A dual-axis intelligent dynamic measurement and health management method for intraocular pressure

Abstract

This invention belongs to the technical field of intelligent healthcare and relates to a dual-axis intelligent sensing method for dynamic measurement and health management of intraocular pressure (IOP). The method includes the following steps: Preliminary screening of fused physiological signals containing continuous IOP data and pressure state data within a synchronous time period; identification of the preliminary screening of fused physiological signals to generate an IOP event profile characterizing risk factors; analysis of the IOP event profile to generate a comprehensive intervention profile including coping strategies, with differentiated behavioral intervention instructions; execution of the comprehensive intervention profile to generate real-time feedback data on the intervention's effects; evaluation of the real-time feedback data on the intervention's effects to generate graded drug release decisions; and execution of the graded drug release decisions to generate proportionalized drug release control signals. This invention solves the problem that timed instillation cannot accurately match actual IOP peaks, potentially leading to drug waste and accumulated side effects when IOP is normal.

Description

Technical Field

[0001] This invention belongs to the technical field of intelligent medical care and relates to a dual-axis intelligent sensing method for dynamic measurement and health management of intraocular pressure. Background Technology

[0002] The monitoring and treatment of intraocular pressure-related diseases such as glaucoma face severe challenges. The core issue lies in the contradiction between the dynamic fluctuations of intraocular pressure and the static limitations of traditional monitoring methods. Intraocular pressure is not only affected by pathological factors, but also closely related to an individual's emotions, stress, daily routines, and other physiological activities. Traditional clinic intraocular pressure measurements can only provide sporadic time-point data and cannot capture key intraocular pressure peaks in daily life. This makes it difficult for doctors to comprehensively assess the condition and treatment effects, thus making it impossible to develop truly personalized treatment plans and form a complete diagnosis and treatment loop.

[0003] For intraocular pressure management, the industry's current common solution is to administer eye drops according to a doctor's prescription at fixed times and in fixed quantities. This is a passive and non-personalized treatment model, and patients need to strictly follow a fixed schedule to take the medication on their own. Continuous monitoring of intraocular pressure mainly relies on some emerging smart contact lenses that are in the research or early clinical stages. Although these devices can achieve continuous data collection, their functions are usually relatively simple, limited to monitoring intraocular pressure itself, and lack the ability to deeply analyze the causes of intraocular pressure fluctuations.

[0004] Based on the above problems, timed instillation of intraocular pressure cannot be precisely matched with the actual peak of intraocular pressure. This may lead to drug waste and accumulation of side effects when the intraocular pressure is normal, while it cannot provide timely intervention when the intraocular pressure rises sharply, resulting in low treatment efficiency and risks. Summary of the Invention

[0005] To address the aforementioned problems, this invention provides a dual-axis intelligent sensing method for dynamic measurement and health management of intraocular pressure.

[0006] A dual-axis intelligent dynamic measurement and health management method for intraocular pressure includes the following steps:

[0007] S1. Obtain the preliminary screened fusion physiological signals, which include continuous intraocular pressure data and pressure state data within the synchronous time period;

[0008] S2. Identify the initially screened fused physiological signals to generate an intraocular pressure event profile characterizing the risk factors;

[0009] S3. Analyze intraocular pressure event profiles to generate comprehensive intervention profiles containing coping strategies, with differentiated behavioral intervention instructions included in the comprehensive intervention profiles;

[0010] S4. Implement comprehensive intervention records to generate real-time feedback data on the post-intervention effects;

[0011] S5. Real-time feedback data to evaluate the effects of intervention in order to generate graded drug release decisions;

[0012] S6. Execute graded drug release decisions to generate proportional drug release control signals.

[0013] A further aspect of the present invention involves initially screening fused physiological signals that include continuous intraocular pressure data and pressure state data within a synchronous time period, comprising the following steps:

[0014] High signal-to-noise ratio continuous intraocular pressure data is obtained by using an asymmetric Wheatstone bridge structure built into a flexible contact lens. The structure arranges a deformation-sensitive linear measurement arm along the corneal ring towards the peripheral area and a deformation-insensitive serpentine reference arm along the radial and central areas of the cornea to physically cancel common-mode interference.

[0015] Simultaneously, adrenaline concentration data reflecting acute stress and cortisol concentration data reflecting chronic stress were collected using skin sweat patches to constitute stress state data.

[0016] Peak values ​​exceeding a preset safety baseline in continuous intraocular pressure data are marked, and intraocular pressure data within a specified time period before and after the peak value are extracted along with pressure status data within the same time period. After aligning the timestamps, preliminary fusion physiological signals are generated.

[0017] A further aspect of the present invention, for generating an intraocular pressure event profile characterizing the causes of risk, includes the following steps:

[0018] Analyze the continuous intraocular pressure data in the initially screened fused physiological signals and classify their peak morphology into sharp narrow peak morphology or gentle wide peak morphology.

[0019] By analyzing the stress state data in the initially screened fused physiological signals, the rapid rising inflection point of the adrenaline concentration curve and the sustained high plateau period of the cortisol concentration curve were identified.

[0020] The peak shape is temporally correlated with the rapid rising inflection point of the adrenaline concentration curve or the sustained high plateau period of the cortisol concentration curve. If the peak shape is a sharp and narrow peak and is closely associated with the rapid rising inflection point, an acute stress profile is generated as an intraocular pressure event profile. If the peak shape is a flat and wide peak and is synchronized with the sustained high plateau period, a chronic cumulative profile is generated as an intraocular pressure event profile.

[0021] A further aspect of this invention involves integrating differentiated behavioral intervention instructions into the intervention file, including the following steps:

[0022] When the intraocular pressure event profile is an acute stress profile, a sequence of instructions to guide the user to perform deep breathing or close-eye meditation is generated and pushed through the user terminal's interactive interface as the first level of behavioral intervention instructions.

[0023] When the intraocular pressure event profile is a chronic cumulative profile, lifestyle suggestions including adjusting work and rest and planning long-term stress reduction are generated and pushed as a second-level behavioral intervention instruction;

[0024] The first-level or second-level behavioral intervention instructions are linked to the corresponding intraocular pressure event profile to form a comprehensive intervention profile containing coping strategies.

[0025] A further aspect of the present invention, for generating real-time feedback data on the post-intervention effect, includes the following steps:

[0026] After executing the first-level behavioral intervention instructions, an independent observation timer is started, and changes in intraocular pressure and acute stress state are continuously monitored during this period to generate real-time feedback data on the post-intervention effect.

[0027] While implementing the second-level behavioral intervention instructions, the risk level of the chronic cumulative profile is raised, and this status is recorded in the real-time feedback data of the intervention effect.

[0028] A further aspect of the present invention for generating graded drug release decisions includes the following steps:

[0029] If the real-time feedback data of the intervention shows that the intraocular pressure or acute stress state has not returned to the preset safety baseline when the independent observation timer ends, the behavioral intervention is deemed ineffective and a first-level drug release decision is generated.

[0030] Based on the increased risk level recorded in the real-time feedback data of the intervention effect, it is directly determined that drug intervention is needed, and a secondary drug release decision is generated.

[0031] A further aspect of the present invention, for generating graded drug release decisions, further includes the following steps:

[0032] Before generating a primary or secondary drug release decision, check whether the cumulative drug dose within 24 hours has reached the preset maximum total dose within 24 hours.

[0033] If the cumulative drug dose within 24 hours reaches the maximum total dose limit within 24 hours, a drug release lock function is generated and a safety lock decision is made to force a medical visit, replacing the graded drug release decision.

[0034] A further aspect of the present invention, for generating a proportionalized drug release control signal, includes the following steps:

[0035] Upon receiving a first-level drug release decision, the system waits for the natural decrease in tear pH due to the continued stress state to trigger the release of an initial small dose of drug from the pH-responsive hydrogel containing the drug. The control logic of this process is the first-level proportional drug release control signal.

[0036] A further aspect of the present invention, for generating a proportionalized drug release control signal, further includes the following steps:

[0037] Upon receiving a secondary drug release decision, a microcurrent of a specific frequency is applied through a wireless microcoil inside the contact lens. This microcurrent, in conjunction with the decrease in tear pH, acts on the dual-response hydrogel to trigger the release of a second-level, higher dose of drug. The control logic of this process is the second-level proportional drug release control signal.

[0038] In summary, the present invention has the following beneficial technical effects:

[0039] 1. By integrating intraocular pressure (IOP) with multidimensional pressure biomarkers, we can achieve a precise profile of the causes of IOP elevation events, fundamentally changing the situation of traditional monitoring that only knows what happens but not why. It can effectively distinguish between transient IOP fluctuations caused by acute stress and persistent high IOP risk caused by chronic pressure accumulation. This etiology-based classification makes subsequent intervention measures more targeted, avoiding the crude approach of using a single drug to treat all IOP elevations, thus achieving a leap from passive monitoring to active cause identification.

[0040] 2. For acute stress-induced intraocular pressure elevation, non-invasive behavioral interventions such as relaxation guidance are prioritized, and drug therapy is only escalated when these interventions are ineffective. This not only reduces unnecessary drug use, lowers the risk of side effects and the patient's financial burden, but also reflects a patient-centered treatment philosophy. The system's built-in total drug dosage safety lock mechanism provides ultimate protection for the medication safety of the closed-loop treatment system, effectively preventing the risk of drug overdose due to algorithm or equipment malfunction.

[0041] 3. The designed proportional drug release mechanism demonstrates a high degree of intelligence and responsiveness. It is no longer a simple "on" or "off" operation, but rather executes different "levels" of drug release strategies based on the severity of the risk assessment. Whether it's low-dose sustained release triggered naturally by physiological changes or high-dose rapid release triggered by active electrical signals in conjunction with physiological signals, it achieves dynamic matching of drug dosage with the patient's needs. This refined drug delivery method maximizes therapeutic efficacy while minimizing drug exposure. Attached Figure Description

[0042] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. The drawings are used to provide a further understanding of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0043] Figure 1 A flowchart illustrating an embodiment of this application is disclosed.

[0044] Figure 2 Structural schematic diagrams of embodiments of this application are disclosed. Detailed Implementation

[0045] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0046] The following is in conjunction with the appendix Figures 1-2 A preferred description of the present invention is provided below.

[0047] See attached document Figure 1 This invention proposes a dual-axis intelligent sensing method for dynamic measurement and health management of intraocular pressure, comprising the following steps:

[0048] S1. Obtain the preliminary screened fusion physiological signals, which include continuous intraocular pressure data and pressure state data within the synchronous time period;

[0049] S2. Identify the initially screened fused physiological signals to generate an intraocular pressure event profile characterizing the risk factors;

[0050] S3. Analyze intraocular pressure event profiles to generate comprehensive intervention profiles containing coping strategies, with differentiated behavioral intervention instructions included in the comprehensive intervention profiles;

[0051] S4. Implement comprehensive intervention records to generate real-time feedback data on the post-intervention effects;

[0052] S5. Real-time feedback data to evaluate the effects of intervention in order to generate graded drug release decisions;

[0053] S6. Execute graded drug release decisions to generate proportional drug release control signals.

[0054] In one embodiment of the present invention, step S1 includes the following steps:

[0055] By incorporating an asymmetric Wheatstone bridge structure within the flexible contact lens, the structure arranges a deformation-sensitive linear measurement arm along the corneal ring towards the peripheral region, and a deformation-insensitive serpentine reference arm along the corneal radial and central regions. This physically cancels out common-mode interference generated by actions such as blinking and rubbing the eyes, thereby obtaining continuous intraocular pressure data with a high signal-to-noise ratio.

[0056] Simultaneously, adrenaline concentration data reflecting acute stress and cortisol concentration data reflecting chronic stress were collected using a skin sweat patch to constitute stress state data. Peak values ​​exceeding a preset safety baseline in continuous intraocular pressure data were marked. Intraocular pressure data within a specified time period before and after this peak value were extracted and time-stamped with the stress state data of the same time period to generate a preliminary fused physiological signal that integrates intraocular pressure changes and multidimensional stress states.

[0057] Specifically, the user first needs to wear a flexible contact lens with integrated sensing capabilities and attach a skin-sweat patch to a suitable area of ​​the body, such as the wrist or chest. Once activated, both devices begin working synchronously. The key to the flexible contact lens is its asymmetric Wheatstone bridge structure. This structure consists of two sets of resistive arms: one set is a linear measuring arm, highly sensitive to physical stretching deformation, precisely positioned in the peripheral circumferential region of the cornea, where the most significant stretching deformation occurs when intraocular pressure increases, leading to increased corneal curvature. The other set is a serpentine reference arm, whose curved, serpentine design results in minimal length change under external force, making it insensitive to deformation. These are positioned in the radial and central regions of the cornea, areas relatively less prone to deformation due to changes in intraocular pressure. When the user blinks or rubs their eyes, the resulting pressure acts evenly across the entire cornea, causing similar changes in the resistance of the measuring and reference arms. This shared change, known as common-mode interference, is physically canceled out in the Wheatstone bridge circuit. When intraocular pressure increases, only the resistance of the linear measuring arm, which is sensitive to deformation, changes significantly. The output is an electrical signal that reflects the true change in intraocular pressure. After calibration and conversion, continuous intraocular pressure data with a high signal-to-noise ratio is formed.

[0058] Meanwhile, the sweat patch continuously monitors two key biomarkers in sweat using built-in electrochemical sensors. It measures adrenaline concentration, reflecting the user's instantaneous, acute stress state, and cortisol concentration, characterizing long-term, chronic stress levels. These two sets of data together constitute the stress status data. Finally, the central processing unit receives and processes the aforementioned two data streams. The processing unit continuously scans the continuous intraocular pressure data, and if any data point exceeds a preset safety baseline, it marks that data point as an abnormal peak.

[0059] The system automatically extracts all intraocular pressure (IOP) data, including the peak value and a specified time period before and after it. Based on the timestamps of this IOP data, the system finds the exact corresponding time period within the synchronously acquired stress state data, and also extracts adrenaline and cortisol concentration data within the same time period. By precisely aligning the extracted IOP and stress state data segments in time, the system fuses them into a single data record. This record serves as the initial filtered fused physiological signal, providing accurate and comprehensive raw data for subsequent risk analysis.

[0060] Flexible contact lenses are transparent lenses made of medical-grade hydrogel or silicone hydrogel, worn directly on the surface of the cornea, and contain embedded microelectronic components. The asymmetric Wheatstone bridge structure is a microcircuit built into the flexible contact lens. The four resistive arms form the bridge, characterized by different response capabilities of the measuring and reference arms to deformation, used to convert physical deformation into a measurable voltage signal. The linear measuring arm is the resistance wire used as a sensing element in the bridge; its linear shape causes a significant change in resistance when stretched, functioning to detect minute deformations of the cornea. The peripheral corneal region, the ring-shaped area at the edge of the cornea, experiences the most significant deformation when intraocular pressure increases, making it an ideal location for deploying the measuring arm.

[0061] The serpentine reference arm is a resistance wire used as a reference in a bridge circuit. Its meandering geometry makes it difficult to stretch under stress, and its function is to provide a stable resistance reference to compensate for signal drift caused by factors other than intraocular pressure. The radial and central regions of the cornea refer to the areas radiating outwards from the center of the cornea and the very center of the cornea; these locations experience relatively small deformations. Common-mode interference refers to interference signals unrelated to the measurement target that act simultaneously on both the measuring arm and the reference arm, such as the pressure signal generated by blinking. High signal-to-noise ratio continuous intraocular pressure data is a digital sequence acquired at a fixed time frequency, with an effective signal strength much greater than the background noise. The data structure is a series of (timestamp, intraocular pressure value) tuples. A sweat patch is a thin, flexible wearable device that adheres to the skin surface, enabling the collection and analysis of chemical components in trace amounts of sweat using microfluidic technology. Adrenaline concentration data reflecting acute stress is a data sequence of (timestamp, adrenaline concentration value) used to quantify the level of stress or excitement experienced by a user over a short period.

[0062] Cortisol concentration data reflecting chronic stress is a sequence of (timestamp, cortisol concentration value) data used to assess a user's physiological and psychological stress levels over a longer period (hours to days). Stress status data is a dataset containing the aforementioned adrenaline and cortisol concentration data, structured as a triplet of (timestamp, adrenaline concentration value, cortisol concentration value). The preset safety baseline is a constant intraocular pressure threshold, such as 21 mmHg, set based on the clinically recognized upper limit of normal intraocular pressure, used for preliminary assessment of whether intraocular pressure is abnormal.

[0063] The specified duration is a preset time window length, such as 30 seconds before and after. This setting is based on ensuring complete capture of the entire process of typical intraocular pressure fluctuation events while avoiding the introduction of excessive irrelevant data. It is set based on statistical analysis of intraocular pressure fluctuation events in 500 groups of glaucoma patients. The initially screened fused physiological signals are structured data blocks containing all relevant information during abnormal intraocular pressure events. The data is organized as sets, which contain truncated sequences of (timestamp, intraocular pressure value) and truncated sequences of (timestamp, adrenaline concentration value, cortisol concentration value).

[0064] For example, suppose that at 14:30:00, the user's flexible contact lens collected a high signal-to-noise ratio continuous intraocular pressure (IOP) of 18 mmHg. Simultaneously, a sweat patch collected pressure status data showing an adrenaline concentration of 50 pg / mL and a cortisol concentration of 10 ng / mL. At 14:30:45, the continuous IOP rose to 25 mmHg, exceeding the preset safety baseline of 21 mmHg. The system immediately marked this as a peak and, according to a specified duration of 30 seconds before and after, extracted continuous IOP data from 14:30:15 to 14:31:15. Simultaneously, the system also extracted pressure status data within the same time period, at which point the adrenaline concentration might have surged to 200 pg / mL. By integrating these two 60-second intervals with perfectly aligned timestamps, a preliminary fused physiological signal is generated. This signal records the complete process from normal IOP to an abnormal surge and then a decline, along with the simultaneous changes in stress hormones.

[0065] In one embodiment of the present invention, step S2 includes the following steps:

[0066] The peak shape of the preliminarily screened fused physiological signals and intraocular pressure data is analyzed and classified into sharp narrow peaks or gentle wide peaks. The changing trend of the preliminarily screened fused physiological signals and stress state data is analyzed to identify the rapid rising inflection point of the adrenaline concentration curve and the sustained high plateau period of the cortisol concentration curve. The peak shape of the intraocular pressure data and the changing trend of the stress state data are temporally correlated and matched. Based on the preset matching rules, an intraocular pressure event profile characterizing the risk factors is generated.

[0067] If the intraocular pressure data shows a sharp, narrow peak, and the timing of its appearance is closely related to the rapid inflection point of the adrenaline concentration curve, an acute stress profile is generated; if the intraocular pressure data shows a flat, wide peak or remains high, and is synchronized with the sustained high plateau phase of the cortisol concentration curve, a chronic cumulative profile is generated.

[0068] Specifically, the algorithm analyzes intraocular pressure data in the initially screened fused physiological signals, focusing on the shape characteristics of its peaks. By calculating the slope of the peak's upward trajectory and the duration of the peak's high position, the algorithm can classify the peak shape. Peaks with extremely steep slopes and short high-position durations are identified as sharp, narrow peaks; conversely, peaks with gentler slopes and longer high-position durations are identified as gentle, broad peaks. Using the same initially screened fused physiological signal, the algorithm then analyzes the changing trends of the stress state data, examining the curve of adrenaline concentration data over time. By identifying the point where the curve rises most rapidly, it pinpoints the rapid inflection point of the adrenaline concentration curve, representing the onset of acute stress.

[0069] The algorithm examines the cortisol concentration data curve to check for continuous time periods during which cortisol concentrations consistently remain above a preset high threshold, thus identifying a sustained high plateau in the cortisol concentration curve. The algorithm then performs time-series correlation matching on the analysis results from the first two steps. Based on preset matching rules, the algorithm determines the root cause of the intraocular pressure abnormality and generates a corresponding intraocular pressure event profile.

[0070] The first rule is that if the intraocular pressure data shows a sharp, narrow peak, and the time when this peak appears is almost synchronous with or closely connected to the inflection point of the rapid rise in the adrenal D concentration curve in terms of timestamps, then the system will characterize the event and generate an acute stress profile.

[0071] Rule two states that if the intraocular pressure data shows a flat, broad peak pattern, or if the intraocular pressure value remains at a high level, and the time period of this state coincides with the sustained high plateau period of the cortisol concentration curve, then the system generates a chronic cumulative profile.

[0072] The resulting intraocular pressure event profile is a data object with clear risk causal labels, providing a basis for decision-making for subsequent differentiated interventions.

[0073] Among them, intraocular pressure event profiles are structured data labels that not only record the occurrence of intraocular pressure events, but more importantly, classify and characterize their potential physiological triggers. The data structure can include records containing "profile type" (e.g., acute stress type or chronic cumulative type), "related timestamps," and "related physiological data indexes." Peak morphology describes the graphical characteristics of abnormally elevated intraocular pressure data and is a key indicator for distinguishing different types of intraocular pressure fluctuations. A sharp, narrow peak morphology indicates a rapid spike in intraocular pressure followed by a rapid drop, usually associated with sudden physiological stimulation. A gentle, broad peak morphology indicates a slow rise in intraocular pressure to a higher level, sustained for a period, and then a slow decline, often associated with sustained physiological stress.

[0074] The trend is a summary of the dynamic evolution of stress state data over time. The rapid rise in the adrenaline concentration curve is the point in time when the adrenaline concentration suddenly shifts from a stable or slow increase to a sharp rise, marking the indicative starting point of an acute stress response. The sustained high plateau in the cortisol concentration curve refers to a period where the cortisol concentration remains stable at a high level exceeding the normal range, a typical physiological manifestation of chronic stress accumulation. The preset matching rules are built-in logical judgment conditions used to infer risk types based on the temporal correlation between intraocular pressure and stress data, such as "when the time difference between event A (sharp narrow peak) and event B (adrenaline inflection point) is less than 5 seconds, it is classified as X (acute stress type)."

[0075] Acute stress profiles are a specific type of intraocular pressure (IOP) profile, clearly indicating that the elevation in IOP was triggered by a brief but intense stress event. Chronic cumulative profiles are another type of IOP profile, showing that elevated or persistently high IOP is related to long-term physiological or psychological stress accumulation.

[0076] For example, the generated preliminary filtered fused physiological signal records physiological data from 14:30:15 to 14:31:15, with intraocular pressure (IOP) peaking at 25 mmHg at 14:30:45. The analysis module first examines the peak shape of the IOP data, finding that the IOP surged from 18 mmHg to 25 mmHg in just a few seconds and then rapidly declined, thus classifying it as a sharp, narrow peak shape.

[0077] The module analyzed the synchronized intraocular pressure data and detected a rapid inflection point in the adrenaline concentration curve at 14:30:43, while the cortisol concentration data did not show a sustained high plateau during this period. Based on the preset matching rules, since the timing of the sharp, narrow peak in the intraocular pressure data (14:30:45) is closely correlated with the timing of the rapid inflection point in the adrenaline concentration curve (14:30:43), the system generated an acute stress profile, clearly labeling this intraocular pressure fluctuation event as being caused by acute stress.

[0078] In one embodiment of the present invention, step S3 includes the following steps:

[0079] When the intraocular pressure event profile is determined to be acute stress type, instead of triggering drug release, a sequence of instructions to guide the user to perform deep breathing or close-eye meditation is generated and pushed through the user terminal's interactive interface as the first level of behavioral intervention instructions.

[0080] When the intraocular pressure event profile is determined to be chronic cumulative, lifestyle recommendations including adjusting daily routines and planning long-term stress reduction are generated and pushed as second-level behavioral intervention instructions. The behavioral intervention instructions are linked to the corresponding intraocular pressure event profile to form a comprehensive intervention file containing coping strategies.

[0081] Specifically, upon receiving the generated intraocular pressure event profile, the system first determines its type. If the system identifies the intraocular pressure event profile as acute stress-related, it executes the first level of intervention. At this first level, the system actively inhibits the activation of the drug release mechanism and instead generates and pushes one or more specific instruction sequences through an application connected to the user's smartphone or other terminal device, on the user's terminal interface. These instruction sequences are carefully designed to guide the user to immediately perform behaviors that can quickly alleviate the acute stress response, such as instructing the user to perform deep diaphragmatic breathing exercises or close their eyes for several minutes of meditation. These instructions constitute the first level of behavioral intervention. Conversely, if the system identifies the intraocular pressure event profile as chronic cumulative, it initiates the second level of intervention.

[0082] Instead of pushing immediate relaxation instructions, the system generates more macro-level lifestyle recommendations focused on long-term health management. These recommendations may include adjusting sleep schedules, ensuring sufficient sleep, increasing physical activity, and planning long-term, personalized stress reduction programs, such as recommending yoga classes or psychological counseling services. These collectively constitute the second-level behavioral intervention instructions. To ensure the traceability and completeness of the intervention measures, the system firmly links the generated first- or second-level behavioral intervention instructions to the specific intraocular pressure event profile that triggered them. This linking process tightly connects the behavioral intervention instructions, risk factors, and original physiological data fluctuations, forming a data package. This data package is a comprehensive intervention profile containing coping strategies, providing a complete historical record for subsequent intervention effectiveness evaluation and closed-loop management.

[0083] Among these, behavioral intervention instructions are explicit instructions or suggestions conveyed to users digitally, aimed at guiding them to change their behavior to improve their health. The user terminal's interactive interface refers to the graphical interface through which users exchange information with smart devices, such as the application screen of a smartphone or the display of a smartwatch.

[0084] A sequence of instructions is a set of steps arranged in a specific order to guide a user through an action or task. For example, "Step 1: Inhale slowly, count to 4 silently; Step 2: Hold your breath, count to 7 silently; Step 3: Exhale slowly, count to 8 silently." Level 1 behavioral intervention instructions are immediate and highly actionable behavioral guidelines designed for acute, sudden health risks. Level 2 behavioral intervention instructions are long-term, planned lifestyle adjustment suggestions designed for chronic, cumulative health risks.

[0085] Lifestyle recommendations are comprehensive guidance covering various aspects such as daily life, diet, exercise, and psychological adjustment. A comprehensive intervention record is a structured data record that integrates the complete information chain of a specific health event, including the initial physiological signals, the analyzed risk profile, and the intervention strategy generated accordingly. Structurally, the data can be a composite object containing an intraocular pressure event profile and corresponding behavioral intervention instructions (text or code).

[0086] For example, following the generated acute stress profile, after determining it to be acute stress type, the first-level intervention is executed, without activating drug release. Instead, on the user's smartphone paired with the system, a push notification from the health management application sends a message to the user. Upon opening, the user's terminal interface displays a sequence of instructions, which reads: "Your intraocular pressure has been detected to have increased momentarily. Please follow these steps for 5 minutes of deep breathing relaxation."

[0087] The first step is to find a quiet and comfortable place to sit down... This instruction is the first level of behavioral intervention. The system then binds this instruction with the previously generated intraocular pressure event profile marked "14:30:45, acute stress type" to form a comprehensive intervention profile containing coping strategies, and stores it in the user's health log.

[0088] In one embodiment of the present invention, step S4 includes the following steps:

[0089] After executing the first-level behavioral intervention instruction, the system starts an independent observation timer and continuously monitors changes in intraocular pressure and acute stress state during this period, generating real-time feedback data on the post-intervention effect.

[0090] While executing the second-level behavioral intervention instruction, the system immediately raises the risk level of the chronic cumulative profile and records this status in the real-time feedback data of the intervention effect, preparing to initiate immediate drug intervention.

[0091] Specifically, after generating a comprehensive intervention profile containing coping strategies, the system enters the execution and feedback phase, generating real-time feedback data on the post-intervention effects. This step is divided into two different execution paths based on the instruction hierarchy contained in the comprehensive intervention profile. If the system executes a comprehensive intervention profile containing first-level behavioral intervention instructions, i.e., an intervention for acute stress events, then while pushing deep breathing or meditation instructions to the user, the system will immediately start an independent observation timer in its internal program. This timer is set to a fixed observation duration, such as five minutes. From the start of the timer until its end, the flexible contact lens and the skin sweat patch are instructed to continuously monitor the user's intraocular pressure and adrenaline concentration data reflecting the acute stress state in a high-frequency mode.

[0092] All collected intraocular pressure and adrenaline concentration readings are precisely timestamped and organized into a time-series dataset. This dataset constitutes the real-time feedback data on the post-intervention effects in this scenario. In another scenario, if the system executes a comprehensive intervention profile containing second-level behavioral intervention instructions—that is, an intervention targeting chronic cumulative events—the system will perform internal operations while pushing lifestyle suggestions such as adjusting daily routines to the user. This operation directly accesses and modifies the data record of the chronic cumulative profile bound to that instruction, raising the risk level field from its initial state to a higher level, such as from "Warning" to "High Risk." This risk level upgrade, the new level, and the timestamp of the operation are packaged into a status update record by the system.

[0093] This record was then defined as real-time feedback data on the post-intervention effect in this situation, and served as a clear signal that the system was ready to initiate immediate drug intervention without waiting for the effects of long-term behavioral intervention to become apparent.

[0094] The independent observation timer is a software timer, whose function is to define a specific time window for evaluating the effect of immediate behavioral interventions. The duration is usually set based on the average time for the relaxation response to take effect physiologically; for example, based on the test results of relaxation training in 300 healthy individuals, it is set to 5 minutes. The real-time feedback data on the post-intervention effect is a dataset used to record changes in physiological state after the intervention is initiated; the specific data structure varies depending on the type of intervention.

[0095] Following the first level of intervention, the data stream consists of a triplet sequence containing (timestamp, intraocular pressure value, adrenaline concentration value); following the second level of intervention, the data stream consists of a single-point status record containing information such as (timestamp, profile ID, new risk level). The risk level is a data label attached to the chronic cumulative profile to quantify the severity of the health risk it represents. Its data attributes are ordered categorical variables, such as "routine," "warning," and "high risk."

[0096] For example, in the generated comprehensive intervention profile containing first-level behavioral intervention instructions, after the system pushes the deep breathing instruction to the user, it starts a separate observation timer for 5 minutes, starting at 14:31:30. During these 5 minutes, intraocular pressure (IOP) and adrenaline concentration are continuously recorded. For instance, at 14:32:00, IOP is recorded as 23 mmHg and adrenaline as 150 pg / mL; at 14:34:00, IOP is recorded as decreasing to 20 mmHg and adrenaline as decreasing to 80 pg / mL. The set of all continuously recorded data points (timestamp, IOP, adrenaline) throughout the 5 minutes constitutes the real-time feedback data for the post-intervention effect, used for subsequent evaluation of the user's relaxation response.

[0097] In one embodiment of the present invention, step S5 includes the following steps:

[0098] For acute stress profiles, real-time feedback data on several prognostic effects show that if intraocular pressure or acute stress state has not returned to the safe baseline by the end of the observation timer, the behavioral intervention is deemed ineffective, and a first-level drug release decision is generated.

[0099] For chronic cumulative profiles, without waiting for the effects of behavioral intervention, the system directly determines the need for drug intervention based on the high-risk level records and generates a secondary drug release decision. All release decisions are limited to the total dose limit within 24 hours. Once the limit is reached, a safety lock decision is generated to lock the drug release function and forcibly remind patients to seek medical attention.

[0100] Specifically, for scenarios involving acute stress profiles, the decision engine first examines the real-time feedback data related to the intervention's post-intervention effects, which includes physiological data from the observation period. It locates the point in time when the independent observation timer ends and reads the intraocular pressure (IOP) and acute stress state indicator, namely adrenaline concentration, at that moment. The engine then compares these two values ​​to their respective preset safety baselines. If the IOP remains above the safety baseline, or the adrenaline concentration has not returned to normal levels, the system determines that the user's self-intervention was ineffective. Based on this determination, the decision engine generates a primary drug release decision.

[0101] In another scenario, namely for chronic cumulative risk profiles, the decision engine's logic is more direct. It reads real-time feedback data on the post-intervention effects generated in the previous step. This data records that the risk level of this profile has been raised to high risk. Without waiting for the user to implement lifestyle recommendations or assess their effects, the engine directly determines that the user needs medication intervention based on this recorded high-risk level. Accordingly, the decision engine generates a secondary drug release decision. Finally, to ensure medication safety, the system has a global safety monitoring mechanism. Whenever a primary or secondary drug release decision is generated, the system increments the corresponding drug dose within 24 hours using a total dose counter.

[0102] Before generating any new release decision, the system checks whether the total dose has reached the preset 24-hour total dose limit. Once the system detects that the total dose is equal to or exceeds this limit, it will no longer generate any drug release decision, but instead generate a special safety lock decision. This decision will immediately lock and disable the drug release function of the contact lens, and forcefully push an emergency reminder to the user's terminal, advising the user to contact a doctor or go to the hospital immediately.

[0103] Among them, graded drug release decisions are directive data objects that specify whether drug release is necessary, as well as the urgency and dosage level of the release. The data attributes are "no release," "Level 1 release," "Level 2 release," or "safety lockout." Level 1 drug release decisions are directives that authorize the initiation of a low-dose, sustained-release drug intervention program specifically for addressing acute intraocular pressure elevation following behavioral intervention failure. Level 2 drug release decisions are directives that authorize the initiation of a higher-dose, more potent drug intervention program directly addressing chronic cumulative intraocular pressure problems identified as high-risk.

[0104] The maximum total dose within 24 hours is a preset drug safety threshold, for example, set at 0.5 mg. This value is determined based on the pharmacokinetic characteristics of the drug used and clinical safe medication guidelines, aiming to prevent drug overdose and side effects. The safety lockout decision is the highest priority system safety instruction. Its function is to forcibly stop all drug release activities and guide the user to professional medical channels, serving as the final safety line for drug intervention.

[0105] For example, in an intervention targeting an acute stress profile, the independent 5-minute observation timer ends at 14:36:30. Real-time feedback data assessing the intervention's effect at this point shows an intraocular pressure of 22 mmHg, still above the safety baseline of 21 mmHg. The system determines the intervention is ineffective and generates a Level 1 drug release decision. Conversely, if the system previously identified a chronic cumulative profile and executed a Level 2 behavioral intervention, the real-time feedback data recording the intervention's effect would indicate a risk level of "high risk." Upon receiving this data, the assessment module, without any waiting time, directly generates a Level 2 drug release decision based on the "high risk" record.

[0106] In any scenario, assuming the dose corresponding to the first-level decision is 0.05 mg, the second-level decision is 0.1 mg, and the maximum total dose within 24 hours is 0.5 mg. If 0.45 mg of drug has been released cumulatively within the past 20 hours, and the system generates a second-level drug release decision, since 0.45 + 0.1 > 0.5, the system will not execute the second-level release. Instead, it will generate a safety lockout decision, locking the drug release function and reminding the user to seek immediate medical attention.

[0107] In one embodiment of the present invention, step S6 includes the following steps:

[0108] Upon receiving a first-level drug release decision, the system does not apply an external signal but waits for the tear pH to decrease naturally due to the continued stress state, in order to trigger the release of an initial small dose of drug by the pH-responsive hydrogel encapsulating the drug. The control logic of this process is the first-level proportional drug release control signal.

[0109] Upon receiving a secondary drug release decision, a microcurrent of a specific frequency is actively applied through a wireless microcoil inside the contact lens. This microcurrent, in conjunction with the decrease in tear pH, acts on the dual-response hydrogel to trigger the release of a second-level, higher dose of drug. The control logic of this process is the second-level proportional drug release control signal.

[0110] Specifically, when the microcontroller receives a primary drug release decision, it enters a passive waiting mode. In this mode, the system does not apply any active external signals. It relies on a physiological and chemical triggering mechanism, namely, that sustained acute stress usually causes the pH of human tears to naturally decrease, i.e., become more acidic. The drug reservoir layer encapsulating the drug uses a pH-responsive hydrogel. When the pH of the tear drops below a specific trigger threshold of the hydrogel, the molecular structure of the hydrogel changes from a compact state to a loose state, allowing the initially small dose of drug encapsulated inside to slowly permeate and be released onto the ocular surface. This drug release process, driven entirely by physiological changes and requiring no external energy input, has its own control logic defined as the first-level proportional drug release control signal.

[0111] When the microcontroller receives a secondary drug release decision, it initiates an active intervention procedure. This procedure first commands the wireless microcoil integrated inside the contact lens to activate, applying a microcurrent at a specific frequency. This microcurrent directly acts on the hydrogel of the drug reservoir. The drug reservoir uses a more advanced dual-responsive hydrogel, which is sensitive not only to pH levels but also to electrical signals of a specific frequency.

[0112] The decrease in tear pH caused by chronic stress, combined with the actively applied microcurrent, synergistically triggers a more rapid structural change in the hydrogel. This synergistic effect leads to the release of a second-level drug at a higher dose and faster rate than the first-level release. The underlying control logic of this complex process, which combines passive physiological triggering and active electrical signal control, is defined as a second-level proportional drug release control signal.

[0113] Proportional drug release control signals are a term describing the control method of drug release, meaning that the drug release dose and rate are proportionally adjusted according to the risk level, rather than a simple on / off mode. "Level 1" and "Level 2" represent two different proportional release levels. Acid-base responsive hydrogels are smart materials whose physical properties (such as swelling degree and permeability) reversibly change with changes in environmental pH, serving as a "valve" for drug release. The natural decrease in tear pH is a physiological response to stress; the normal pH of tears is approximately 7.4, but may drop below 7.0 under stress.

[0114] Wireless microcoils are miniature inductive components embedded within contact lenses, capable of receiving external commands and generating weak electromagnetic fields or currents. Microcurrents of a specific frequency refer to precisely set alternating current signals, their frequency optimized to most efficiently excite the response of dual-response hydrogels. Dual-response hydrogels are more advanced smart materials that simultaneously sense and respond to two or more different external stimuli (in this case, pH and current). Synergistic effects refer to the phenomenon where the combined effect of two factors is greater than the sum of their individual effects; in this case, it refers to the combined acceleration of drug release by an acidic environment and microcurrents.

[0115] The first level of proportionalized drug release control signal represents a low-dose, sustained-release control strategy, with the release amount determined by the hydrogel's natural response rate to physiological changes. The second level of proportionalized drug release control signal represents a high-dose, rapid-release control strategy, with the release amount and rate determined by both the intensity of the actively applied electrical signal and physiological changes, significantly higher than the first level.

[0116] For example, a first-level drug release decision is generated. Upon receiving this decision, the contact lens's microcontroller performs no action and enters a monitoring standby state. Assume the user's stress persists, causing their tear pH to drop from 7.4 to 6.9 after 10 minutes. This pH change triggers the pH-responsive hydrogel encapsulating the drug, causing its pores to open and slowly releasing 0.05 mg of glaucoma medication over the next 30 minutes. This process, relying on the natural decrease in tear pH to trigger the initial small-dose drug release, constitutes the complete first-level proportional drug release control signal.

[0117] Upon receiving a decision to generate a secondary drug release response, the microcontroller immediately drives a wireless microcoil to apply a 100 kHz microcurrent. At this time, the user's tear pH may also be low at 7.1 due to prolonged stress. Under the synergistic effect of the 100 kHz microcurrent and the acidic environment, the dual-response hydrogel rapidly disintegrates, releasing 0.1 mg of drug within just 5 minutes. This process of actively applying a microcurrent and synergistically decreasing tear pH to trigger the release of a second, higher dose of drug from the dual-response hydrogel is controlled by a second-level proportional drug release control signal.

[0118] See appendix Figure 2 This invention also proposes a dual-axis intelligent intraocular pressure dynamic measurement and health management system, comprising the following modules:

[0119] The fusion physiological signal acquisition module acquires preliminarily screened fusion physiological signals, which include continuous intraocular pressure data and pressure state data within the synchronous time period.

[0120] The intraocular pressure event profiling module identifies preliminarily screened fused physiological signals to generate an intraocular pressure event profile that characterizes the risk factors.

[0121] The differentiated behavioral intervention instruction generation module analyzes the intraocular pressure event profile to generate a comprehensive intervention profile containing coping strategies. The comprehensive intervention profile contains differentiated behavioral intervention instructions.

[0122] The comprehensive intervention record execution and feedback module executes the comprehensive intervention record to generate real-time feedback data on the post-intervention effects;

[0123] The post-intervention effect assessment and decision-making module evaluates real-time feedback data on the post-intervention effect in order to generate graded drug release decisions.

[0124] The graded drug release execution module performs graded drug release decisions to generate proportional drug release control signals.

[0125] Each of the modules can be implemented in whole or in part through software, hardware, or a combination thereof. It supports hardware embedded in or independent of the processor in the computer device, and also supports software stored in the memory of the computer device, so that the processor can call and execute the operations corresponding to each of the above modules.

[0126] It should be noted that the human information (including but not limited to human device information and personal information) and data (including but not limited to data used for analysis, data stored and data displayed) involved in this invention are all information and data authorized by the human body or fully authorized by all parties. The collection, use and processing of related data require relevant legal standards.

[0127] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A dual-axis intelligent sensing method for dynamic measurement and health management of intraocular pressure, characterized in that, Includes the following steps: S1. Obtain the preliminary screened fusion physiological signals, which include continuous intraocular pressure data and pressure state data within the synchronous time period; High signal-to-noise ratio continuous intraocular pressure data is obtained by using an asymmetric Wheatstone bridge structure built into a flexible contact lens. The structure arranges a deformation-sensitive linear measurement arm along the corneal ring towards the peripheral area and a deformation-insensitive serpentine reference arm along the radial and central areas of the cornea to physically cancel common-mode interference. Simultaneously, adrenaline concentration data reflecting acute stress and cortisol concentration data reflecting chronic stress were collected using skin sweat patches to constitute stress state data. Peak values ​​exceeding a preset safety baseline in continuous intraocular pressure data are marked, and intraocular pressure data within a specified time period before and after the peak value and pressure status data within the synchronous time period are extracted, and after aligning the timestamps, a preliminary fused physiological signal is generated. S2. Identify the initially screened fused physiological signals to generate an intraocular pressure event profile characterizing the risk factors; S3. Analyze intraocular pressure event profiles to generate comprehensive intervention profiles containing coping strategies, with differentiated behavioral intervention instructions included in the comprehensive intervention profiles; S4. Implement comprehensive intervention records to generate real-time feedback data on the post-intervention effects; S5. Real-time feedback data to evaluate the effects of intervention in order to generate graded drug release decisions; S6. Execute graded drug release decisions to generate proportional drug release control signals.

2. The method for dynamic measurement and health management of intraocular pressure with dual-axis intelligent sensing according to claim 1, characterized in that, To generate a profile of intraocular pressure events characterizing the causes of risk, the following steps are included: Analyze the continuous intraocular pressure data in the initially screened fused physiological signals and classify their peak morphology into sharp narrow peak morphology or gentle wide peak morphology. By analyzing the stress state data in the initially screened fused physiological signals, the rapid rising inflection point of the adrenaline concentration curve and the sustained high plateau period of the cortisol concentration curve were identified. The peak shape is temporally correlated with the rapid rising inflection point of the adrenaline concentration curve or the sustained high plateau period of the cortisol concentration curve. If the peak shape is a sharp and narrow peak and is closely associated with the rapid rising inflection point, an acute stress profile is generated as an intraocular pressure event profile. If the peak shape is a flat and wide peak and is synchronized with the sustained high plateau period, a chronic cumulative profile is generated as an intraocular pressure event profile.

3. The method for dynamic measurement and health management of intraocular pressure with dual-axis intelligent sensing according to claim 1, characterized in that, The comprehensive intervention file contains differentiated behavioral intervention instructions, including the following steps: When the intraocular pressure event profile is an acute stress profile, a sequence of instructions to guide the user to perform deep breathing or close-eye meditation is generated and pushed through the user terminal's interactive interface as the first level of behavioral intervention instructions. When the intraocular pressure event profile is a chronic cumulative profile, lifestyle suggestions including adjusting work and rest and planning long-term stress reduction are generated and pushed as a second-level behavioral intervention instruction; The first-level or second-level behavioral intervention instructions are linked to the corresponding intraocular pressure event profile to form a comprehensive intervention profile containing coping strategies.

4. The method for biaxial intelligent dynamic measurement and health management of intraocular pressure according to claim 1, characterized in that, To generate real-time feedback data on the effects of intervention, the following steps are included: After executing the first-level behavioral intervention instructions, an independent observation timer is started, and changes in intraocular pressure and acute stress state are continuously monitored during this period to generate real-time feedback data on the post-intervention effect. While implementing the second-level behavioral intervention instructions, the risk level of the chronic cumulative profile is raised, and this status is recorded in the real-time feedback data of the intervention effect.

5. The method for dynamic measurement and health management of intraocular pressure with dual-axis intelligent sensing according to claim 1, characterized in that, To generate a graded drug release decision, the following steps are included: If the real-time feedback data of the intervention shows that the intraocular pressure or acute stress state has not returned to the preset safety baseline when the independent observation timer ends, the behavioral intervention is deemed ineffective and a first-level drug release decision is generated. Based on the increased risk level recorded in the real-time feedback data of the intervention effect, it is directly determined that drug intervention is needed, and a secondary drug release decision is generated.

6. The method for biaxial intelligent dynamic measurement and health management of intraocular pressure according to claim 5, characterized in that, To generate a graded drug release decision, the following steps are also included: Before generating a primary or secondary drug release decision, check whether the cumulative drug dose within 24 hours has reached the preset maximum total dose within 24 hours. If the cumulative drug dose within 24 hours reaches the maximum total dose limit within 24 hours, a drug release lock function is generated and a safety lock decision is made to force a medical visit, replacing the graded drug release decision.

7. The method for biaxial intelligent dynamic measurement and health management of intraocular pressure according to claim 1, characterized in that, To generate a scaled-up drug release control signal, the following steps are included: Upon receiving a first-level drug release decision, the system waits for the natural decrease in tear pH due to the continued stress state to trigger the release of an initial small dose of drug from the pH-responsive hydrogel containing the drug. The control logic of this process is the first-level proportional drug release control signal.

8. The method for dynamic measurement and health management of intraocular pressure with dual-axis intelligent sensing according to claim 1, characterized in that, To generate a scaled-up drug release control signal, the method further includes the following steps: Upon receiving a secondary drug release decision, a microcurrent of a specific frequency is applied through a wireless microcoil inside the contact lens. This microcurrent, in conjunction with the decrease in tear pH, acts on the dual-response hydrogel to trigger the release of a second-level, higher dose of drug. The control logic of this process is the second-level proportional drug release control signal.

Images