Monitoring methods, devices, media and products for preventing stroke

Abstract

The disclosed embodiments provide a monitoring method, device, medium, and product for preventing stroke, including: obtaining sleep time data of a user entering a current sleep state, and collecting real-time physiological index data related to stroke from the user during sleep; predicting a first stroke risk value of the user at a predicted moment in the future based on the prediction information through a prediction model; obtaining a second stroke risk value of the user based on the real-time physiological index data through a real-time monitoring model; determining a stroke warning event and triggering an intervention action based on an indication of the consistency of the change trends of the first stroke risk value and the second stroke risk value with respect to a stroke warning event. By performing a consistency analysis on the change trends of the first and second stroke risk values, abnormal conditions that may indicate a stroke can be identified at an early stage, providing a valuable time window for timely intervention measures.

Description

Technical Field

[0001] The present disclosure relates to the field of medical health monitoring technology, and in particular to monitoring methods, devices, media, and products for preventing stroke. Background Art

[0002] The sudden onset and rapid progression of stroke pose a significant challenge to early warning technologies. Commonly used clinical stroke risk assessment methods rely on regular physical examination indicators (such as blood pressure and lipid levels) and imaging studies. These methods have significant lags and are unable to detect physiological abnormalities in the hours before onset. Stroke risk is particularly elevated during sleep, when changes in the body's autonomic nervous system regulate the body's function. Summary of the Invention

[0003] In view of the above-mentioned shortcomings of the prior art, the purpose of the present disclosure is to provide a monitoring method, device, medium and product for preventing stroke, so as to solve the problems in the related art.

[0004] A first aspect of the present disclosure provides a monitoring method for preventing stroke, wherein the monitoring method comprises:

[0005] Acquiring sleep time data of the user entering the current sleep state, and collecting real-time physiological indicator data related to stroke from the user during sleep; wherein the physiological indicator data related to stroke includes facial data and snoring data collected during the sleep state;

[0006] Predicting a first stroke risk value of a user at a predicted time in the future based on prediction information using a prediction model; wherein the prediction information includes the user's disease profile data and the sleep time data; and

[0007] Obtaining a second stroke risk value of the user according to the real-time physiological indicator data through a real-time monitoring model;

[0008] A stroke warning event is determined and an intervention action is triggered based on an indication of consistency of the change trends of the first stroke risk value and the second stroke risk value with respect to a stroke warning event.

[0009] In an embodiment of the first aspect, further comprising:

[0010] Detecting user's posture changes during sleep;

[0011] In response to determining, according to the posture change data, that the user moves from a lying position to a standing position, performing fall event detection;

[0012] In response to the occurrence of a fall event, a corresponding alarm action is executed.

[0013] In an embodiment of the first aspect, the determining of the stroke warning event according to the indication of consistency of the change trends of the first stroke risk value and the second stroke risk value with respect to the stroke warning event includes one or more of the following:

[0014] 1) When the change trend of the second stroke risk value continues to rise within a preset time range or reaches a value within a preset range from the first stroke risk value, determining a stroke warning event and triggering an intervention action;

[0015] 2) when the second stroke risk value reaches a preset risk threshold at a determination time that is a preset time period from the prediction time, determining a stroke warning event and triggering an intervention action; wherein the preset risk threshold is set according to a preset ratio of the first stroke risk value;

[0016] 3) When the change trend of the second stroke risk value gradually decreases, it is determined that the stroke warning event has not occurred and no intervention action is triggered;

[0017] 4) Perform at least one of the following interventions based on the stroke warning event:

[0018] Send risk warning information to relevant personnel associated with the user;

[0019] The user's location information and physiological indicator data related to stroke are sent to the medical institution associated with the user to prepare for medical treatment.

[0020] In an embodiment of the first aspect, obtaining facial data from the user's stroke-related physiological indicator data includes:

[0021] In a sleeping state, a facial image of the user is acquired through a facial acquisition device, and facial feature points are extracted based on the facial image;

[0022] The user's eye state information and mouth state information are extracted based on facial feature points, and the facial data is obtained based on the eye state information and the mouth state information.

[0023] In an embodiment of the first aspect, the eye state information is defined as a height difference between a left eyelid and a right eyelid of the user; and the mouth state information is defined as a displacement difference between left and right corners of the mouth.

[0024] In an embodiment of the first aspect, the facial acquisition device includes an infrared camera; and / or the facial acquisition device is set to face the user's face.

[0025] In an embodiment of the first aspect, obtaining snoring data from the user's physiological indicator data related to stroke includes:

[0026] Acquiring voice information of a user in a sleeping state through at least one microphone, and extracting snoring audio of the user according to the voice information to obtain snoring characteristics;

[0027] The snoring data is obtained according to the snoring features through a trained snoring recognition model; wherein the snoring data includes pause ventilation information and ambiguous pronunciation information in the user's snoring.

[0028] A second aspect of the present disclosure provides a computer device, comprising:

[0029] processor and memory;

[0030] The memory stores program instructions;

[0031] The processor is configured to run the program instructions to execute any one of the monitoring methods described above.

[0032] A third aspect of the present disclosure provides a computer-readable storage medium, wherein program instructions are stored, and the program instructions are executed to perform any of the monitoring methods described above.

[0033] A fourth aspect of the present disclosure provides a computer program product, which includes: program instructions for executing any of the monitoring methods described above.

[0034] The beneficial effects of the present disclosure are as follows: by performing consistency analysis on the changing trends of the first and second stroke risk values, abnormal conditions that may indicate a stroke can be identified at an early stage, providing a valuable time window for timely intervention measures, and at the same time, more accurately determining whether there are consistent stroke warning events. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 A schematic diagram showing the overall process of a monitoring method for preventing stroke in one embodiment of the present disclosure.

[0036] Figure 2 A schematic diagram of the flow chart corresponding to fall monitoring in a monitoring method for preventing stroke in one embodiment of the present disclosure is shown.

[0037] Figure 3 A schematic diagram showing the process of acquiring facial data in a monitoring method for preventing stroke in one embodiment of the present disclosure.

[0038] Figure 4 A schematic diagram showing the process of obtaining snoring data in a monitoring method for preventing stroke in one embodiment of the present disclosure is shown.

[0039] Figure 5 A schematic diagram showing the structure of a computer device in one embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0040] The following describes the embodiments of the present disclosure through specific examples. Those skilled in the art can easily understand the other advantages and effects of the present disclosure from the information disclosed in this disclosure. The present disclosure can also be implemented or applied through different specific embodiments. The details of the present disclosure can also be modified or changed according to different viewpoints and application modules without departing from the spirit of the present disclosure. It should be noted that the embodiments and features in the embodiments of the present disclosure can be combined with each other unless there is a conflict.

[0041] The following is a detailed description of the embodiments of the present disclosure with reference to the accompanying drawings so that those skilled in the art can easily implement the present disclosure. The present disclosure can be embodied in many different forms and is not limited to the embodiments described herein.

[0042] Throughout the present disclosure, reference to terms such as "one embodiment," "some embodiments," "examples," "specific examples," or "some examples" means that a specific feature, structure, material, or characteristic described in conjunction with that embodiment or example is included in at least one embodiment or example of the present disclosure. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or a group of embodiments or examples. Furthermore, those skilled in the art may combine and integrate different embodiments or examples, and features of different embodiments or examples, as described in the present disclosure, without conflicting requirements.

[0043] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features being referred to. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one such feature. In the context of this disclosure, "a group" means two or more, unless otherwise specifically defined.

[0044] In order to clearly describe the present disclosure, components not related to the description are omitted, and the same or similar components throughout the specification are denoted by the same reference numerals.

[0045] Throughout this specification, when a device is said to be "connected" to another device, this includes not only "direct connection" but also "indirect connection" with other elements interposed therebetween. Furthermore, when a device is said to "include" a certain component, unless otherwise stated, this does not exclude the inclusion of other components but rather implies that the device may include other components.

[0046] Although the terms first, second, etc. are used in this document to represent various elements in some examples, these elements should not be limited by these terms. These terms are only used to distinguish one element from another. For example, the first interface and the second interface, etc. are represented. Furthermore, as used in this document, the singular forms "one", "an", and "the" are intended to also include the plural forms, unless there is a contrary indication in the context. It should be further understood that the terms "comprise" and "include" indicate the presence of features, steps, operations, elements, modules, projects, types, and / or groups, but do not exclude the presence, occurrence, or addition of one or a group of other features, steps, operations, elements, modules, projects, types, and / or groups. The terms "or" and "and / or" used herein are interpreted as inclusive, or mean any one or any combination. Therefore, "A, B, or C" or "A, B, and / or C" means "any of the following: A; B; C; A and B; A and C; B and C; A, B, and C". Exceptions to this definition only occur when the combination of elements, functions, steps, or operations is inherently mutually exclusive in some way.

[0047] The technical terms used herein are intended only to refer to specific embodiments and are not intended to limit the present disclosure. The singular form used herein also includes the plural form unless the statement explicitly indicates otherwise. The term "comprising" as used in this specification is intended to specify specific features, regions, integers, steps, operations, elements, and / or components and does not exclude the presence or addition of other features, regions, integers, steps, operations, elements, and / or components.

[0048] Although not defined differently, all terms used herein, including technical and scientific terms, have the same meanings as those generally understood by those skilled in the art to which this disclosure belongs. Terms defined in commonly used dictionaries are additionally interpreted as having meanings consistent with relevant technical literature and the current message. Unless otherwise defined, they should not be overly interpreted as ideal or highly formalized meanings.

[0049] First, related technologies generally do not address stroke monitoring during sleep. However, many early symptoms (such as facial asymmetry, apnea, and hypopnea) can manifest during sleep. Second, most stroke risk assessment methods rely on static data such as historical medical records, lifestyle factors, and regular physical examination results. These methods often overlook the importance of real-time physiological data, particularly key physiological indicators during sleep.

[0050] In order to solve the above problems, a monitoring method for preventing stroke is provided in one embodiment of the present disclosure, wherein, please refer to Figure 1In one embodiment, a forward-looking risk assessment is provided by combining long-term medical history and current sleep status to help identify high-risk individuals. A real-time monitoring model is used to calculate a second stroke risk value based on real-time facial and snoring data, dynamically reflecting current health status. By simultaneously considering both the first and second risk values, dual verification of potential health issues is achieved. A single data source can be biased by multiple factors, but combining two different types of data can effectively reduce false positives and false negatives. Continuous monitoring of the user's physiological state during sleep can trigger an alert at the first onset of symptoms. Once possible signs of a stroke are detected, the system can automatically notify emergency contacts or healthcare providers, ensuring the user receives necessary medical assistance as quickly as possible, even during the night. The first few hours after a stroke is a critical treatment window, particularly for acute ischemic stroke, where thrombolytic therapy is most effective. Prompt treatment within this timeframe significantly improves the user's survival rate and recovery outcomes.

[0051] like Figure 1 In an embodiment, the monitoring method comprises:

[0052] Step S1: Obtain the sleep time data of the user entering the current sleep state, and collect real-time physiological index data related to stroke from the sleeping user in real time; wherein the physiological index data related to stroke includes facial data and snoring data collected in the sleep state.

[0053] Specifically, in some embodiments, a smart mattress or wearable device (such as a smart bracelet or smart watch) can be used to monitor a user's sleep status. These devices can detect the user's activity level and heart rate changes through built-in sensors (such as accelerometers and heart rate sensors), thereby determining whether the user has entered a sleep state. Sleep time data includes bedtime, which records the time from the user getting into bed to actually falling asleep. Difficulty falling asleep (insomnia) is associated with an increased risk of stroke. People with cardiovascular disease may take longer to fall asleep than healthy people due to possible physical discomfort. Although bedtime itself does not directly cause stroke, as part of overall sleep patterns, it does have a significant impact on stroke risk.

[0054] Furthermore, facial data can reveal facial nerve damage or impaired muscle control, which can lead to facial asymmetry, a common condition among users who have experienced a stroke. Real-time monitoring of facial asymmetry can help identify possible stroke symptoms early. Specific snoring patterns (such as high-frequency, high-amplitude snoring) may indicate upper airway obstruction, a hallmark of obstructive sleep apnea (OSA). OSA is an independent risk factor for stroke.

[0055] In addition to facial and snoring data, real-time physiological indicator data can also include sleep interruptions, blood oxygen saturation, and respiratory rate. Frequent nocturnal awakenings or sleep interruptions (such as those caused by apnea or other sleep disorders) are associated with a higher risk of stroke. Sleep apnea can cause intermittent hypoxia, triggering a series of pathological reactions, including oxidative stress, increased inflammation, and endothelial dysfunction, all of which can damage the cardiovascular system and lead to stroke. Furthermore, pulse oximetry can be used to continuously monitor blood oxygen concentration. A decrease in blood oxygen saturation (e.g., below 90%) may signal hypoxia, which is particularly important in stroke, especially ischemic stroke. Hypoxia can cause brain cell damage or even death, so real-time monitoring of blood oxygen saturation can help early detection and treatment of potential stroke risks. Respiratory rate can also be indirectly inferred using chest or abdominal motion sensors or audio signals (such as snoring). Abnormal respiratory rate (excessively high or low) may indicate respiratory problems, which in some cases may also affect cardiovascular health. For example, rapid, shallow breathing may be a sign of anxiety or acute respiratory distress syndrome, whereas slow, deep breathing may be associated with central nervous system depression.

[0056] Specifically, in Figure 2 In an embodiment, the facial data can be obtained by the following steps:

[0057] Step S111: While the user is asleep, a facial image of the user is captured using a facial capture device, and facial feature points are extracted from the image. In some embodiments, the facial capture device includes an infrared camera or other non-invasive imaging device installed in the bedroom to ensure clear facial images in low-light conditions. When the ambient light intensity falls below a preset value, an infrared light source is automatically activated to provide fill light (this wavelength is invisible to the human eye and prevents pupil constriction from interfering with sleep monitoring). After the user falls asleep, the facial capture device begins capturing facial images of the user periodically or continuously. These images need to cover the entire sleep cycle to capture facial changes during different sleep stages.

[0058] Furthermore, in some embodiments, the captured facial images undergo preprocessing, including but not limited to: adaptive histogram equalization (CLAHE) to enhance contrast for more accurate facial feature recognition. Denoising is also performed to reduce the impact of image noise on subsequent analysis. Facial feature points are extracted from the preprocessed images. These feature points are distributed in key areas such as the eyes, eyebrows, nose, and mouth, providing basic data for subsequent analysis.

[0059] Step S112: extracting the user's eye state information and mouth state information based on facial feature points, and synthesizing the eye state information and mouth state information to obtain the facial data.

[0060] A stroke affecting the pons or midbrain (e.g., basilar artery occlusion) can damage the oculomotor nuclei, leading to ptosis (loss of levator palpebrae superioris control) or oculomotor movement disorders (horizontal gaze palsy). Although the user is unconscious during sleep, nerve damage can cause abnormal relaxation of the orbicularis oculi muscle, manifesting as unilateral incomplete eyelid closure (known as the "rabbit eye sign"). Infarction in the middle cerebral artery territory can damage the cortical brainstem tract, resulting in contralateral lower facial muscle weakness (e.g., ptosis of the corner of the mouth) while sparing the frontalis muscle (due to bilateral innervation). A stroke affecting the facial nerve nuclei in the pons (e.g., basilar artery branch occlusion) can lead to ipsilateral complete facial paralysis. A posterior circulation stroke may damage the respiratory rhythm center in the medulla oblongata, causing abnormal breathing patterns (e.g., long inspiratory breathing), which can manifest as outward signs such as lip tremors. While awake, the user can mask mild neurological deficits through subjective efforts (e.g., forcefully closing the eyes or deliberately controlling facial expression). However, during sleep, the muscles relax completely, preserving pathological signs.

[0061] Specifically, in some embodiments, the eye status information is defined as the height difference between the user's left eyelid and right eyelid. Specifically, the positions of the upper and lower edges of the left and right eyelids are determined based on eye feature points, and the height of each eyelid is calculated. The heights of the left and right eyelids are compared, and the difference between them is calculated to obtain the eye status information. If the difference exceeds a preset threshold, it may indicate facial nerve damage or other neurological problems.

[0062] Optionally, the mouth state information is defined as the displacement difference between the left and right corners of the mouth. Similarly, the specific coordinate positions of the left and right corners of the mouth are determined based on the mouth feature points. The horizontal and vertical displacement differences between the left and right corners of the mouth are calculated. If these differences exceed a preset threshold, this may indicate facial muscle weakness or paralysis.

[0063] Specifically, in Figure 3 In the embodiment, monitoring snoring during sleep not only helps to detect potential stroke risks at an early stage, but also ensures that the user can get necessary medical help in the shortest possible time, thereby greatly improving the survival rate and recovery effect. Further, the snoring data can be obtained by the following steps:

[0064] Step S121: acquiring voice information of a user in a sleeping state through at least one microphone, and extracting the user's snoring audio according to the voice information to obtain snoring features.

[0065] Specifically, a microphone array is installed in the user's bedroom to ensure that clear snoring signals can be captured. The microphone array can help locate the sound source and reduce background noise interference. The collected audio signals are preprocessed, including noise reduction, filtering and other operations to improve the accuracy of subsequent analysis. To ensure data integrity, the recording should cover the entire sleep cycle. Audio signal processing technology is used to separate snoring audio clips from the collected voice information. These snoring audio clips usually appear as periodic high-frequency vibration patterns, which are significantly different from the sounds of normal speech.

[0066] Step S122: obtaining the snoring data according to the snoring features through a trained snoring recognition model; wherein the snoring data includes pause ventilation information and ambiguous pronunciation information in the user's snoring.

[0067] Specifically, the snoring features extracted from step S121 are fed into a trained snoring recognition model for analysis. The model can detect long pauses in snoring, which may be a sign of obstructive sleep apnea. Obstructive sleep apnea is an independent risk factor for stroke. The model can also detect abnormal changes in snoring sound quality, such as slurred speech, which may be caused by partial airway obstruction or other neurological problems. Slurred speech is sometimes associated with facial nerve damage, a potential marker for stroke.

[0068] Optionally, in Figure 1 In an embodiment, step S2: predicting a first stroke risk value of a user for a stroke at a predicted time in the future based on prediction information using a prediction model; wherein the prediction information includes the user's disease profile data and the sleep time data.

[0069] Specifically, a user's medical profile data includes, but is not limited to, past medical history, family medical history, chronic diseases (such as hypertension, diabetes, etc.), past medical records, medication use, etc. Specifically, hypertension is one of the most important controllable risk factors for stroke. Long-term hypertension can lead to arteriosclerosis and vascular damage, increasing the risk of thrombosis. Hypertension can damage blood vessel walls, making them more susceptible to rupture or blockage, leading to ischemic or hemorrhagic stroke. High blood sugar levels can damage vascular endothelial cells, promote the formation of atherosclerosis, and increase blood viscosity, thereby increasing the likelihood of thrombosis. Atrial fibrillation can cause irregular blood flow within the heart, making it easy for blood clots to form. Once these clots break off and enter the brain, they can cause ischemic stroke. Other heart diseases may also increase the risk of thrombosis through similar mechanisms.

[0070] Family medical history may reflect the influence of genetic factors, and certain gene variants may increase an individual's susceptibility to stroke. In addition, family members often share similar lifestyles and environmental factors, which may also be one of the reasons for the increased risk. Smoking damages vascular endothelial cells and promotes the development of atherosclerosis. It also increases blood viscosity, leading to an increased risk of thrombosis. Obesity is often accompanied by problems such as high blood pressure, diabetes, and high cholesterol, which are all risk factors for stroke. In addition, obesity can affect the body's inflammatory response, further damaging blood vessel health. Alcohol can increase blood pressure, damage liver function, and interfere with coagulation mechanisms, all of which increase the risk of stroke. As we age, blood vessels gradually age and their elasticity decreases, making them more susceptible to arteriosclerosis and thrombosis.

[0071] The prediction model is trained by leveraging historical data from other stroke patients. Multi-source data, including medical records and sleep duration, is collected from electronic health records, clinical research databases, and wearable devices. Basic features such as age, gender, hypertension, diabetes, and atrial fibrillation are extracted. Dynamic features (such as sleep duration) and lifestyle characteristics (such as smoking and drinking habits) are combined. An appropriate machine learning or deep learning model, such as a random forest, gradient boosting decision tree, or neural network, is selected. The dataset is divided into training, validation, and test sets for model training. The prediction model is used to predict stroke risk in new patients, promoting early intervention measures and effectively reducing the risk of stroke.

[0072] exist Figure 1 In the embodiment, step S3 is further included: obtaining a second stroke risk value of the user according to the real-time physiological indicator data through a real-time monitoring model.

[0073] Specifically, in some embodiments, data is collected from a variety of sensors, including facial status information (such as eyelid height difference and mouth corner displacement difference), snoring data, heart rate data, blood oxygen saturation, and respiratory rate. Using a pre-trained real-time monitoring model, these real-time physiological indicators are used as input to calculate the user's stroke risk. The real-time monitoring model can promptly capture changes in the user's health status.

[0074] Step S4: determining a stroke warning event and triggering an intervention action based on the indication of consistency of the change trends of the first stroke risk value and the second stroke risk value with respect to the stroke warning event.

[0075] The first stroke risk value reflects the user's long-term health trends and potential risk factors, while the second stroke risk value captures current, real-time physiological changes. The combination of the two can verify each other and reduce the false positives or false negatives that may be caused by a single model.

[0076] Optionally, the first stroke risk value and the second stroke risk value are usually standardized to the same numerical range (such as 0 to 1 or 0% to 100%) for easy comparison. For example, the first risk value may be based on historical data analysis to derive the probability of a user suffering a stroke in the future (such as 0.25, i.e. 25%), while the second risk value is based on real-time monitoring data to calculate the current short-term risk (such as 0.30, i.e. 30%). Whether it is the first risk value or the second risk value, the ultimate goal is to assess the possibility of the user suffering a stroke, so the two are essentially different dimensional descriptions of the same health event. In some embodiments, determining a stroke warning event includes one or more of the following:

[0077] 1) When the change trend of the second stroke risk value keeps increasing within a preset time range or reaches a value within a preset range from the first stroke risk value, a stroke warning event is determined and an intervention action is triggered.

[0078] Specifically, determine an appropriate time period (e.g., 30 minutes, 1 hour, or several hours) for monitoring the changing trend of the second stroke risk value. Monitor the changing trend of the second stroke risk value within the selected time range (e.g., within the past 1 hour). If the risk value shows a continuous upward trend, further analyze its cumulative change and fluctuation range. When the second stroke risk value gradually approaches the first stroke risk value and enters the preset range (e.g., ±0.05), it is considered that there is a higher risk of stroke. If the second stroke risk value continues to rise within the preset time range (e.g., within 1 hour), and the rate of increase exceeds a certain threshold (e.g., an increase of 0.05 per hour), it is also considered a high-risk signal; a stroke warning event is determined and an intervention action is triggered.

[0079] 2) When the second stroke risk value reaches a preset risk threshold at a determination time that is a preset time period from the prediction time, a stroke warning event is determined and an intervention action is triggered.

[0080] Specifically, the preset risk threshold is set based on the first stroke risk value and is typically a proportional value (e.g., 0.8 times the first stroke risk value). If, at the time of determination, the second stroke risk value reaches or exceeds the preset risk threshold, then a higher stroke risk is considered present.

[0081] 3) When the changing trend of the second stroke risk value gradually decreases, it is determined that the stroke warning event has not occurred and no intervention action is triggered.

[0082] Specifically, if the second stroke risk value continues to increase, it indicates that the user's stroke risk is increasing. If the second stroke risk value remains relatively stable, it indicates that the user's stroke risk remains stable. If the second stroke risk value gradually decreases, it indicates that the user's stroke risk is decreasing.

[0083] If the changing trend of the second stroke risk value gradually decreases, it is considered that there is no acute risk of stroke at present, and no warning or intervention actions will be triggered.

[0084] Optionally, one or more of the following interventions are performed based on the stroke warning event:

[0085] Send risk warning information to relevant personnel associated with the user;

[0086] Specifically, in some embodiments, when a stroke warning event is identified, a notification is automatically sent to these contacts to inform them of the current situation. The warning message can include a brief description (e.g., "Your family / friend is at high risk of stroke. Please pay attention to their health"), recommended initial measures, and further assistance information (e.g., emergency phone number).

[0087] Send the user's location information and physiological indicator data related to stroke to the medical institution associated with the user to prepare for medical treatment. Specifically, use GPS or other positioning technologies to accurately obtain the user's current location and incorporate it into the warning information. Send the user's location information and physiological indicator data to a pre-designated medical institution or emergency center. Medical institutions can prepare in advance based on the information received, such as arranging an ambulance, preparing appropriate treatment equipment or drugs, etc. By notifying medical institutions in advance, the time from the discovery of the warning to the start of medical treatment can be significantly shortened, which is crucial for the treatment of acute stroke.

[0088] Alternatively, when getting up at night, the body changes from a lying position to a standing position, which may cause a sudden change in blood pressure. For users with high blood pressure, this change in body position may cause orthostatic hypotension (i.e., a sudden drop in blood pressure when changing from a lying position to a standing position), which may lead to dizziness, vertigo, or even falls, which may cause a stroke. Figure 4 In an embodiment, the monitoring method further comprises:

[0089] Step S5: Detecting the posture change data of the sleeping user.

[0090] Step S6: In response to determining according to the posture change data that the user has changed from a lying position to a standing position, performing fall event detection.

[0091] Step S7: In response to the occurrence of the fall event, executing a corresponding alarm action.

[0092] Specifically, sensors integrated into smart mattresses or wearable devices capture the user's change from lying to standing posture in real time. When a posture change is detected, fall event detection is automatically performed, and the movement amplitude, pattern, and subsequent posture are analyzed to determine whether a fall has occurred. If a fall event is confirmed, the corresponding alarm action is triggered, including sending a risk notification to the emergency contact associated with the user, and automatically calling the emergency center when necessary, while providing the user's location information and related physiological indicator data for timely medical assistance preparation. This method not only covers the early warning of stroke risk, but also enhances the real-time monitoring and response capabilities of potential safety risks (such as falls) during sleep.

[0093] Furthermore, in some embodiments, it usually takes a certain amount of time (e.g., 3-5 seconds or longer) for a healthy user to transition from a lying position to a standing position, and the movements are relatively smooth and continuous. If the user stands up too quickly or the movements are discontinuous due to weakness, dizziness, or other reasons, it may indicate a potential problem. Standing up quickly may cause orthostatic hypotension, which in turn may lead to a temporary loss of consciousness or loss of balance. Based on the user's historical data, a reasonable range of "normal standing time" is set (e.g., 3-10 seconds). If the time from lying to standing is lower than a certain threshold (e.g., <2 seconds), it is considered to be "standing up quickly", which may indicate abnormal behavior, and the fall event detection mode is entered.

[0094] In another embodiment, posture change data can be collected simultaneously by a camera and hand sensors to achieve more comprehensive and accurate posture monitoring. The camera captures full-body or partial images of the user and analyzes the user's posture changes. Wearable sensors are worn on the user's wrist or finger to record the hand's movement trajectory and posture changes in real time. By analyzing hand movement patterns (such as sudden falls, violent shaking, etc.), abnormal behavior can be determined. The camera provides global posture information, while the hand sensors provide local detailed information. The combination of the two can overcome the shortcomings of a single data source. After the camera captures the user's posture changes, the data from the wearable sensors can be combined to perform fall event detection.

[0095] For users at risk of stroke, waking up mid-sleep and being unable to call or move is extremely dangerous, especially for those living alone. To provide additional safety in this situation, an image acquisition device can monitor the user's eye movements and detect eye changes during sleep. Based on this eye change data, the system can detect an eye abnormality event based on changes in eye feature data when the user transitions from sleep to wakefulness. In response to the occurrence of an eye abnormality event, a corresponding alarm action is triggered. For example, if the user experiences eye deviation, such as gazing to one side, for a certain duration (e.g., 5 seconds), an alarm action is triggered. This alarm mechanism automatically sends a notification to a pre-determined emergency contact, such as a family member or caregiver, and can also automatically call emergency services and provide the user's location and health data. This method can identify potential stroke events early, providing immediate response and necessary assistance, making it particularly suitable for users living alone who are unable to actively seek help.

[0096] like Figure 5 FIG. 1 is a schematic diagram showing the structure of a computer device in one embodiment of the present disclosure.

[0097] The computer device 100 may be exemplified as a processing terminal in a cloud platform, such as a server, a desktop computer, a laptop computer, a tablet computer, a smart phone, or other terminals.

[0098] The computer device 100 includes a bus 101, a processor 102, and a memory 103. The processor 102 and the memory 103 can communicate with each other via the bus 101. The memory 103 can store program instructions. The processor 102 executes the program instructions in the memory 103 to implement the steps of the monitoring method in the previous embodiment, such as Figure 1 .

[0099] Bus 101 may be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus. Buses can be classified into address buses, data buses, and control buses. For ease of illustration, although only one thick line is used in the figure, this does not mean that there is only one bus or only one type of bus.

[0100] In some embodiments, processor 102 may be implemented as a central processing unit (CPU), a microprocessor unit (MCU), a system on a chip (SoC), or a field programmable gate array (FPGA). Memory 103 may include volatile memory, such as random access memory (RAM), for temporarily storing data while running programs.

[0101] The memory 103 may also include a non-volatile memory (non-volatile memory) for data storage, such as a read-only memory (ROM), a flash memory, a hard disk drive (HDD), or a solid-state disk (SSD).

[0102] In some embodiments, the computer device 100 may further include a communicator 104. The communicator 104 is used to communicate with the outside world. In a specific example, the communicator 104 may include one or a group of wired and / or wireless communication circuit modules. For example, the communicator 104 may include one or more of a wired network card, a USB module, a serial interface module, etc. The wireless communication protocols followed by the wireless communication module include, for example, near field communication (NFC) technology, infrared (IR) technology, Global System for Mobile communications (GSM), General Packet Radio Service (GPRS), Code Division Multiple Access (CDMA), Wideband Code Division Multiple Access (WCDMA), Time-Division Code Division Multiple Access (TD-SCDMA), Long Term Evolution (LTE), Bluetooth (BT), Global Navigation Satellite System (GNSS), etc. One or more of the following.

[0103] In an embodiment of the present disclosure, a computer-readable storage medium may be provided, storing program instructions, which implement the monitoring method in any of the previous embodiments when executed.

[0104] That is, the method steps in the above embodiments are implemented as software or computer code that can be stored in a recording medium (such as a CD ROM, RAM, floppy disk, hard disk or magneto-optical disk), or are implemented as computer code originally stored in a remote recording medium or a non-transitory machine-readable medium downloaded via a network and to be stored in a local recording medium, so that the method represented herein can be stored in such software processing on a recording medium using a general-purpose computer, a dedicated processor or programmable or dedicated hardware (such as an ASIC or FPGA).

[0105] A computer program product may also be provided in an embodiment of the present disclosure, including program instructions for executing the monitoring method described in any of the above embodiments.

[0106] The above embodiments are merely illustrative of the principles and effects of this disclosure and are not intended to limit this disclosure. Anyone skilled in the art may modify or alter the above embodiments without departing from the spirit and scope of this disclosure. Therefore, any equivalent modifications or alterations made by a person skilled in the art without departing from the spirit and technical concepts disclosed herein shall be encompassed by the scope of protection of this disclosure.

Claims

1. A monitoring method for preventing stroke, characterized in that: The monitoring method comprises: Acquiring sleep time data of the user entering the current sleep state, and collecting real-time physiological indicator data related to stroke from the user during sleep; wherein the physiological indicator data related to stroke includes facial data and snoring data collected during the sleep state; Predicting a first stroke risk value of a user at a predicted time in the future based on prediction information using a prediction model; wherein the prediction information includes the user's disease profile data and the sleep time data; and Obtaining a second stroke risk value of the user according to the real-time physiological indicator data through a real-time monitoring model; determining a stroke warning event and triggering an intervention action based on an indication of consistency between the changing trends of the first stroke risk value and the second stroke risk value with respect to a stroke warning event; The determination of the stroke warning event based on the consistency of the change trends of the first stroke risk value and the second stroke risk value with respect to the stroke warning event includes at least one of the following: 1) When the change trend of the second stroke risk value continues to rise within a preset time range or reaches a value within a preset range from the first stroke risk value, determining a stroke warning event and triggering an intervention action; 2) when the second stroke risk value reaches a preset risk threshold at a determination time that is a preset time period from the prediction time, determining a stroke warning event and triggering an intervention action; wherein the preset risk threshold is set according to a preset ratio of the first stroke risk value; 3) When the change trend of the second stroke risk value gradually decreases, it is determined that the stroke warning event has not occurred and no intervention action is triggered; 4) Perform at least one intervention action based on the stroke warning event.

2. The monitoring method according to claim 1, characterized in that: Also includes: Detecting user's posture changes during sleep; In response to determining, according to the posture change data, that the user moves from a lying position to a standing position, performing fall event detection; In response to the occurrence of a fall event, a corresponding alarm action is executed.

3. The monitoring method according to claim 1, characterized in that The performing of at least one intervention operation according to the stroke warning event includes: Send risk warning information to relevant personnel associated with the user; The user's location information and physiological indicator data related to stroke are sent to the medical institution associated with the user to prepare for medical treatment.

4. The monitoring method according to claim 1, characterized in that: The step of obtaining facial data from the user's stroke-related physiological indicator data includes: In a sleeping state, a facial image of the user is acquired through a facial acquisition device, and facial feature points are extracted based on the facial image; The user's eye state information and mouth state information are extracted based on facial feature points, and the facial data is obtained based on the eye state information and the mouth state information.

5. The monitoring method according to claim 4, characterized in that: The eye state information is defined as the height difference between the left eyelid and the right eyelid of the user; and the mouth state information is defined as the displacement difference between the left and right corners of the mouth.

6. The monitoring method according to claim 4, characterized in that: The facial acquisition device includes an infrared camera; and / or the facial acquisition device is arranged to face the user's face.

7. The monitoring method according to claim 1, characterized in that: The step of obtaining the snoring data from the user's physiological index data related to stroke includes: Acquiring voice information of a user in a sleeping state through at least one microphone, and extracting snoring audio of the user according to the voice information to obtain snoring characteristics; The snoring data is obtained according to the snoring features through a trained snoring recognition model; wherein the snoring data includes pause ventilation information and ambiguous pronunciation information in the user's snoring.

8. A computer device, characterized in that: include: processor and memory; The memory stores program instructions; The processor is configured to run the program instructions to execute the monitoring method according to any one of claims 1 to 7.

9. A computer-readable storage medium, characterized in that Program instructions are stored, and the program instructions are executed to perform the monitoring method according to any one of claims 1 to 7.

10. A computer program product, characterized in that include: Program instructions for executing the monitoring method according to any one of claims 1 to 7.

Images