designation, and components with similar structures or functions are indicated by similar numerical designations. The dimensions and thicknesses of each component shown in the drawings are arbitrary, and the present invention does not limit the dimensions and thicknesses of each component. To make the illustrations clearer, the thickness of components is schematically exaggerated in some places in the drawings.

[0039] The applicant designed a method for real-time closed-loop control of subretinal injection blister formation, such as... Figure 1 As shown, it includes the following steps: S100. System initialization and preparation: calibrate the OCT system, install the injection equipment, and locate the target area for subretinal injection; specifically including: S110, OCT system calibration: The OCT system is calibrated, including light source intensity calibration, scan path calibration, and image contrast calibration, to ensure imaging accuracy and resolution; S120. Preparation of injection equipment: Install the injection equipment, use the infusion pump, and set the initial injection pressure according to the clinical safety threshold to overcome the adhesion between the retina and the RPE. The range of the clinical safety threshold is 4 psi to 8 psi, and the initial injection pressure is set to 6 psi. S130. Target area localization: Determine the target area for subretinal injection using OCT imaging, and locate the injection point in the target area using OCT B-scan images.

[0040] S200, OCT image acquisition and speckle dynamic monitoring: Injection is performed into the target area under the retina. During the injection process, OCT B-scan images are acquired in real time, speckle dynamics are calculated, speckle dynamic curves are plotted, and real-time speckle dynamic signals are generated; specifically including: S210, Injection begins. Begin injecting the target area under the retina with the initial injection pressure. S220, OCT image acquisition: During the injection process, OCT B-scan images are acquired in real time at an acquisition frequency of 80kHz; S230. Dynamic calculation of speckle pattern: Calculate the speckle decorrelation value or speckle displacement; calculate the speckle decorrelation value to quantify the temporal change of speckle and reflect the microstructural changes of the tissue. The formula is as follows: ; in, Indicates the location and time t The speckle decorrelation value, Indicates the location and time t OCT signal strength, For time intervals; The speckle displacement, reflecting the mechanical deformation of the tissue, is calculated using the optical flow method, as shown in the following formula: ; in, The gradient of the OCT image. =( x, y) is the speckle displacement vector. This is the time derivative.

[0041] S240. Real-time feedback signal generation: Based on the speckle decorrelation value or speckle displacement, a speckle dynamic curve is plotted, generating a real-time speckle dynamic signal. The speckle dynamic curve is smoothed to reduce noise interference.

[0042] S300, Real-time injection pressure adjustment: When the real-time speckle dynamic signal reaches a preset threshold, the injection pressure is reduced; when the injection pressure is reduced to the level necessary to maintain stable blister expansion, the injection pressure is maintained; specifically including: S310, Injection pressure adjustment: When the real-time speckle dynamic signal reaches the preset threshold, the injection pressure is reduced. The preset threshold is set according to the statistical characteristics of speckle decorrelation value or speckle displacement. The preset threshold is set as the mean of speckle decorrelation value or speckle displacement plus twice the standard deviation. The injection pressure reduction is achieved through the flow control of the injection device. When the injection pressure is too high, the flow rate and flow of the injection pump are reduced to ensure stable expansion of the blister. The injection pressure to maintain stable blister expansion is 4 psi. S320, Injection pressure maintenance: When the injection pressure decreases to the level required to maintain stable blister expansion, maintain the injection pressure.

[0043] S400, Injection complete. When the real-time speckle dynamic signal rises or falls below the set fluctuation range threshold of 0.02 within 3 seconds, the injection is complete. Stop the injection, shut down the injection equipment and OCT system; otherwise, return to step S300. Specifically, this includes: S410. Injection complete. When the real-time speckle dynamic signal rises or falls less than the set fluctuation range threshold of 0.02 within 3 seconds, it indicates that the separation between the retina and RPE has been completed and no new separation has occurred. It is determined that the injection process has achieved the expected effect. Injection complete. Proceed to step S420. Otherwise, return to step S300. S420, System shutdown, stop injection, shut down injection equipment and OCT system.

[0044] To verify the effectiveness of this embodiment, the applicant conducted an experiment. The applicant used an optical coherence tomography (OCT) imaging system, an injection system, and a control system to implement the method for real-time closed-loop control of subretinal injection blister formation in this embodiment.

[0045] Specifically, optical coherence tomography (OCT) imaging systems include: The light source uses a superluminescent diode (SLD) with a center wavelength of 840 nm and a spectral bandwidth of 155 nm, achieving an axial resolution of approximately 2.5 µm. The interferometer uses a 50:50 fiber optic coupler to split the light into a sample arm and a reference arm; The detector, using a commercial OCT spectrometer, provides an imaging range of approximately 2.6 mm with an A-line rate of 80 kHz; The scanning module uses a two-axis galvanometer to achieve rapid scanning of OCT images.

[0046] The imaging lens uses an objective lens with a focal length of 50 mm to achieve a lateral resolution of approximately 30 µm.

[0047] The injection system includes: The syringe pump uses a micro-injection pump, which connects to a 2.5 mL syringe and connects to a computer via RS-485 protocol to achieve real-time pressure control; Use 25-gauge / 38-gauge retractable PolyTip injection needles to ensure precise injection location; The fluid delivery module delivers balanced salt solution (BSS) to the injection needle via a disposable infusion tubing.

[0048] The control system includes: A computer is used to control the OCT imaging system and the injection pump, process the speckle dynamic signal in real time, and adjust the injection pressure according to a preset threshold. The data acquisition card uses multi-functional I / O devices to generate galvanometer scanning signals and control the operation of the syringe pump; Supporting software, including dedicated software, is developed to display OCT images in real time, calculate speckle dynamic signals, generate feedback signals, and control injection pressure.

[0049] The experiment used an isolated pig eye model for verification, and the experimental steps are as follows: 1. Sample Preparation: Using fresh excised pig eyes, surrounding tissues (fat, muscle, and optic nerve) were removed via an open-window surgery, and the posterior ocular surface was adhered to a 3D-printed orbital socket. The anterior segment was removed through a scleral incision, the vitreous body was preserved, and the posterior segment was filled with saline to maintain structural integrity. 2. Experimental Procedure: The experiment was conducted according to both the open-loop control method and the steps outlined in this embodiment. 3. Experimental Process Analysis: Following this embodiment, the experiment utilized OCT speckle dynamic monitoring to achieve subretinal vesicle formation, such as... Figure 2 As shown, OCT B-scan recorded the dynamics of retinal layer separation during fluid injection from time t1 to t5. t1-t5 refers to five time points from the start of injection to the formation of a subretinal vesicle and the cessation of injection. At time t1, the needle tip entered the retina to begin injection; at time t2, injection continued at the original injection pressure; at time t3, a subretinal vesicle began to form, at which point the speckle dynamic signal reached a threshold, triggering a stop command from the computer-controlled infusion pump; at time t4, the infusion pump stopped injection; and at time t5, a stable state was reached after injection cessation. Microscopic tissue deformation was quantified using corresponding speckle decorrelation, such as... Figure 3As shown, at time t1, the needle tip enters the retina, and the tissue shows only slight local deformation. At time t2, as the liquid continues to be injected, the deformation spreads from the needle tip contact point to the surrounding area, and signs of tissue separation begin to appear. At time t3, subretinal vesicles begin to form, the deformation intensifies significantly, the area expands rapidly, and the deformation rate reaches its peak. At time t4, the injection is stopped, but the deformation continues, the area expands slowly, and the rate gradually slows down. At time t5, the deformation tends to stabilize, the vesicle morphology is fixed, and the tissue reaches a new equilibrium state. This embodiment adjusts the injection pressure to control the dynamic process of real-time speckle pattern, such as... Figure 4 As shown, during the injection process, the injection pressure is described by the black curve. The pressure is controlled by the injection pump to ensure that the liquid can be smoothly injected into the subretinal space. The blue curve reflects the real-time changes in speckle dynamics. The changes in speckle signal are used to monitor the microscopic deformation of retinal tissue. At times t1 and t2, the changes in speckle dynamic signal are not obvious. At time t3, due to the formation of subretinal vesicles (the start of retinal-retinopathy of prematurity separation), the speckle dynamic signal begins to increase. Between t3 and t4, the speckle dynamic signal reaches a preset threshold, triggering the injection pump to stop injection. At this time, the injection pressure gradually decreases. At time t5, the injection pump stops and tends to stabilize, avoiding over-injection that could lead to further deformation or damage to the retinal tissue.

[0050] 4. Evaluation of test results: The test results of the open-loop control method are as follows: Figure 5 As shown, the speckle decorrelation trajectory exhibits a pattern consistent with previous uncontrolled experiments. OCT B-scan images reveal that the subretinal vesicle gradually expands during injection, ultimately leading to retinal detachment (area indicated by the white box in the image). In this case, due to the lack of real-time feedback control, the injection process cannot be adjusted according to the actual separation of retinal tissue, resulting in excessive vesicle expansion and increasing the risk of retinal damage. The results of the experiment in this embodiment are as follows... Figure 6 As shown, the speckle dynamic signal uses speckle decorrelation value. Before injection, the system measures the speckle dynamic signal for the first 2 seconds, calculates the mean and standard deviation of the speckle decorrelation value, and sets a preset threshold as the mean plus twice the standard deviation. At time t2, the system detects that the speckle decorrelation value exceeds the preset threshold and immediately triggers an injection stop. Subsequent OCT imaging confirms that the blister morphology is stable and there is no further volume change (the area shown in the white box in the figure). This indicates that closed-loop feedback control can effectively prevent excessive blister expansion and avoid the risk of retinal detachment. The results of experiments conducted using the open-loop control method and this embodiment are compared as follows: Figure 7As shown, the blue curve (with control) represents the speckle dynamic signal in this embodiment, while the red curve (without control) represents the speckle dynamic signal under the open-loop control method. Under open-loop control, the speckle decorrelation trajectory shows a gradually increasing trend, indicating that the bubble is continuously expanding, eventually leading to retinal detachment. Under closed-loop feedback control, when the system detects that the decorrelation signal reaches the threshold at time t2, it immediately triggers the injection pump to stop injecting. The speckle decorrelation signal stabilizes, and the bubble morphology remains stable without further volume changes, avoiding over-injection that could cause retinal tissue deformation or damage. Comparative experiments demonstrate that this embodiment can complete speckle decorrelation and displacement calculations in a short time, thereby achieving rapid pressure feedback control.

[0051] The preferred embodiments of the present invention have been described in detail above. It should be understood that those skilled in the art can make numerous modifications and variations based on the concept of the present invention without creative effort. Therefore, all technical solutions that can be obtained by those skilled in the art based on the concept of the present invention through logical analysis, reasoning, or limited experimentation on the basis of existing technology should be within the scope of protection defined by the claims.

Claims

1. A method for real-time closed-loop control of subretinal injection bubble formation, characterized in that, The steps include: S100. System initialization and preparation: calibrate the OCT system, install the injection equipment, and locate the target area for subretinal injection. S200, OCT image acquisition and speckle dynamic monitoring, injecting into the target area of ​​the subretinal injection, during the injection process, real-time acquisition of OCT B-scan images, calculation of speckle dynamics, plotting of speckle dynamic curves, and generation of real-time speckle dynamic signals. S300, real-time adjustment of injection pressure: when the real-time speckle dynamic signal reaches a preset threshold, the injection pressure is reduced; when the injection pressure is reduced to the injection pressure that maintains stable blister expansion, the injection pressure is maintained. S400, Injection complete. When the real-time speckle dynamic signal meets the injection completion condition, if the rise or fall is less than the set fluctuation range threshold within 3 seconds, the injection is complete, the injection is stopped, and the injection device and the OCT system are turned off; otherwise, return to step S300.

2. The method for real-time closed-loop control of subretinal injection blister formation as described in claim 1, characterized in that, Step S100 includes: S110, OCT system calibration: The OCT system is calibrated to ensure imaging accuracy and resolution; S120. Prepare the injection equipment, install the injection equipment, and set the initial injection pressure; S130, Target area localization: Determine the target area for subretinal injection using OCT imaging, and locate the injection point in the target area for subretinal injection using the OCT-scan image.

3. The method for real-time closed-loop control of subretinal injection bubble formation as described in claim 2, characterized in that, The calibration of the OCT system includes: light source intensity calibration, scan path calibration, and image contrast calibration.

4. The method for real-time closed-loop control of subretinal injection blister formation as described in claim 2, characterized in that, The initial injection pressure is set according to a clinical safety threshold to overcome the adhesion between the retina and the RPE.

5. The method for real-time closed-loop control of subretinal injection bubble formation as described in claim 1, characterized in that, Step S200 includes: S210, Injection begins, and injection is initiated into the target area of ​​subretinal injection using the initial injection pressure; S220, OCT image acquisition: During the injection process, the OCT B-scan images are acquired in real time according to the acquisition frequency; S230, Dynamic calculation of speckle pattern, calculating speckle decorrelation value or speckle displacement: S240. Real-time feedback signal generation: Based on the speckle decorrelation value or the speckle displacement, the speckle dynamic curve is plotted to generate the real-time speckle dynamic signal.

6. The method for real-time closed-loop control of subretinal injection bubble formation as described in claim 5, characterized in that, In step S230, the speckle decorrelation value is calculated to quantify the temporal change of speckle and reflect the changes in the microstructure of the tissue. The formula is as follows: ； in, Indicates the location and time t The speckle decorrelation value, Indicates the location and time t OCT signal strength, For time intervals.

7. The method for real-time closed-loop control of subretinal injection bubble formation as described in claim 5, characterized in that, In step S230, the speckle displacement is calculated using the optical flow method, and the formula is as follows: ； in, The gradient of the OCT image. =( x, y) is the speckle displacement vector. This is the time derivative.

8. The method for real-time closed-loop control of subretinal injection bubble formation as described in claim 1, characterized in that, Step S300 includes: S310, Injection pressure adjustment: When the real-time speckle dynamic signal reaches the preset threshold, the injection pressure is reduced. S320, Injection pressure maintenance: When the injection pressure decreases to the injection pressure required to maintain stable blister expansion, the injection pressure is maintained.

9. The method for real-time closed-loop control of subretinal injection blister formation as described in claim 1, characterized in that, The preset threshold is set based on the speckle decorrelation value or the statistical characteristics of the speckle displacement.

10. The method for real-time closed-loop control of subretinal injection bubble formation as described in claim 1, characterized in that, Step S400 includes: S410. Injection complete. When the real-time speckle dynamic signal meets the injection completion condition, it indicates that the separation between the retina and RPE has been completed and no new separation has occurred. It is determined that the injection process has achieved the expected effect and the injection is complete. Execute step S420. Otherwise, return to step S300. S420. System shutdown, injection stopped, injection device and OCT system shut down.

Images