An intervention system for screen-related dry eye syndrome

Abstract

This invention relates to an intervention system for dry eye syndrome associated with screen-related syndromes. The system includes a personal computer with a front-facing camera and a humidifier. The personal computer uses its front-facing camera to collect dynamic changes in the user's eye state. An analysis unit on the personal computer determines the user's total blink count based on these dynamic changes. The analysis unit then determines a composite index representing poor eye-use behavior associated with dry eye syndrome based on the number of incomplete blinks and blink frequency within the total blink count over a predetermined time period. The analysis unit determines intervention reminders based on changes in these composite indexes relative to baseline data and generates an intervention plan to drive the humidifier. The intervention plan is then sent to the humidifier, which performs humidification according to the intervention plan. This invention addresses the problem of failing to alleviate symptoms of dry eye syndrome associated with screen-related syndromes due to neglecting individual differences and environmental factors.

Description

Technical Field

[0001] This invention relates to the field of health data processing technology, and more particularly to technology for processing eye health data, specifically to an intervention system for dry eye syndrome related to screen syndrome. Background Technology

[0002] Numerous epidemiological reports on dry eye have been published worldwide, indicating an incidence rate of approximately 5% to 50% in different regions. For example, see the 2022 report "Research Progress of Artificial Intelligence in Dry Eye Diagnosis" by Han Xue, Ding Jingjuan, Lu Shuting, Jiang Qin, Yang Weihua, and Xue Jingsong (International Journal of Ophthalmology (12), 2063-2067); and the 2007 report "Research on the Impact of Video Display Terminals on the Visual System" by She Xijin, Huang Zhongning, Huang Duru, Yin Dongming, Wen Xianzong, Qiu Chuangyi, and Huang Li (Chinese Journal of Occupational Medicine (05), 392-394). The 2019 report "Does Digital Screen Exposure Cause Dry Eye?" by Mufti M., Sayeeed SI, Jaan I., and Nazir S. The study (see Indian Journal of Clinical Anatomy and Physiology, doi:10.18231 / 2394-2126.2019.0017) indicated that prolonged exposure to digital screens (especially smartphones and tablets) was significantly associated with the occurrence of dry eye symptoms among medical graduates, with an incidence rate of 55.6%.

[0003] The 2013 research article "Blink Rate, Incomplete Blinks and Computer Vision Syndrome" published by Portello, J., Rosenfield, M., & Chu, C. et al. (see Optometry and Vision Science, 90, 482–487) explored the relationship between Computer Vision Syndrome (CVS) and blink rate and incomplete blinks. The article found a significant positive correlation between the percentage of incomplete blinks and the total symptom score (p = 0.002), which means that the more incomplete blinks, the more severe the symptoms associated with dry eye. In particular, increasing the average blink rate to 23.5 times / minute through external cues did not significantly change the symptom score, indicating that the method of increasing blink rate to improve dry eye symptoms simply by adding external cues is ineffective.

[0004] A 2021 article by Al-Mohtaseb, Z., Schachter, S., Lee, B., Garlich, J., & Trattler, W. et al., entitled "The Relationship Between Dry Eye Disease and Digital Screen Use" (see Clinical Ophthalmology (Auckland, NZ), 15, 3811–3820), indicated that daily smartphone usage time (odds ratio [OR] = 1.86) and total daily digital screen usage time (OR = 1.82) were associated with an increased risk of dry eye disease. In contrast, daily computer or television usage time was not found to be associated with dry eye disease, based on a large-scale Korean study (N = 916) aged 7–12 years, in which smartphone usage prevalence was 96.7% among children with dry eye disease, compared to 55.4% among children without dry eye disease. The article also provides behavioral and environmental recommendations for preventing or alleviating dry eye symptoms in digital screen users, including blinking exercises, the "blind working" strategy, the 20-20-20 rule, and using a desktop humidifier. Humidifiers can alleviate dry eye symptoms and help prevent dry eye syndrome by reducing tear evaporation. Similarly, Mehra D. and Galor A., ​​in their 2020 paper "Digital Screen Use and Dry Eye: A Review" (see Asia-Pacific Journal of Ophthalmology, 9:491-497, doi:10.1097 / APO.0000000000000328), pointed out that blue light radiation from digital screens and abnormal blink rate are the main mechanisms of tear film evaporation-type dry eye, and suggested alleviating symptoms through blinking training and blue light filtering screens. However, for workers, frequent beeping sounds can interfere with their work, and they often choose to ignore the beeping sounds or even turn off the reminder function. Furthermore, the aforementioned studies also indicate that simply increasing blinking frequency does not actually improve dry eye symptoms.

[0005] CN116864105A discloses an artificial intelligence-based method for predicting the risk of dry eye syndrome. This method includes: acquiring a target facial video of a target object; inputting each image frame from the target facial video into a preset segmentation model to segment the eye region, obtaining an eye mask combination; calculating the blink frequency and incomplete blink data for each eye based on the target facial video and each eye mask combination; and determining the dry eye syndrome risk prediction result for each eye of the target object based on the blink frequency and incomplete blink data for each eye. This patent application details using the maximum eye-opening amplitude (i.e., the maximum distance between the upper and lower eyelids) to determine the presence of incomplete blinking, for example, when the maximum eye-opening amplitude is not 0, because at this time part of the eyeball is exposed outside the eyelid. The patent application also discloses the conditions for judging dry eye syndrome: a frequency condition and / or an incomplete blinking condition, further illustrating that the dry eye syndrome condition is less than 15 blinks per minute (i.e., the frequency condition), or an incomplete blinking frequency not being 0 blinks per second (i.e., the incomplete blinking condition). However, such simple condition settings lack any literature support and do not conform to objective reality. Furthermore, this technical solution assesses the risk of dry eye solely through blink frequency and incomplete blink data. It not only fails to consider the objective fact that subjects have extremely large differences in blink frequency, but also ignores the influence of test conditions (such as the content viewed and the content carrier), the influence of subjective consciousness on blink frequency and incomplete blink, and fails to mention the significance of the relationship between incomplete blink and blink frequency for the development of dry eye disease.

[0006] On the other hand, the technical solution disclosed in patent application CN116864105A is in obvious contradiction with existing research articles, such as "Marked reduction and distinct patterns of eye blinking in patients with moderately dry eyes during video display terminal use" published by Schlote, T., Kadner, G., & Freudenthaler, N. in 2004 (see Graefe's Archive for Clinical and Experimental Ophthalmology, 242, 306-312). This article investigated the spontaneous eye blink rate (SEBR). During conversations, the average SEBR was 16.8 blinks / minute. During initial use of a video display terminal (VDT), the SEBR significantly decreased to 6.6 ± 4.8 blinks / minute (P < 0.001). After 30 minutes of use, the SEBR was measured again at 5.9 ± 4.6 blinks / minute (P < 0.001), showing no significant difference compared to the initial VDT use period (P = 0.65). Furthermore, significant inter-individual variability in SEBR was observed during both conversations and VDT use. Therefore, this article concludes that the reduction in SEBR during VDT use is primarily determined by significant visual attention, rather than dry eye syndrome. In their 2011 paper, "Blink rate, blink amplitude, and tear film integrity during dynamic visual display terminal tasks" (Curr Eye Res. 2011; 36(3):190–197, doi:10.3109 / 02713683.2010.544442), Cardona G, Garcia C, Seres C, Vilaseca M, and Gispets J. pointed out that the influence of blink rate, blink amplitude, and incomplete blink on the presented content is not negligible. For example, blink rate drops to nearly 1 / 3 and 1 / 2 of baseline levels during fast-paced and slow-paced games, respectively, while the proportion of incomplete blink is even greater during dynamic tasks. The research content of the above two articles indicates that there is currently no unified standard for evaluating dry eye based on blink rate and incomplete blink rate, nor has any related research shown that a fixed condition can be used to evaluate the risk of dry eye in most people.The differences in blinking patterns and their impact on tear film stability in different studies are complex and varied, making it difficult to apply a single, fixed condition to the diagnosis of dry eye in most populations.

[0007] Furthermore, on the one hand, there are differences in understanding among those skilled in the art; on the other hand, the applicant studied a large number of documents and patents when making this invention, but due to space limitations, not all details and contents were listed in detail. However, this does not mean that the present invention does not possess the features of these prior art. On the contrary, the present invention already possesses all the features of the prior art, and the applicant reserves the right to add relevant prior art to the background art. Summary of the Invention

[0008] Relying solely on blink frequency or incomplete blink frequency as a diagnostic criterion for dry eye syndrome lacks sufficient reliability and universality. Blinking is a complex reflex activity involving both the central and peripheral nervous systems, encompassing multiple levels of the autonomic and somatic nervous systems. The autonomic nervous system plays a crucial role in tear secretion and ocular surface protection mechanisms. Composed of the sympathetic and parasympathetic nervous systems, these two components work synergistically in regulating lacrimal gland function. The parasympathetic nervous system promotes blinking by increasing lacrimal secretion, helping to lubricate the eyes. The sympathetic nervous system, on the other hand, reduces tear secretion and blink frequency under stress to maintain alertness. Blinking is also controlled by the somatic nervous system, particularly neural circuits in the cerebral cortex and brainstem. The cerebral cortex can consciously control the frequency and intensity of blinking. Specifically, the cerebral cortex can regulate intentional blinking, such as reducing blink frequency when focusing attention is required. The cerebral cortex (especially the primary motor cortex) controls blinking through neuronal connections. These neurons send signals that are transmitted to the brainstem via the corticobulbar tract, thus influencing blinking. The prefrontal cortex (especially the dorsal prefrontal cortex) is involved in planning and executing complex motor sequences, including conscious blinking, and can regulate the timing and frequency of blinks. Furthermore, the cerebral cortex exhibits neuroplasticity, allowing for adjustments to blinking control through learning and experience. For example, practice can improve the coordination and accuracy of blinking. Regarding blinking parameters, subjects subjectively control eye closure to better complete tasks; that is, the number of blinks is subjectively adjusted by the subject (see Li Yifan's 2023 Master's thesis, "A Study on Visual Fatigue in Naked-Eye 3D Dynamic Display Based on Human Eye Visual Characteristics," Beijing University of Posts and Telecommunications).

[0009] Existing technologies typically rely on parameters such as blink frequency to directly determine whether a user suffers from dry eye syndrome and propose solutions based on the results. For example, CN118155803A proposes a solution for correcting eye habits in relation to dry eye syndrome. However, this solution suffers from problems such as overly complex data sources, making it impractical, especially given its high technical difficulty. For instance, data collection primarily involves monitoring the eyelash overlap by taking photos from the side of the eyeglass frame to obtain blink frequency and amplitude data. This method may be affected by various factors in practical applications, such as the wearer's head movements and changes in ambient light, which could impact the accuracy and reliability of the data. Furthermore, ensuring data accuracy requires specially designed eyeglass frames and a matching sensor system, increasing product cost and technical barriers, making it difficult for ordinary users to accept or afford. In addition, its data processing is extremely complex, integrating and analyzing data from different sources (such as blink frequency, near-vision distance, and air humidity), requiring not only powerful computing capabilities but also complex algorithms to ensure the correct interpretation of the correlations between the data. Furthermore, the reminder mechanism of this patent application is overly complex, containing numerous meaningless reminders, leading to low patient compliance. This can also cause "reminder fatigue," causing users to ignore truly important reminders and intervention information. In addition, it lacks a personalized adjustment mechanism, failing to mention how to customize reminder settings based on individual circumstances. This means all users receive the same reminder strategy, ignoring individual differences and reducing user experience; it also increases the psychological burden on patients, which is detrimental to maintaining good eye habits. Worse still, the patent application does not adjust for personalized baselines in its eye behavior monitoring data, resulting in excessive false alarms. This is because dry eye symptoms vary from person to person, and a universal standard cannot be applied to everyone. Without establishing personalized baselines based on individual characteristics (such as occupation), it is difficult to distinguish between normal and abnormal eye behaviors, easily leading to misjudgments. This also raises the issue of insufficient dynamic adaptation, ignoring dynamic changes in the patient's health status, such as the dynamic changes brought about by environmental factors. This is because, in addition to physiological factors, the external environment (such as work environment, climate conditions, etc.) also greatly affects eye behavior.

[0010] To address the shortcomings of existing technologies, this invention provides an intervention system for dry eye syndrome related to screen viewing syndrome. The system includes: a personal computer with a front-facing camera; and a humidifier that communicates with the personal computer.

[0011] The personal computer uses its front-facing camera to capture changes in the user's dynamic eye state. The computer's analysis unit determines the user's total blink count based on these changes. This analysis unit then uses the incomplete blink count and blink frequency within the total blink count over a predetermined time period to determine a composite index representing poor eye-use behavior associated with dry eye syndrome.

[0012] The analysis unit determines intervention alert information based on the change of the composite index relative to the baseline data, generates an intervention plan to drive the humidifier based on the intervention alert information, and sends the intervention plan to the humidifier.

[0013] The humidifier performs humidification operations to intervene in poor eye habits according to the intervention plan.

[0014] According to a preferred embodiment, the baseline data of the intervention system is a standardized reference value determined based on the eye dynamics of users with the same application scenario.

[0015] According to a preferred embodiment, the baseline data of the intervention system is a user-customized standardized reference value generated by combining the user's symptom scores after a professional examination at a hospital.

[0016] According to a preferred embodiment, the composite index of the intervention system can dynamically change through different application scenarios displayed on the user's personal computer screen, wherein the different application scenarios include scenarios where the screen displays rapidly changing content, scenarios where the screen displays static content, and scenarios where the screen displays slowly changing content.

[0017] Preferably, a scenario where the screen displays rapidly changing content refers to an image or video content on the screen that has a high frame rate, rapid changes in visual elements, and frequent scene switching.

[0018] Preferably, scenarios where static content is displayed on the screen refer to scenarios where the images, text, or other visual elements displayed on the screen remain unchanged or change only slightly over a period of time. Examples of static content include document reading, code editing, spreadsheet operations, and static webpage browsing.

[0019] Preferably, scenarios where the screen displays slowly changing content refer to images or videos on the screen that change at a slow rate, typically between 24 FPS and 30 FPS. Users do not need to frequently adjust their gaze or focus, and the frequency of visual element changes is low. Common examples of slowly changing content include video conferencing, movie watching, slideshow presentations, and online courses.

[0020] According to a preferred embodiment, the analysis unit is configured to adjust the humidification level of the humidifier and / or change the spray pattern of the humidifier according to different application scenarios displayed on the user's personal computer screen.

[0021] According to a preferred embodiment, the analysis unit is configured to: in scenarios where the screen displays rapidly changing content,

[0022] When the user's composite index is lower than the threshold of the baseline data, the humidifier is set to a low humidification level.

[0023] When the user's composite indicators are within the normal range of the baseline data, the humidifier should be set to the medium humidification level.

[0024] When a user's composite index exceeds the baseline threshold, the humidifier is set to a high humidification level, and the humidification rate is adjusted according to the user's breathing rhythm.

[0025] Among them, the humidification frequency of the high humidification setting is higher than that of the medium humidification setting, and the humidification frequency of the medium humidification setting is higher than that of the low humidification setting.

[0026] According to a preferred embodiment, the analysis unit is configured to: in a scenario where static content is displayed on the screen,

[0027] When the user's composite index falls below the threshold of the baseline data, the humidifier is turned off.

[0028] When the user's composite indicators are within the normal range of the baseline data, the humidifier should be set to a low humidification level.

[0029] When the user's composite index is higher than the threshold of the baseline data, the humidifier is set to the medium humidification level.

[0030] According to a preferred embodiment, the analysis unit is configured to: in a scenario where static content is displayed on the screen,

[0031] When the user's composite index falls below the threshold of the baseline data, the humidifier is turned off.

[0032] When the user's composite indicators are within the normal range of the baseline data, the humidifier should be set to a low humidification level.

[0033] When the user's composite index is higher than the threshold of the baseline data, the humidifier is set to the medium humidification level, and the spray pattern is set to a wave-shaped spray mode.

[0034] According to a preferred embodiment, the analysis unit is configured to: in scenarios where the screen displays slowly changing content,

[0035] When the user's composite index is lower than the threshold of the baseline data, the humidifier setting is adjusted to medium humidification.

[0036] When the user's composite indicators are within the normal range of the baseline data, the humidifier should be set to a low humidification level.

[0037] When the user's composite index is higher than the threshold of the baseline data, the humidifier is adjusted to a high humidification level, and the spray pattern is controlled to be either circular or cloud-shaped.

[0038] According to a preferred embodiment, the personal computer determines that the user's current application scenario is being performed using the collaborative work of its integrated display driver unit and processor.

[0039] Technical Effects of the Invention: This invention provides an intervention system for dry eye syndrome associated with Video Visualization Syndrome (VDT Syndrome). Instead of directly diagnosing dry eye syndrome based on blink frequency or incomplete blink frequency, this system monitors the user's blinking behavior and dynamically intervenes in their eye habits to prevent and alleviate dry eye symptoms caused by VDT Syndrome. The intervention system provides by collecting patients' eye behavior data (such as blink frequency and the proportion of incomplete blinks) and combining this data with the patient's work scenario to establish personalized baseline data. This baseline data is dynamically adjusted based on changes in the corresponding patient's daily eye behavior, and the adjustment is determined by the concentration of trends in composite indicators. This personalized baseline data adjustment mechanism ensures the accuracy and reliability of the intervention system's assessment results, allowing patients to receive truly important intervention information at the appropriate time. This design not only improves user comfort but also avoids psychological stress caused by excessive reminders, helping users better maintain good eye habits. Attached Figure Description

[0040] Figure 1 This is a block diagram of the intervention system provided by the present invention;

[0041] Figure 2 This is a flowchart of the intervention system provided by the present invention;

[0042] Figure 3 This invention provides an intervention scheme generated by an intervention system for scenarios where the screen displays rapidly changing content.

[0043] Figure 4 This invention provides an intervention scheme generated by an intervention system for scenarios where static content is displayed on a screen.

[0044] Figure 5 This invention provides an intervention scheme generated by an intervention system for slowly changing scenes displayed on a screen.

[0045] Figure 6 This is an application scenario diagram of the intervention system provided by the present invention;

[0046] Figure 7 It is a graph showing the trend of composite indicators over time for scenarios where the screen displays rapidly changing content.

[0047] Figure 8 It is a graph showing the trend of composite indicators over time for scenarios where static content is displayed on the screen.

[0048] Figure 9 It is a graph showing the trend of composite indicators over time for scenarios where the screen display changes slowly.

[0049] Figure 10 This is a diagram illustrating how a personal computer's front-facing camera captures blinking behavior.

[0050] Figure 11 This is a scene depicting a patient undergoing a professional eye examination at a hospital to obtain an eye score.

[0051] List of reference numerals

[0052] 100: Personal computer; 110: Front-facing camera; 120: Screen; 130: Analysis unit; 140: Display driver unit; 200: Humidifier. Detailed Implementation

[0053] The following is a detailed explanation with reference to the accompanying drawings.

[0054] In this invention, the intervention system is described using the monitoring behavior of users during continuous work as an example.

[0055] The personal computer 100 in this invention is an independent computer system with computing power, data processing capabilities, and multitasking functions. The front-facing camera 110 integrated into the personal computer 100 can capture the user's eye dynamics, used to monitor the user's blinking behavior, eye health, and other biometric information. For example, the personal computer 100 can be a desktop PC, laptop computer, smartphone, or other similar device.

[0056] In this invention, unhealthy eye habits associated with dry eye syndrome refer to behaviors that may lead to or aggravate tear film instability, ocular surface dryness, abnormal blinking frequency, and other phenomena. These behaviors typically interfere with normal tear secretion and distribution, resulting in insufficient lubrication of the cornea and conjunctiva, thereby triggering or exacerbating dry eye symptoms. Unhealthy eye habits associated with dry eye syndrome include prolonged screen time (prolonged use of electronic devices such as computers, mobile phones, and tablets).

[0057] Example 1

[0058] This embodiment provides an intervention system specifically for dry eye syndrome related to screen viewing syndrome, such as... Figures 1-6 As shown. The intervention system in this embodiment is completely different from the passive diagnostic systems in the prior art that assess and diagnose dry eye syndrome based on various parameters. Existing systems typically rely on several physiological indicators to determine whether someone has dry eye syndrome, ignoring individual differences and environmental factors, resulting in inaccurate assessments of the user's eye health. They also neglect interventions for poor eye habits, as many cases of dry eye syndrome are closely related to user behavior. The intervention system provided in this embodiment forms an active intervention strategy through dynamic monitoring and real-time feedback, thereby effectively alleviating symptoms related to dry eye syndrome associated with screen-related disorders.

[0059] The intervention system in this embodiment includes: a personal computer 100 with a front-facing camera 110; and a humidifier 200 that communicates with the personal computer 100, such as... Figure 6 As shown. Communication between the humidifier 200 and the personal computer 100 is achieved, for example, by connecting the humidifier 200 to the USB port of the personal computer 100 via a USB cable. The USB interface not only provides power but can also be used for data transfer. Preferably, the humidifier 200 has a built-in Wi-Fi module, enabling communication with the personal computer 100 via a home or office wireless network. The personal computer 100 can send control commands to the humidifier 200 via a local area network (LAN) or the Internet. Preferably, the personal computer 100 and the humidifier 200 can also be paired via Bluetooth to achieve wireless communication over short distances. In this embodiment, the communication connection methods established between the personal computer 100 with the front-facing camera 110 and the humidifier 200 are not listed exhaustively.

[0060] According to this embodiment, the personal computer 100 uses its front-facing camera 110 to collect changes in the dynamic state of the user's eyes, and the analysis unit 130 of the personal computer 100 determines the total number of blinks of the user based on these changes. The analysis unit 130 determines a composite index representing poor eye habits associated with dry eye syndrome based on the number of incomplete blinks and blink frequency within the total number of blinks over a predetermined time period. The analysis unit 130 determines intervention reminder information based on the change of this composite index relative to baseline data, generates an intervention plan for driving the humidifier 200 based on the intervention reminder information, and sends the intervention plan to the humidifier 200, wherein the humidifier 200 performs a humidification operation to intervene in poor eye habits according to the intervention plan.

[0061] Preferably, the front-facing camera 110 of the personal computer 100 collects the user's eye dynamic status information at specific intervals. For example, the front-facing camera 110 collects the user's eye dynamic status information every 20 minutes, 30 minutes, and 40 minutes.

[0062] According to this embodiment, the dynamic state information of the eye includes the total number of blinks, the number of incomplete blinks, etc.

[0063] The specific steps for a personal computer to process information are as follows:

[0064] S1: Data Acquisition

[0065] While the user is using the computer, the front-facing camera 110 activates and begins capturing images of the user's face and eyes, such as... Figure 10 As shown, the camera continuously captures a video stream at a frequency of 30 frames per second and transmits it to the analysis unit 130 (in this embodiment, the processor of the personal computer 100).

[0066] S2: Video Preprocessing

[0067] The acquired video frames undergo preprocessing, including noise reduction, contrast enhancement, and edge detection, to ensure image quality meets analytical requirements, especially in low-light or complex background environments. Computer vision algorithms (such as Haar cascade classifiers or deep learning models) are used to detect and locate the user's binocular regions. This ensures accurate capture of dynamic changes in the eyes during each analysis. Blinking behavior is identified by analyzing the motion characteristics of the eye regions. Specifically, the blinking behavior identification method includes detecting changes in the position of the upper and lower eyelids, calculating the blink amplitude, and calculating the blink duration. The blinking behavior identification method is a mature existing technology and will not be elaborated upon in this embodiment.

[0068] S3: Data Analysis

[0069] Blink count statistics: Within a predetermined time period (e.g., every minute or every 10 minutes), the total number of blinks by the user is counted. This includes both complete and incomplete blinks.

[0070] Blink frequency calculation: Based on the total number of blinks and the time period, the blink frequency (blinks / minute) is calculated. Changes in blink frequency can help determine whether a user is overusing their eyes or exhibiting abnormal blinking behavior.

[0071] Incomplete blinking ratio calculation: The proportion of incomplete blinking to the total number of blinks is calculated. A high proportion of incomplete blinking indicates that the user's eyes are under stress and the tear film stability is poor.

[0072] A composite index is generated based on blink frequency and incomplete blink ratio to represent a composite index of poor eye use behaviors associated with dry eye syndrome.

[0073] S4: Feedback and Intervention

[0074] When the composite index exceeds the baseline data, the analysis unit 130 of the personal computer 100 determines the intervention reminder information based on the change of the composite index relative to the baseline data, generates an intervention plan for driving the humidifier 200 based on the intervention reminder information, and sends the intervention plan to the humidifier 200 connected to the personal computer 100 to control the humidifier 200 to perform humidification operation for intervening in poor eye behavior according to the intervention plan.

[0075] In this embodiment, the baseline data refers to standard reference values ​​used to assess a user's blinking behavior and eye health. The system determines whether an intervention alert needs to be issued by comparing the changes in the user's real-time composite indicators with the baseline data, and generates a corresponding intervention plan.

[0076] Preferably, the baseline data in this embodiment is a standardized reference value determined through big data analysis and machine learning models based on the eye dynamics of a large number of users with the same application scenarios. This model is applicable to a wide range of users, providing a universal benchmark to help the system identify common poor eye-use behaviors. Specifically, when a user uses a personal computer 100, the front-facing camera 110 collects their eye dynamic data (such as blink count, incomplete blink count, blink frequency, incomplete blink ratio, etc.) and periodically uploads this data anonymously to a cloud server. After receiving the uploaded data from multiple users, the cloud server processes the data using big data analysis models (such as random forest, support vector machine, neural network, and other machine learning models). Through cluster analysis, regression analysis, and other methods, typical eye dynamic characteristics under different application scenarios are identified, and corresponding baseline data is generated for each scenario.

[0077] According to this embodiment, the intervention reminder information is a signal generated by the analysis unit 130 (processor) of the personal computer 100 in the intervention system based on the user's dynamic eye state change information, used to control the humidifier 200 to perform humidification operation. The core purpose of these reminder messages is to intervene in the user's poor eye-use behaviors related to dry eye syndrome by adjusting the humidification rhythm and spray pattern of the humidifier 200, so as to alleviate the adverse symptoms.

[0078] According to this embodiment, the personal computer 100 determines the user's current application scenario through the collaborative work of the display driving unit 140, such as the graphics processing unit (GPU), and the CPU. In this embodiment, the GPU is not only responsible for processing the displayed content (such as videos, games, web pages, etc.), but also works with the CPU to help determine the current usage scenario. By monitoring the GPU's working status (such as load, frame rate, rendering mode, etc.), the CPU can infer the type of content the user is currently viewing (such as office documents, videos, games, etc.) and load corresponding calculation schemes for blink frequency and incomplete blink count based on different content types.

[0079] Specifically, the steps for identifying the application scenarios of the personal computer 100 and the loading method of the computing scheme are as follows:

[0080] S1: GPU operating status monitoring

[0081] The CPU monitors the GPU's operating status in real time through a communication interface (such as the PCIe bus), including load, frame rate, and rendering mode. Based on this information, the CPU can infer the type of content the user is currently viewing.

[0082] S2: Loading the calculation scheme

[0083] Based on the identified content type, the CPU loads the corresponding calculation scheme for blink frequency and incomplete blink count, with each content type corresponding to different baseline data and intervention thresholds.

[0084] S3: Blink Behavior Analysis

[0085] Video acquisition and preprocessing, eye region localization, blink detection and statistics, and blink frequency and ratio changes. This step is existing technology and will not be described in detail in this embodiment.

[0086] S4: Generation of Composite Indicators

[0087] Baseline data comparison: The analysis unit 130 compares the user's blink frequency, incomplete blink ratio, and other data with the corresponding baseline data to generate a composite index representing poor eye use behavior related to dry eye syndrome.

[0088] Scenario-adaptive adjustments: Based on different content types, analysis unit 130 will dynamically adjust the thresholds of composite indicators.

[0089] This embodiment generates test values ​​for a composite index by simulating the eye use of a patient working for 8 hours. Since the content viewed by the subject varies—for example, screen 120 displays rapidly changing content, static content, and slowly changing content—the same patient can be tested for three consecutive days. For instance, on the first day, the patient can view rapidly changing content to simulate a normal 8-hour workday. Within each hour, the composite index test value is calculated at 10-minute intervals. This testing method is conducted based on the patient's normal working hours, thus obtaining data that more closely reflects reality.

[0090] S5: Feedback and Intervention

[0091] When the composite index exceeds the baseline data, the analysis unit 130 of the personal computer 100 determines the intervention reminder information based on the change of the composite index relative to the baseline data, generates an intervention plan for driving the humidifier 200 based on the intervention reminder information, and sends the intervention plan to the humidifier 200 connected to the personal computer 100 to control the humidifier 200 to perform humidification operation for intervening in poor eye behavior according to the intervention plan.

[0092] According to this embodiment, the intervention scheme includes adjusting the humidification level of the humidifier 200 and the humidification mode with a reminder function.

[0093] Preferably, the humidification level adjustment of the humidifier 200 includes a high humidification level, a medium humidification level, and a low humidification level, wherein the humidification frequency of the high humidification level is higher than that of the medium humidification level, and the humidification frequency of the medium humidification level is higher than that of the low humidification level.

[0094] Preferably, the humidification mode of the reminder function includes accelerating the humidification rate, decelerating the humidification rate, changing the spray pattern of the humidifier 200, or turning off the humidifier 200 according to the user's breathing rhythm.

[0095] Example 2

[0096] This embodiment is a further improvement on embodiment 1, and repeated content will not be described again.

[0097] This embodiment provides an intervention system for poor eye-use behaviors related to dry eye syndrome in scenarios where the screen 120 displays rapidly changing content (such as games or sports videos).

[0098] The analysis unit 130 of the personal computer 100 is configured to calculate the corresponding composite index C. The formula for calculating the composite index C is as follows:

[0099] C = w1·F ratio +w2·I change ,

[0100] Among them, F ratio The blink rate ratio represents the change in blink rate relative to the initial blink rate when using a video display terminal (VDT), reflecting the relative change in blink rate over time; I change The change in the proportion of incomplete blinks represents the degree of change in the proportion of incomplete blinks from the start to the current sampling time point or the end time point; that is, it directly measures the change in the proportion of incomplete blinks. w1 and w2 are weighting coefficients used to adjust F. ratio and I change The importance of this is that w1 + w2 = 1.

[0101] Here, Among them, F initial F is the initial instantaneous frequency. final The current or final blink frequency after a specific time interval (e.g., 30 minutes).

[0102] Here, I change =I final -I initial , among which, I initial For the initial incomplete blink ratio, I final The proportion of the current or final incomplete blinks after a specific time interval (e.g., 30 minutes).

[0103] According to this embodiment, the weighting coefficients w1 and w2 are determined based on expert knowledge and clinical experience or historical data.

[0104] In this scenario, due to heightened concentration, blinking frequency decreases, while the proportion of incomplete blinks increases significantly. The composite index calculation scheme in this embodiment effectively captures the changes in these two variables, particularly through the blink frequency ratio F. ratio To reflect the change in blink frequency relative to the initial state and the change in incomplete blink frequency I chage This method quantifies the increase in the proportion of incomplete blinking. Therefore, it can effectively identify this type of poor eye behavior and provide a comprehensive assessment indicator.

[0105] The experimental data used in this embodiment to determine the risk threshold of the composite index was obtained from a single subject. Specifically, in this embodiment, the same subject was tested continuously for 8 hours (watching content that changes rapidly, such as sports videos) to obtain a trend chart of the composite index over time, as shown in the figure below. Figure 7 As shown.

[0106] Figure 7In the graph, the horizontal axis represents time, and the vertical axis represents the test value of the composite index. Each curve represents the test value of the composite index within one hour. However, the slope of the curves in this graph is relatively dispersed, and the test value of the composite index at the same time point fluctuates greatly. When the system makes a judgment (comparing it with the risk threshold of the composite index), due to the dispersed trend, the composite index is judged to exceed the risk threshold. Therefore, when the system controls the intervention plan of humidifier 200, it cannot intervene more accurately or appropriately for the patient. Instead, it may affect the patient due to frequent false alarms.

[0107] Example 3

[0108] This embodiment is a further improvement of embodiment 1, and the repeated content will not be described again.

[0109] This embodiment provides an intervention system for poor eye-use behaviors related to dry eye syndrome in scenarios where static content is displayed on screen 120.

[0110] The analysis unit 130 of the personal computer 100 is configured to calculate the corresponding composite index C. The formula for calculating the composite index C is as follows:

[0111] C = w1·F + w2·(1-I),

[0112] Where w1 and w2 are weighting coefficients, satisfying w1+w2=1; F is the blink / blink frequency (times / minute); I is the proportion of incomplete blinks to the total number of blinks (incomplete blink ratio, expressed as a percentage), ranging from 0 to 1; 1-I represents the proportion of complete blinks. When I is close to 0, it indicates that most blinks are complete; when I is close to 1, it indicates that most blinks are incomplete.

[0113] According to this embodiment, the weighting coefficients w1 and w2 are determined based on expert knowledge and clinical experience or historical data.

[0114] For static content, users' blinking behavior is more stable, and the proportion of incomplete blinking changes less. In this case, the composite index calculation scheme provided in this embodiment considers both blinking frequency F and the proportion of complete blinking 1-I, and the importance of the two can be balanced by adjusting the weights w1 and w2, for example, w1 = 0.6, w2 = 0.4. That is, this scheme can sensitively capture small changes in blinking frequency, and can reflect potential poor eye behavior even if the proportion of incomplete blinking remains relatively constant.

[0115] The experimental data used in this embodiment to determine the risk threshold of the composite indicator was obtained from a single subject. Specifically, in this embodiment, the same subject was tested continuously for 8 hours (viewing static content, such as Word text content) to obtain a trend graph of the composite indicator over time, as shown below. Figure 8 As shown.

[0116] Figure 8 In the graph, the horizontal axis represents time, and the vertical axis represents the test value of the composite index. Each curve represents the test value of the composite index within one hour. Although the curves in the graph show relatively consistent trends, the test values ​​of the composite index fluctuate significantly at the same time point. When the system makes a judgment (comparing with the risk threshold of the composite index), the relatively dispersed changes in the test values ​​of the composite index reduce the uncertainty of the judgment, and the setting of the threshold may also be biased. Therefore, when the system controls the intervention plan of the humidifier 200, it is more likely to have frequent false alarms or require adjustments to the plan, which may affect the patient's work.

[0117] Example 4

[0118] The baseline data in this embodiment is a user-customized standardized reference value generated by combining a unified model with the user's symptom scores after a professional examination at a hospital. This method of determining baseline data can more accurately reflect individual differences among users and provide more personalized intervention plans. Specifically, users can go to a hospital for a professional ophthalmological examination, such as... Figure 11 As shown, doctors will give a dry eye score based on the user's actual situation. This score is uploaded to a personal computer or cloud server as an important part of personalized baseline data.

[0119] This embodiment is a further improvement of embodiment 1, and the repeated content will not be described again.

[0120] This embodiment provides an intervention system for poor eye-use behaviors related to dry eye syndrome in scenarios where screen 120 displays slowly changing content.

[0121] The analysis unit 130 of the personal computer 100 is configured to calculate the corresponding composite index C. The formula for calculating the composite index C is as follows:

[0122] C = F × (1 - I) and / or

[0123] C = w1·F reduced +w2·I score ,

[0124] For C = F × (1-I), where F is the number of complete blinks per minute, i.e., blink / blink frequency; I is the proportion of incomplete blinks to the total number of blinks (incomplete blink ratio), ranging from 0 to 1; 1-I represents the proportion of complete blinks.

[0125] Regarding C = w1·F reduced +w2·I score Where w1 and w2 are weighting coefficients used to adjust F ratio and Ichange The importance of F is that it satisfies w1 + w2 = 1; reduced The degree to which blink frequency is reduced relative to normal conditions; I score This is a weighted average of the proportion of incomplete blinking and the symptom score.

[0126] According to this embodiment,

[0127] Among them, F VDT The blink rate during the use of the video display terminal; F normal This represents the blinking frequency under normal conditions.

[0128] According to this embodiment, I score =I×S,

[0129] Where I represents the proportion of incomplete blinking, and S represents the symptom score of the user during a professional examination at the hospital.

[0130] According to this embodiment, C = F × (1-I) captures the combined effect between blink frequency and the quality of complete blinks. Specifically, when F is large and I is small, that is, when the blink frequency is high and most blinks are complete, the value of C will be high, which represents a good ocular surface moisture state. Conversely, if F is small or I is large, even if the blink frequency is not high or many blinks are incomplete, the value of C will be low, which indicates insufficient ocular surface moisture and a higher risk of increased dry eye.

[0131] When blinking behavior exhibits significant uncertainty or randomness, intervention systems require a composite index that captures both blinking frequency and the impact of incomplete blinking. The calculation scheme C = F × (1-I) multiplies the blinking frequency F by the proportion of complete blinks 1-I, forming a single product term. This method may be more advantageous when dealing with highly volatile data because it emphasizes the simultaneous effect of both factors. On the other hand, to retain more flexibility and allow for independent assessment of the degree of blinking frequency reduction, this embodiment also utilizes C = w1·F reduced +w2·I score The calculation scheme takes into account the reduction of the instantaneous frequency F. reduced and incomplete blinking score I score And adjust it using weights w1 and w2.

[0132] The trial data used in this embodiment to determine the risk threshold of the composite indicator was obtained from a single patient. Specifically, in this embodiment, the same subject was tested continuously for 8 hours (with slowly changing parameters) to obtain a trend graph of the composite indicator over time, as shown below. Figure 9 As shown. Figure 9The graph shows the changing trends of the first composite index (C1) and the second composite index (C2) over time. The first hour to the eighth hour in the graph represent eight trials over eight hours.

[0133] Because each individual's blinking behavior, eye dynamics, and reactions to viewed content are unique, this embodiment controls for individual differences by conducting tests on the same patient for three consecutive days, avoiding interference from differences in physiological characteristics, eye habits, and environmental adaptability. Testing the same patient within the same time period ensures more stable behavioral patterns (work intensity, rest frequency, etc.), reducing result bias caused by behavioral differences between patients. Furthermore, by using data from the same patient, baseline data suitable for that patient can be determined based on different scenarios and individual eye conditions. This baseline data can more accurately assess the impact of different screen content on blinking behavior and composite indicators, preventing excessively large baseline differences between patients that could lead to unsuitable intervention plans.

[0134] Figure 9 The data shows that the test values ​​of the first composite index at the same time point are more dispersed, while the test values ​​of the second composite index at the same time point fluctuate less (are more convergent). This indicates that the trend of change of the second composite index is more concentrated, meaning that the test values ​​of the second composite index can better reflect the overall trend of change in the patient's ocular condition, with fewer individual abnormalities. The setting of baseline data (risk threshold) needs to be based on a more consistent overall data distribution and trend. In this embodiment, the second composite index C2 is determined by collecting the changing trends of the first and second composite indices and can be used to adjust the parameters of C1. The concentration of the slope of the second composite index means that the changing trend of the data is more consistent, thereby improving prediction accuracy. This means that the value of the patient's second composite index can be predicted more accurately at future time points, thus better assessing the patient's ocular condition. Furthermore, since the slope and threshold of the second composite index can more accurately reflect and predict the patient's ocular condition, the parameters of the first composite index can be adjusted using the second composite index. In addition, the baseline data or risk threshold determined based on the second composite index can avoid frequent false alarms from the system that could interfere with the patient.

[0135] According to this embodiment, the slope and threshold of a trend graph with a more consistent or convergent trend are used as two indicators to judge the patient's eye use status, which effectively improves the system's judgment and prediction of users' poor eye use behavior related to dry eye syndrome, thereby generating an eye intervention plan that is more suitable for the corresponding patient.

[0136] Example 5

[0137] This embodiment is a further improvement on embodiment 4, and the repeated content will not be described again.

[0138] This embodiment provides yet another intervention system for poor eye-use behaviors related to dry eye syndrome based on a scenario where the screen 120 displays slowly changing content.

[0139] Screen 120 displays slowly changing content, and the user's blinking frequency and incomplete blinking ratio change randomly. This means that further analysis is needed to determine if there are any unhealthy eye behaviors associated with dry eye syndrome. To improve the accuracy of the prediction, this invention first uses the calculation scheme C = F × (1-I) for preliminary evaluation, and then uses C = w1·F reduced +w2·I score The calculation scheme is used for verification. This method combines the advantages of two different strategies, providing more comprehensive and detailed results.

[0140] This embodiment first uses the calculation scheme C = F × (1-I) for preliminary evaluation to screen out potentially risky situations; then it uses C = w1·F reduced +w2·I score The calculation scheme is used for review, and the introduction of physician diagnostic symptom scores means that medical expertise can be used to supplement objective measurement data. At this point, in addition to continuing to monitor blinking behavior, professional symptom scores based on ophthalmologist diagnoses, clinical experience, and patient self-reports can be used to confirm whether humidification intervention is necessary. Therefore, combining the two calculation schemes can yield more accurate risk assessment conclusions, allowing for a scientific determination of whether humidification intervention is required.

[0141] This combined approach not only fully leverages the advantages of automated monitoring but also ensures that critical decisions are supported by professional diagnosis, thereby improving the reliability and effectiveness of the entire intervention system. Furthermore, this method provides valuable experience for subsequent dry eye research on current users, helping to continuously optimize personalized assessment algorithms to better suit the actual needs of each user.

[0142] It should be noted that the specific embodiments described above are exemplary. Those skilled in the art can devise various solutions inspired by the disclosure of this invention, and these solutions all fall within the scope of this invention and its protection. Those skilled in the art should understand that this specification and its accompanying drawings are illustrative and not intended to limit the scope of the claims. The scope of protection of this invention is defined by the claims and their equivalents. This specification contains multiple inventive concepts; phrases such as "preferredly" or "according to a preferred embodiment" indicate that the corresponding paragraph discloses an independent concept. The applicant reserves the right to file divisional applications based on each inventive concept.

Claims

1. An intervention system for screen-related dry eye syndrome, the intervention system comprising: A personal computer (100) with a front-facing camera (110); and a humidifier (200) that communicates with the personal computer (100). The feature is that the personal computer (100) uses its front-facing camera (110) to collect the dynamic changes in the current user's eye state, and the analysis unit (130) of the personal computer (100) determines the total number of blinks of the current user based on the dynamic changes in the eye state. The analysis unit (130) determines a composite index representing poor eye-use behavior related to dry eye syndrome based on the number of incomplete blinks and blink frequency in the total number of blinks within a predetermined time period. The analysis unit (130) determines intervention reminder information based on the changes in the composite index relative to baseline data, generates an intervention plan for driving the humidifier (200) based on the intervention reminder information, and sends the intervention plan to the humidifier (200). The determined baseline data is dynamically adjusted based on changes in the corresponding patient's daily eye use behavior, and the adjustment of the baseline data is determined based on the concentration of the changing trends of the composite index. The humidifier (200) performs a humidification operation to intervene in poor eye behavior according to the intervention plan. The composite indicators of the intervention system can dynamically change through different application scenarios displayed on the screen (120) of the user's personal computer (100). The different application scenarios include scenarios where the screen (120) displays content that changes rapidly, scenarios where the screen (120) displays static content, and scenarios where the screen (120) displays content that changes slowly. The analysis unit (130) is configured to adjust the humidification level of the humidifier (200) and / or change the spray pattern of the humidifier (200) according to different application scenarios displayed on the screen (120) of the user's personal computer (100).

2. The intervention system for screen-related dry eye syndrome according to claim 1, characterized in that, The baseline data of the intervention system is a standardized reference value determined based on the eye dynamics of users with the same application scenario.

3. The intervention system for screen-related dry eye syndrome according to claim 1, characterized in that, The baseline data of the intervention system is a user-customized standardized reference value generated by combining the user's symptom scores after professional examinations at the hospital.

4. The intervention system for screen-related dry eye syndrome according to claim 1, characterized in that, The analysis unit (130) is configured to: in scenarios where the screen (120) displays rapidly changing content, When the user's composite index is lower than the threshold of the baseline data, the humidifier (200) is set to a low humidification level. When the user's composite index is within the normal range of the baseline data, the humidifier (200) is set to the medium humidification level. When the user's composite index exceeds the threshold of the baseline data, the humidifier (200) is set to a high humidification level, and the humidification rate of the humidifier (200) is adjusted according to the user's breathing rhythm. Among them, the humidification frequency of the high humidification setting is higher than that of the medium humidification setting, and the humidification frequency of the medium humidification setting is higher than that of the low humidification setting.

5. The intervention system for screen-related dry eye syndrome according to claim 1, characterized in that, The analysis unit (130) is configured to: in a scenario where static content is displayed on the screen (120), When the user's composite index is lower than the threshold of the baseline data, the humidifier is turned off (200). When the user's composite index is within the normal range of the baseline data, the humidifier (200) is set to the low humidification level. When the user's composite index is higher than the threshold of the baseline data, the humidifier (200) is set to the medium humidification level.

6. The intervention system for screen-related dry eye syndrome according to claim 1, characterized in that, The analysis unit (130) is configured to: in a scenario where static content is displayed on the screen (120), When the user's composite index is lower than the threshold of the baseline data, the humidifier is turned off (200). When the user's composite index is within the normal range of the baseline data, the humidifier (200) is set to the low humidification level. When the user's composite index is higher than the threshold of the baseline data, the humidifier (200) is set to the medium humidification level, and the spray pattern of the humidifier (200) is set to the wave spray mode.

7. The intervention system for screen-related dry eye syndrome according to claim 1, characterized in that, The analysis unit (130) is configured to: in a scenario where the screen (120) displays slowly changing content, When the user's composite index is lower than the threshold of the baseline data, the humidifier (200) is adjusted to the medium humidification level. When the user's composite index is within the normal range of the baseline data, the humidifier (200) is set to the low humidification level. When the user's composite index is higher than the threshold of the baseline data, the humidifier (200) is adjusted to a high humidification level, and the spray pattern of the humidifier (200) is adjusted to a circular or cloud-shaped spray mode.

8. The intervention system for screen-related dry eye syndrome according to claim 1, characterized in that, The personal computer (100) determines the user's current application scenario by utilizing the collaborative work of its integrated display driver unit (140) and processor.

Images