A fall prediction method, system, terminal device and storage medium
By analyzing the physical examination data and gait trends of the elderly, combined with road surface information, a multi-level fall warning signal is generated, which solves the problem that existing technologies cannot predict fall risks and achieves effective identification and early warning of fall risks for the elderly.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- 深圳腾信百纳科技有限公司
- Filing Date
- 2023-02-21
- Publication Date
- 2026-07-24
AI Technical Summary
Current technologies are unable to effectively predict and warn of fall risks among the elderly, especially in daily life where potential fall hazards cannot be identified and warned in advance due to the combined effects of endogenous and exogenous factors.
By acquiring the physical examination data of the target personnel, it is determined whether there are any pre-set disease types, and their normal walking posture and dynamic walking posture trends are analyzed. Combined with the road surface information coordinate system and potential dangerous locations, level one and level two fall warning signals are generated, taking into account both endogenous and exogenous factors.
It improves the accuracy of fall risk prediction, reduces the occurrence of fall accidents, provides timely warnings and emergency remedial prompts, and enhances the safety of the elderly.
Smart Images
Figure CN116229675B_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of human body recognition technology, and in particular to a fall prediction method, system, terminal device and storage medium. Background Technology
[0002] Currently, with the accelerating aging process in my country, the decline in bodily functions and balance control leads to an increased risk of falls. Falls among the elderly are prone to causing physical injuries, which in turn affect their daily lives. Fall detection is the foundation for implementing fall protection and provides support for fall injury assessment and timely assistance.
[0003] Fall detection primarily uses sensors to perceive human behavior data, preprocesses and extracts features from the data, and ultimately identifies falls. Sensors used for fall detection can be divided into wearable sensors and environmental sensing sensors. Wearable sensors need to be worn on the body and are suitable for outdoor locations during the day. Environmental sensing sensors, including video, infrared, and sound sensors, do not need to be worn and are suitable for specific spaces with a high risk of falls, such as bathrooms, kitchens, and bedrooms.
[0004] In real life, there are many factors that cause people to fall. These factors may be endogenous factors related to the body's own constitution or exogenous factors caused by external impacts. Falls may also occur due to a combination of factors. Since wearable sensors or environmental sensors can only identify and detect falls that are happening or have already happened, they cannot effectively prevent or warn against factors that may cause falls in advance. Summary of the Invention
[0005] To improve the effective identification and prediction of risk factors that may lead to falls, this application provides a fall prediction method, system, terminal device, and storage medium.
[0006] Firstly, this application provides a fall prediction method, comprising the following steps:
[0007] Obtain the physical examination data of the target personnel;
[0008] Based on the physical examination data, determine whether the target person has a preset disease type;
[0009] If the target person has the preset disease type, then obtain the target person's normal walking posture;
[0010] If the target person's current walking posture does not conform to the normal walking posture, then obtain the dynamic walking posture trend corresponding to the current walking posture;
[0011] If the dynamic walking trend meets the preset fall trend standard, a corresponding first-level fall warning signal is generated;
[0012] Based on the first-level fall warning signal, a road surface information coordinate system corresponding to the current walking path of the target person is generated;
[0013] Identify the road surface information coordinate system and generate the corresponding coordinates of potential dangerous locations and the straight line axis of the walking direction of the target person;
[0014] If the walking direction straight line axis coincides with the coordinates of the potential danger location, then determine whether the target distance between the target person and the coordinates of the potential danger location is within the warning distance threshold;
[0015] If the target distance between the target person and the coordinates of the potentially dangerous location is within the warning distance threshold, a corresponding secondary fall warning signal is generated.
[0016] By adopting the above technical solution, in order to comprehensively consider various factors when a target person faces the risk of falling, the following steps are taken: First, the target person's corresponding endogenous factors are analyzed, that is, whether the target person has a pre-set disease type that leads to the risk of falling. If so, it is further determined whether the target person's current walking posture conforms to the normal walking posture in daily life. If not, it indicates that the target person already has the initial manifestations of the risk of falling. Then, it is determined whether the target person's dynamic walking posture trend conforms to the precursor of the risk of falling, that is, the pre-set fall trend standard. If it does, a first-level fall warning signal for the initial fall risk prediction is generated. Based on the generation of the first-level fall warning signal, the target person's exogenous factors, that is, the current walking road environment, are analyzed, and a corresponding road information coordinate system is established. It is determined whether the target person's walking route, that is, the straight line axis of the walking direction, coincides with the coordinates of the potential danger position. If they coincide, a corresponding second-level fall warning signal is generated by combining the actual target distance between the target person and the coordinates of the potential danger position. By combining the target person's endogenous and exogenous factors for comprehensive analysis and judgment, the effective identification and prediction of the risk factors that lead to the risk of falling is improved.
[0017] Optionally, after obtaining the target person's typical gait if the target person has the preset disease type, the following steps are further included:
[0018] If the target person's current walking posture conforms to the normal walking posture, then obtain the target person's current heart rate value;
[0019] Determine whether the current heart rate change value exceeds a preset heart rate change threshold;
[0020] If the current heart rate change value exceeds the preset heart rate change threshold, then the preset heart rate change threshold and the current heart rate change value are combined to generate a corresponding heart rate change difference as a heart rate warning signal.
[0021] By adopting the above technical solution, when the target person's current heart rate change value exceeds the preset heart rate change threshold, a corresponding heart rate change difference is generated as a heart rate warning signal, thereby improving the analysis and prediction of fall risk to a certain extent.
[0022] Optionally, generating a corresponding Level 1 fall warning signal if the dynamic walking trend conforms to a preset fall trend standard includes the following steps:
[0023] If the dynamic walking trend matches the preset fall trend standard, then determine whether the target person has taken preset fall risk medication;
[0024] If the target person takes the preset potential hazard drug, then obtain the corresponding potential hazard drug;
[0025] Identify the target potentially harmful drug and obtain the corresponding associated symptoms;
[0026] Based on the associated symptoms, a corresponding fall risk level is generated as the Level 1 fall warning signal.
[0027] By adopting the above technical solution, based on the target personnel's use of the target potentially hazardous drugs, the corresponding related symptoms can be further inferred, thereby effectively predicting and analyzing the relevant safety factors that may lead to a fall risk.
[0028] Optionally, after generating a corresponding Level 1 fall warning signal if the dynamic walking trend meets the preset fall trend standard, the following steps are further included:
[0029] Based on the Level 1 fall warning signal, obtain the complication symptoms of the target person corresponding to the preset disease type;
[0030] Identify the aforementioned symptoms and output corresponding emergency remedy prompts.
[0031] By adopting the above technical solution, corresponding emergency remedial prompts are output based on the possible complications of the target personnel, thereby reducing the occurrence of fall hazards.
[0032] Optionally, identifying the concurrent symptoms and outputting corresponding emergency remedial prompts includes the following steps:
[0033] If there are multiple complications, then obtain the target person's historical symptom records;
[0034] If the concurrent symptoms exist in the historical symptom record, then obtain the concurrent occurrence number of the concurrent symptoms for the target person in the historical symptom record;
[0035] Based on the number of concurrent occurrences, the output priority of the emergency remedial prompt corresponding to the concurrent symptoms is set, and the number of concurrent occurrences is proportional to the output priority;
[0036] Based on the output priority, the corresponding emergency remedial prompt is output.
[0037] By adopting the above technical solution, the output priority of emergency remedial prompts corresponding to the current possible complications can be set according to the number of concurrent occurrences of the target personnel's historical complications. This can effectively reduce the occurrence of fall hazards caused by untimely output of emergency remedial prompts.
[0038] Optionally, generating a corresponding secondary fall warning signal if the target distance between the target person and the coordinates of the potentially dangerous location is within the warning distance threshold includes the following steps:
[0039] If the target distance between the target person and the coordinates of the potential danger location is within the warning distance threshold, then determine whether there are multiple coordinates of the potential danger location;
[0040] If there are multiple potential hazard location coordinates, the broadcast priority of the potential hazard type corresponding to each potential hazard location coordinate is set according to the target distance between each potential hazard location coordinate and the target person, and the target distance is inversely proportional to the broadcast priority of the hazard type.
[0041] Based on the broadcast priority of the potential hazard type, the potential hazard type corresponding to the coordinates of the potential hazard location is broadcast, and the corresponding secondary fall warning signal is generated.
[0042] By adopting the above technical solution, the potential hazard type corresponding to the location coordinates of the potential hazard is broadcast according to the broadcast priority of the potential hazard type, so as to obtain the potential hazard factors corresponding to the fall risk close to the target personnel in a timely manner, thereby improving the target personnel's ability to predict potential hazard factors.
[0043] Optionally, after broadcasting the potential hazard type corresponding to the potential hazard location coordinates according to the potential hazard type broadcasting priority and generating the corresponding secondary fall warning signal, the method further includes the following steps:
[0044] The potential hazard types are analyzed based on preset potential hazard occurrence criteria to generate corresponding fall hazard occurrence probabilities;
[0045] Based on the probability of the fall, the safety hazard level corresponding to the potential hazard type is set.
[0046] By adopting the above technical solution, a safety hazard level is set according to the probability of falling, which can effectively distinguish the degree of danger of the potential hazard type and further prevent the risk of falling.
[0047] Secondly, this application provides a fall prediction system, including:
[0048] The first acquisition module is used to acquire the physical examination data of the target personnel;
[0049] The first judgment module is used to determine whether the target person has a preset disease type based on the physical examination data;
[0050] The second acquisition module, if the target person has the preset disease type, is used to acquire the target person's corresponding normal gait;
[0051] If the target person's current walking posture does not conform to the conventional walking posture, the third acquisition module is used to acquire the dynamic walking posture trend corresponding to the current walking posture.
[0052] The first generation module generates a corresponding first-level fall warning signal if the dynamic walking trend meets the preset fall trend standard.
[0053] The second generation module is used to generate a road surface information coordinate system corresponding to the current walking path of the target person based on the first-level fall warning signal;
[0054] The third generation module is used to identify the road surface information coordinate system and generate the corresponding coordinates of the potential danger location and the straight line axis of the walking direction of the target person.
[0055] The second judgment module is used to determine whether the target distance between the target person and the coordinates of the potential danger location is within the warning distance threshold if the walking direction straight line axis coincides with the coordinates of the potential danger location.
[0056] The fourth generation module generates a corresponding secondary fall warning signal if the target distance between the target person and the coordinates of the potential danger location is within the warning distance threshold.
[0057] By adopting the above technical solution, in order to comprehensively consider various factors when a target person faces the risk of falling, the first judgment module first analyzes the target person's corresponding endogenous factors, that is, whether the target person has a preset disease type that leads to the risk of falling. If so, it further judges whether the target person's current gait conforms to the normal gait in daily life. If not, it indicates that the target person already has the initial manifestations of the risk of falling. Then, it judges whether the target person's dynamic gait trend conforms to the precursor of the risk of falling, that is, the preset fall trend standard. If it does, the first generation module generates a level one fall risk prediction for the initial fall risk. The fall warning signal, based on the generation of the first-level fall warning signal, analyzes the external factors of the target person, namely the current walking road environment. The second generation module establishes and generates a corresponding road information coordinate system to determine whether the target person's walking route, i.e., the straight line axis of the walking direction, coincides with the coordinates of the potential danger location. If they coincide, the fourth generation module further generates a corresponding second-level fall warning signal by combining the actual target distance between the target person and the coordinates of the potential danger location. By combining the target person's internal and external factors for comprehensive analysis and judgment, the effective identification and prediction of the risk factors that may lead to falls are improved.
[0058] Thirdly, this application provides a terminal device, which adopts the following technical solution:
[0059] A terminal device includes a memory and a processor. The memory stores computer instructions that can be executed on the processor. When the processor loads and executes the computer instructions, it employs the aforementioned fall prediction method.
[0060] By adopting the above technical solution, a fall prediction method is generated into computer instructions and stored in a memory for loading and execution by a processor. This allows for the creation of a terminal device based on the memory and processor, making it convenient to use.
[0061] Fourthly, this application provides a computer-readable storage medium, which adopts the following technical solution:
[0062] A computer-readable storage medium storing computer instructions, wherein when the computer instructions are loaded and executed by a processor, the aforementioned fall prediction method is employed.
[0063] By adopting the above technical solution, a fall prediction method is generated into computer instructions and stored in a computer-readable storage medium for loading and execution by a processor. The computer-readable storage medium facilitates the reading and storage of computer instructions.
[0064] In summary, this application includes at least one of the following beneficial technical effects: To comprehensively consider various factors when a target person faces the risk of falling, the application first analyzes the target person's corresponding endogenous factors, i.e., whether the target person has a pre-existing disease type that could lead to a fall risk. If so, it further determines whether the target person's current gait conforms to a normal gait in daily life. If not, it indicates that the target person already exhibits initial signs of a fall risk. Then, it determines whether the target person's dynamic gait trend conforms to the precursory signs of a fall, i.e., the pre-existing fall trend standard. If so, it generates an initial fall risk indicator. The predicted first-level fall warning signal, based on the analysis of the target person's external factors, i.e., the current walking road environment, is generated to establish a corresponding road surface information coordinate system. It is determined whether the target person's walking route, i.e., the straight line axis of the walking direction, coincides with the coordinates of a potential danger location. If they coincide, a corresponding second-level fall warning signal is generated by combining the actual target distance between the target person and the coordinates of the potential danger location. By combining the target person's internal and external factors for comprehensive analysis and judgment, the effective identification and prediction of risk factors that may lead to falls is improved. Attached Figure Description
[0065] Figure 1 This is a flowchart illustrating steps S101 to S109 of a fall prediction method according to this application.
[0066] Figure 2 This is a flowchart illustrating steps S201 to S203 of a fall prediction method according to this application.
[0067] Figure 3 This is a flowchart illustrating steps S301 to S304 of a fall prediction method according to this application.
[0068] Figure 4 This is a flowchart illustrating steps S401 to S402 in a fall prediction method according to this application.
[0069] Figure 5 This is a flowchart illustrating steps S501 to S504 of a fall prediction method according to this application.
[0070] Figure 6 This is a flowchart illustrating steps S601 to S603 of a fall prediction method according to this application.
[0071] Figure 7 This is a flowchart illustrating steps S701 to S702 in a fall prediction method according to this application.
[0072] Figure 8 This is a schematic diagram of a fall prediction system according to this application.
[0073] Explanation of reference numerals in the attached figures:
[0074] 1. First acquisition module; 2. First judgment module; 3. Second acquisition module; 4. Third acquisition module; 5. First generation module; 6. Second generation module; 7. Third generation module; 8. Second judgment module; 9. Fourth generation module. Detailed Implementation
[0075] The following is in conjunction with the appendix Figure 1-8 This application will be described in further detail.
[0076] This application discloses a fall prediction method, such as... Figure 1 As shown, it includes the following steps:
[0077] S101. Obtain the physical examination data of the target personnel;
[0078] S102. Determine whether the target person has a pre-defined disease type based on the physical examination data;
[0079] S103. If the target person has a preset disease type, then obtain the target person's corresponding normal walking posture;
[0080] S104. If the target person's current walking posture does not conform to the normal walking posture, then obtain the dynamic walking posture trend corresponding to the current walking posture;
[0081] S105. If the dynamic walking trend meets the preset fall trend standard, a corresponding first-level fall warning signal is generated;
[0082] S106. Based on the Level 1 fall warning signal, generate a coordinate system for the road surface information corresponding to the current walking path of the target person;
[0083] S107. Identify the road surface information coordinate system and generate the corresponding coordinates of potential dangerous locations and the straight line axis of the walking direction of the target personnel;
[0084] S108. If the straight line axis of the walking direction coincides with the coordinates of the potential danger location, then determine whether the target distance between the target person and the coordinates of the potential danger location is within the warning distance threshold;
[0085] S109. If the target distance between the target person and the coordinates of the potentially dangerous location is within the warning distance threshold, a corresponding secondary fall warning signal is generated.
[0086] In practical applications, a fall refers to a sudden and unexpected fall to the ground. Falls can occur at any age, but are more common in the elderly. Therefore, in step S101, the target group of this application is the elderly.
[0087] Secondly, the physical examination data in step S101 refers to the periodic physical examination report data of the elderly. The physical examination data can be obtained by entering the physical examination report of the elderly. Through the physical examination data, we can know the various health conditions of the elderly, such as whether they have disabilities or whether they have some underlying diseases that have not been cured for many years.
[0088] Further in step S102, based on the physical examination data, it can be determined whether the elderly person has a pre-defined disease type. A pre-defined disease type refers to a disease type that may cause the elderly person to have a risk of falling while walking. For example, stroke, ischemic attack, seizures, cerebellar diseases, dementia and other neurological diseases are all diseases that increase the risk of falls in the elderly person.
[0089] In step S103, if the target person, i.e. the elderly, has a preset disease type, then their corresponding normal walking posture is further obtained. Normal walking posture refers to the posture of the elderly when walking in daily life, including walking forward in a straight line, walking sideways, walking backward, turning, walking on tiptoe or heel, etc. The normal walking posture of the elderly can be recorded by setting up a corresponding monitoring camera. The installation scenario of the monitoring camera can be determined in combination with the actual application situation.
[0090] In step S104, if the target person's current walking posture does not conform to the normal walking posture, it means that the elderly person's current walking posture has been affected by the above-mentioned preset disease type, thereby increasing the risk of falling. In order to conduct a more detailed analysis of the elderly person's current walking posture, the dynamic walking posture trend corresponding to the elderly person's current walking posture is further obtained. The dynamic walking posture trend refers to the corresponding dynamic walking posture trend of the elderly person within a specified time period.
[0091] In practical applications, among the factors that cause abnormal walking posture in the elderly, neurological disorders are the most common, such as stroke, Parkinson's disease, peripheral neuropathy, and various spinal cord diseases. These can all affect the walking posture of the elderly, causing them to take small, quick steps forward after starting to walk, with their feet not leaving the ground, shuffling along the ground, and their bodies leaning forward, as if they are about to fall to the ground.
[0092] Some of these changes in gait are due to psychological factors, such as depression, which manifests as a fear of falling and a reluctance to move forward even when encountering very small obstacles. Other causes include various bone and joint diseases and hypothyroidism.
[0093] In step S105, if the dynamic walking trend meets the preset fall trend standard, it means that the elderly person is very likely to fall. In order to make a prediction in advance and reduce the occurrence of falls, a corresponding first-level fall warning signal is generated to remind the elderly person or people around them. The preset fall trend standard refers to the various walking trend standards corresponding to the risk of falling that are set in advance. For example, the elderly person takes small steps and rushes forward quickly after starting.
[0094] In step S106, in order to further analyze the risk factors corresponding to the risk of falls among the elderly, based on the analysis of the endogenous factors that cause falls among the elderly, the analysis of the exogenous factors that cause falls among the elderly is further carried out. That is, based on the environmental information of the elderly being in the current location recorded by the camera, a road surface information coordinate system corresponding to the road surface where the elderly are currently walking is generated. The road surface information coordinate system includes the coordinate information of the road surface where the elderly are currently walking and the coordinate information of the objects on the road surface. The road surface here is not limited to the street where people walk, but can also include the ground inside the building.
[0095] In step S107, by identifying the road surface information coordinate system, the coordinates of potential dangerous locations corresponding to the current road surface and the straight line axis of the walking direction corresponding to the target person can be further generated. The coordinates of potential dangerous locations refer to the coordinates of the locations on the road surface where there is a potential risk of falling. Potential risks of falling include inconspicuous protrusions on the ground, smooth ground, door thresholds that are too high or difficult to notice, etc. The straight line axis of the walking direction refers to the straight line axis formed with the current walking direction of the elderly person as the reference.
[0096] In step S108, if the straight line axis of the walking direction coincides with the coordinates of the potential danger location, it means that if the elderly person walks along the straight line without noticing the potential danger location, there is a high risk of falling. In order to make a reasonable risk analysis based on the actual walking situation of the elderly person, it is determined whether the target distance between the elderly person and the coordinates of the potential danger location is within the warning distance threshold. The warning distance threshold refers to the pre-set safe warning distance range between the elderly person and the potential danger location.
[0097] In step S109, if the target distance between the target person and the coordinates of the potential danger location is within the warning distance threshold, it indicates that the elderly person did not notice the potential danger location in time during the walking process. Then, a corresponding secondary fall warning signal is generated to remind the elderly person or people around them.
[0098] For example, if the potential hazard is a room threshold, the elderly person's coordinates on the road information coordinate system are (3,0), the coordinates of the potential hazard location are (5,0), and the corresponding warning distance threshold is 3 centimeters. Here, the unit interval between coordinates on the X-axis is 1 centimeter, so the target distance can be calculated to be 2 centimeters. It can be further determined that the target distance between the elderly person and the room threshold is within the warning distance threshold, and a corresponding secondary fall warning signal is generated. This secondary fall warning signal can be a voice alarm prompt or a fall protection mechanism can be triggered based on the secondary fall warning signal. This fall protection mechanism can be some fall protection devices.
[0099] This embodiment provides a fall prediction method that comprehensively considers various factors when a target person faces the risk of falling. First, it analyzes the target person's endogenous factors, specifically whether the target person has a pre-defined disease type that could lead to a fall. If so, it further determines whether the target person's current gait conforms to a normal gait in daily life. If not, it indicates that the target person already exhibits initial signs of a fall risk. Then, it determines whether the target person's dynamic gait trend conforms to a pre-existing fall hazard, i.e., a pre-defined fall trend standard. If so, it generates a first-level fall warning signal for initial fall risk prediction. Based on the first-level fall warning signal, it analyzes the target person's exogenous factors, i.e., the current walking environment, and establishes a corresponding road surface information coordinate system. It determines whether the target person's walking route (i.e., the straight line axis of the walking direction) coincides with the coordinates of a potential danger location. If they coincide, it further generates a corresponding second-level fall warning signal based on the actual target distance between the target person and the potential danger location coordinates. By combining the target person's endogenous and exogenous factors for comprehensive analysis and judgment, it improves the effective identification and prediction of risk factors that could lead to a fall.
[0100] In one embodiment of this example, such as Figure 2 As shown, after obtaining the target person's normal gait in step S103 (if the target person has a preset disease type), the following steps are also included:
[0101] S201. If the target person's current walking posture conforms to the normal walking posture, then obtain the target person's current heart rate value;
[0102] S202. Determine whether the current heart rate change value exceeds the preset heart rate change threshold;
[0103] S203. If the current heart rate change value exceeds the preset heart rate change threshold, then the preset heart rate change threshold and the current heart rate change value are combined to generate a corresponding heart rate change difference as a heart rate warning signal.
[0104] In step S201, if the target person's current walking posture conforms to the normal walking posture, in order to further improve the prevention of fall risk in the elderly, the target person's current heart rate value is obtained. The current heart rate value refers to the number of times the elderly person's heart beats within a standard time.
[0105] In practical applications, as bodily functions decline with age, it is more common for the elderly to experience arrhythmia, i.e., unstable heart rate. When arrhythmia is severe, the elderly may faint and fall, and there is even a possibility of sudden death. Therefore, by obtaining the current heart rate value of the elderly, the risk of falling can be predicted to a certain extent.
[0106] Among them, the current heart rate value of the elderly can be obtained by wearing a heart rate monitoring terminal, such as a vital signs monitoring bracelet. The bracelet monitors health data such as heart rate and blood pressure through its built-in sensors, and then sends the monitored health data to the server for analysis via Bluetooth.
[0107] In step S202, the preset heart rate change threshold refers to the normal heart rate range for the elderly. By determining whether the elderly person's current heart rate change value exceeds the preset heart rate change threshold, the elderly person's heart rate status can be monitored in real time. In practical applications, if the elderly person's heart rate value exceeds the corresponding preset heart rate change threshold, it may lead to arrhythmia, which is one of the common causes of falls in the elderly.
[0108] In step S203, if the current heart rate change value of the elderly exceeds the preset heart rate change threshold, in order to reduce the risk of falls, the preset heart rate change threshold and the current heart rate change value of the elderly are combined to generate a corresponding heart rate change difference as a heart rate warning signal. Through this heart rate warning signal, the current heart rate of the elderly can be analyzed and judged in a timely manner, thereby effectively reducing the occurrence of falls.
[0109] The fall prediction method provided in this embodiment generates a corresponding heart rate change difference as a heart rate warning signal when the current heart rate change value of the target person exceeds a preset heart rate change threshold, thereby improving the analysis and prediction of fall risk to a certain extent.
[0110] In one embodiment of this example, such as Figure 3 As shown, step S105, which involves generating a corresponding Level 1 fall warning signal if the dynamic walking trend meets the preset fall trend standard, includes the following steps:
[0111] S301. If the dynamic walking trend meets the preset fall trend criteria, determine whether the target person has taken the preset fall risk medication;
[0112] S302. If the target personnel take a pre-set potential hazard drug, then obtain the corresponding potential hazard drug;
[0113] S303. Identify the target potentially hazardous drug and obtain the corresponding associated symptoms;
[0114] S304. Based on the associated symptoms, generate the corresponding fall risk level as a Level 1 fall warning signal.
[0115] In step S301, if the elderly person's dynamic walking trend meets the preset fall trend standard, then in order to further analyze the reasons for the change in the elderly person's walking posture, it is determined whether the elderly person has taken preset fall risk drugs. Preset fall risk drugs refer to drugs that are likely to cause elderly people to fall after taking them.
[0116] In practice, besides aging and decreased balance, some falls in the elderly are caused by certain medications, which are designed to increase the risk of falls. For example, sedative-hypnotic drugs, such as estazolam, zolpidem, and zopiclone, can affect postural stability and balance, increasing the risk of falls when getting up at night. Antidepressants, such as monoamine oxidase inhibitors and tricyclic antidepressants, can cause blurred vision, drowsiness, rapid heartbeat, difficulty urinating, and orthostatic hypotension, all of which increase the risk of falls.
[0117] Furthermore, antiepileptic drugs can cause dizziness, headaches, unsteady gait, and difficulty concentrating in some elderly people as the dosage increases, thus increasing the risk of falls. Antihypertensive drugs are also problematic, as most elderly people have high blood pressure and need to take medication regularly. However, some antihypertensive drugs can cause orthostatic hypotension, especially when getting up or changing positions, which can easily cause instability and increase the risk of falls. Finally, hypoglycemic drugs are necessary for some elderly people with diabetes, requiring medication to control blood sugar in addition to a healthy diet. However, hypoglycemic drugs can cause adverse reactions such as hypoglycemia, which may lead to a strong feeling of hunger, cold sweats, palpitations, general weakness, tremors, and blurred vision, potentially causing falls.
[0118] In steps S302 to S303, if the elderly person has taken the aforementioned preset potential hazard drugs, then the corresponding target potential hazard drugs are further obtained. The target potential hazard drugs refer to the types of drugs among the preset potential hazard drugs that the elderly person has specifically taken.
[0119] Furthermore, identify the target potentially dangerous drug and obtain the corresponding associated symptoms. Associated symptoms refer to the symptoms that appear in the body after taking the target potentially dangerous drug. For example, if the target potentially dangerous drug is an anti-epileptic drug, the corresponding associated symptoms are dizziness, headache, unsteady gait, and inattention.
[0120] In step S304, based on the associated symptoms obtained above, a corresponding fall risk level is generated as a Level 1 fall warning signal. Based on this Level 1 fall warning signal, an accurate judgment can be made in a timely and effective manner regarding any abnormalities in the elderly person's gait, so as to better prevent potential fall risks in the future.
[0121] The fall prediction method provided in this embodiment can further infer the corresponding related symptoms based on the target person's use of the target potentially hazardous medication, thereby effectively predicting and analyzing the relevant safety factors that may lead to a fall risk.
[0122] In one embodiment of this example, such as Figure 4 As shown, after step S105, which generates a corresponding first-level fall warning signal if the dynamic walking trend meets the preset fall trend standard, the following steps are also included:
[0123] S401. Based on the Level 1 fall warning signal, obtain the concurrent symptoms of the target person corresponding to the preset disease type;
[0124] S402. Identify concurrent symptoms and output corresponding emergency remedial prompts.
[0125] In step S401, after a Level 1 fall warning signal is issued, in order to further analyze the risk factors for falls in the elderly, the corresponding complications of the preset disease type suffered by the elderly are obtained and analyzed, because the complications may affect the elderly’s current walking posture. If the elderly’s current walking posture is seriously affected, there is a risk of falling at any time.
[0126] Among them, complications refer to the symptoms that may occur in elderly people with the corresponding preset disease type. For example, if the preset disease type is osteoarthritis, more specifically lumbar muscle strain and degenerative changes, the corresponding complications are a decrease in the spine's ability to compensate for the center of gravity of the lower limbs, leading to reduced joint stability and falls.
[0127] In step S402, the emergency remedial prompt refers to the emergency remedial treatment prompt for falls caused by complications. After identifying the complications, it can be determined which specific physical defects in the elderly person are causing the risk of falling.
[0128] For example, the increased number of chronic diseases among the elderly will significantly increase the risk of falls. If the disease type is arthritis, the corresponding complications include decreased muscle tone and joint mobility impairment, which can change the elderly’s walking posture and even cause falls. The corresponding emergency rescue tips are to stay calm, not to move around if you feel pain or injury, and to shout loudly or pat the ground or wall to attract attention and seek help.
[0129] The fall prediction method provided in this embodiment, by adopting the above-mentioned technical solution, outputs corresponding emergency remedial prompts based on the possible complications of the target person, thereby reducing the occurrence of fall hazards.
[0130] In one embodiment of this example, such as Figure 5 As shown, step S402, which identifies concurrent symptoms and outputs corresponding emergency remedial prompts, includes the following steps:
[0131] S501. If there are multiple concurrent symptoms, obtain the target person's historical symptom records;
[0132] S502. If there are concurrent symptoms in the historical symptom record, then obtain the number of concurrent occurrences of the corresponding concurrent symptoms of the target person in the historical symptom record;
[0133] S503. Based on the number of concurrent occurrences, set the output priority of emergency remedial prompts corresponding to concurrent symptoms. The number of concurrent occurrences is directly proportional to the output priority.
[0134] S504. Output the corresponding emergency remedial prompts according to the output priority.
[0135] In step S501, if there are multiple complications, it indicates that the elderly person has multiple physical abnormalities. In order to improve the accuracy of outputting corresponding emergency remedial prompts for complications, the elderly person's historical symptom records are obtained, which are records of each occurrence of complications in the elderly person.
[0136] In steps S502 to S504, in order to more accurately determine the types of complications and output corresponding emergency remedial prompts in a timely manner, the number of times the corresponding complications occurred in the elderly in the historical symptom records is obtained. Then, based on the number of occurrences, the output priority of the emergency remedial prompts corresponding to the complications is set. The number of occurrences is directly proportional to the output priority. That is to say, the more times the complications occur, the greater the probability that the elderly person will have the same complications.
[0137] For example, if the historical symptom record shows corresponding concurrent symptoms such as cerebral ischemic attack and musculoskeletal disease, with 3 occurrences of cerebral ischemic attack and 1 occurrence of musculoskeletal disease, then the output priority of the emergency remedial prompt for cerebral ischemic attack should be set higher than that of the emergency remedial prompt for musculoskeletal disease. Therefore, the emergency remedial prompt for cerebral ischemic attack should be output first, followed by the emergency remedial prompt for musculoskeletal disease.
[0138] The fall prediction method provided in this embodiment sets the output priority of emergency remedial prompts corresponding to the current possible complications based on the number of concurrent occurrences of the target person's historical complications. This can effectively reduce the occurrence of fall hazards caused by untimely output of emergency remedial prompts.
[0139] In one embodiment of this example, such as Figure 6 As shown, step S109, which involves generating a corresponding secondary fall warning signal if the target distance between the target person and the coordinates of the potentially dangerous location is within the warning distance threshold, includes the following steps:
[0140] S601. If the target distance between the target personnel and the coordinates of the potential danger location is within the warning distance threshold, then determine whether there are multiple potential danger location coordinates;
[0141] S602. If there are multiple potential hazard location coordinates, the broadcast priority of the potential hazard type corresponding to each potential hazard location coordinate shall be set according to the target distance between each potential hazard location coordinate and the target personnel. The target distance is inversely proportional to the broadcast priority of the hazard type.
[0142] S603. Broadcast the potential hazard type corresponding to the potential hazard location coordinates according to the broadcast priority of potential hazard type, and generate the corresponding secondary fall warning signal.
[0143] In steps S601 to S602, in order to comprehensively analyze the potential dangers on the ground information where the elderly are currently walking, if the target distance between the target person and the coordinates of the potential danger location is within the warning distance threshold, it is determined whether there are multiple potential danger location coordinates. If there are multiple potential danger location coordinates, it means that there are multiple potential dangers on the ground where the elderly are currently walking in a straight line.
[0144] Secondly, in order to provide timely risk warnings to the elderly or other personnel, the broadcast priority of the potential hazard type corresponding to each potential hazard location coordinate is set according to the target distance between each potential hazard location coordinate and the target person. The target distance is inversely proportional to the hazard type broadcast priority. That is to say, the closer the elderly are to the target distance between the potential hazard location coordinates, the higher the corresponding potential hazard type broadcast priority, and vice versa.
[0145] In step S603, after the priority for broadcasting potential hazard types corresponding to the potential hazard location coordinates is set, the potential hazard types corresponding to the potential hazard location coordinates are broadcast according to the broadcast priority. The broadcast content includes the specific type of potential hazard and the specific location of the potential hazard. For example, the broadcast content for the potential hazard type corresponding to the potential hazard location coordinates according to the broadcast priority is: the current potential hazard type is slippery ground, and the current distance is 1 meter directly in front.
[0146] The fall prediction method provided in this embodiment broadcasts the potential hazard type corresponding to the coordinates of the potential hazard location according to the broadcast priority of the potential hazard type, thereby timely obtaining the potential hazard factors corresponding to the fall risk close to the target person, and improving the target person's ability to predict potential hazard factors.
[0147] In one embodiment of this example, such as Figure 7 As shown, after step S603, which involves broadcasting the potential hazard type corresponding to the potential hazard location coordinates according to the potential hazard type broadcast priority and generating the corresponding secondary fall warning signal, the following steps are also included:
[0148] S701. Analyze the types of potential hazards based on preset potential hazard occurrence standards, and generate the corresponding probability of fall hazards occurring;
[0149] S702. Based on the probability of a fall, set the safety hazard level corresponding to the type of potential hazard.
[0150] In practical application, the preset potential hazard incident standard in step S701 refers to the standard for the probability of a fall risk caused by a corresponding safety factor. The fall hazard occurrence probability refers to the probability of an elderly person falling due to various types of potential hazards. For example, potential hazard types include high room thresholds and slippery floors. According to the preset potential hazard incident standard, the probability of an elderly person falling due to slippery floors is 80%, and the probability of an elderly person falling due to high room thresholds is 60%.
[0151] In step S702, in order to further indicate the degree of danger of various potential hazard types, the safety hazard level corresponding to the potential hazard type is set according to the probability of falling. The safety hazard level is directly proportional to the probability of falling. That is, the higher the probability of falling for a potential hazard type, the higher the corresponding safety hazard level.
[0152] The fall prediction method provided in this embodiment sets a safety hazard level corresponding to the potential hazard type based on the probability of a fall. The safety hazard level can effectively distinguish the degree of danger of the potential hazard type, and further effectively prevent the risk of falls.
[0153] This application discloses a fall prediction system, such as... Figure 8 As shown, it includes:
[0154] The first acquisition module 1 is used to acquire the physical examination data of the target personnel;
[0155] The first judgment module 2 is used to determine whether the target person has a preset disease type based on the physical examination data;
[0156] If the target person has a preset disease type, the second acquisition module 3 is used to acquire the target person's corresponding normal walking posture.
[0157] If the target person's current walking posture does not conform to the normal walking posture, the third acquisition module 4 is used to acquire the dynamic walking posture trend corresponding to the current walking posture.
[0158] First generation module 5: If the dynamic walking trend meets the preset fall trend standard, then generate the corresponding first-level fall warning signal.
[0159] The second generation module 6 is used to generate a road surface information coordinate system corresponding to the current walking path of the target person based on the first-level fall warning signal;
[0160] The third generation module 7 is used to identify the road surface information coordinate system and generate the corresponding coordinates of potential dangerous locations and the straight line axis of the walking direction of the target personnel.
[0161] If the straight line axis of the walking direction coincides with the coordinates of the potential danger location, the second judgment module 8 is used to determine whether the target distance between the target person and the coordinates of the potential danger location is within the warning distance threshold.
[0162] The fourth generation module 9 generates a corresponding secondary fall warning signal if the target distance between the target person and the coordinates of the potential danger location is within the warning distance threshold.
[0163] The fall prediction method provided in this embodiment, in order to comprehensively consider various factors when a target person faces the risk of falling, firstly, the first judgment module 2 analyzes the target person's corresponding endogenous factors, that is, whether the target person has a preset disease type that leads to the risk of falling. If so, it further judges whether the target person's current gait conforms to the normal gait in daily life. If not, it indicates that the target person already has the initial manifestations of the risk of falling. Then, it judges whether the target person's dynamic gait trend conforms to the precursor of the risk of falling, that is, the preset fall trend standard. If it does, the first generation module 5 generates a level one of the initial fall risk prediction. The fall warning signal, based on the generation of the first-level fall warning signal, analyzes the external factors of the target person, namely the current walking road environment. The second generation module 6 establishes and generates a corresponding road information coordinate system to determine whether the target person's walking route, i.e., the straight line axis of the walking direction, coincides with the coordinates of the potential danger location. If they coincide, the fourth generation module 9 further generates a corresponding second-level fall warning signal by combining the actual target distance between the target person and the coordinates of the potential danger location. By combining the target person's internal and external factors for comprehensive analysis and judgment, the effective identification and prediction of the risk factors that may lead to falls are improved.
[0164] It should be noted that the fall prediction system provided in this application embodiment also includes each module and / or corresponding sub-module corresponding to the logical function or logical step of any of the above fall prediction methods, to achieve the same effect as each logical function or logical step, which will not be elaborated here.
[0165] This application also discloses a terminal device, including a memory, a processor, and computer instructions stored in the memory and capable of running on the processor, wherein the processor executes the computer instructions using any of the fall prediction methods described in the above embodiments.
[0166] The terminal device can be a computer device such as a desktop computer, a laptop computer, or a cloud server. The terminal device includes, but is not limited to, a processor and a memory. For example, the terminal device may also include input / output devices, network access devices, and buses.
[0167] The processor can be a central processing unit (CPU). Of course, depending on the actual use, it can also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), off-the-shelf programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor, etc., and this application does not limit it.
[0168] The memory can be an internal storage unit of the terminal device, such as a hard disk or RAM of the terminal device, or an external storage device of the terminal device, such as a plug-in hard disk, smart memory card (SMC), secure digital card (SD), or flash memory card (FC) equipped on the terminal device. Furthermore, the memory can be a combination of internal storage units and external storage devices of the terminal device. The memory is used to store computer instructions and other instructions and data required by the terminal device. The memory can also be used to temporarily store data that has been output or will be output. This application does not limit this.
[0169] In this terminal device, any of the fall prediction methods in the above embodiments can be stored in the terminal device's memory and loaded and executed on the terminal device's processor for convenient use.
[0170] This application also discloses a computer-readable storage medium, which stores computer instructions, wherein when the computer instructions are executed by a processor, any of the fall prediction methods described in the above embodiments are employed.
[0171] The computer instructions can be stored in a computer-readable medium. The computer instructions include computer instruction code, which can be in the form of source code, object code, executable file, or certain middleware. The computer-readable medium includes any entity or device capable of carrying computer instruction code, recording media, USB flash drive, portable hard drive, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. It should be noted that the computer-readable medium includes, but is not limited to, the above-mentioned components.
[0172] In this computer-readable storage medium, any one of the fall prediction methods in the above embodiments can be stored in the computer-readable storage medium and loaded and executed on the processor to facilitate the storage and application of the above methods.
[0173] The above are all preferred embodiments of this application, and are not intended to limit the scope of protection of this application. Therefore, all equivalent changes made in accordance with the structure, shape and principle of this application should be covered within the scope of protection of this application.
Claims
1. A fall prediction method, characterized in that, Includes the following steps: Obtain the physical examination data of the target personnel; Based on the physical examination data, determine whether the target person has a preset disease type; If the target person has the preset disease type, then obtain the target person's normal walking posture; If the target person's current walking posture does not conform to the normal walking posture, then obtain the dynamic walking posture trend corresponding to the current walking posture; If the dynamic walking trend meets the preset fall trend standard, a corresponding first-level fall warning signal is generated; Based on the first-level fall warning signal, a road surface information coordinate system corresponding to the current walking path of the target person is generated; Identify the road surface information coordinate system and generate the corresponding coordinates of potential dangerous locations and the straight line axis of the walking direction of the target person; If the walking direction straight line axis coincides with the coordinates of the potential danger location, then determine whether the target distance between the target person and the coordinates of the potential danger location is within the warning distance threshold; If the target distance between the target person and the coordinates of the potentially dangerous location is within the warning distance threshold, a corresponding secondary fall warning signal is generated.
2. The fall prediction method according to claim 1, characterized in that, After obtaining the target person's typical gait if the target person has the preset disease type, the following steps are also included: If the target person's current walking posture conforms to the normal walking posture, then obtain the target person's current heart rate value; Determine whether the current heart rate change value exceeds a preset heart rate change threshold; If the current heart rate change value exceeds the preset heart rate change threshold, then the preset heart rate change threshold and the current heart rate change value are combined to generate a corresponding heart rate change difference as a heart rate warning signal.
3. The fall prediction method according to claim 1, characterized in that, If the dynamic walking trend conforms to the preset fall trend standard, generating the corresponding first-level fall warning signal includes the following steps: If the dynamic walking trend matches the preset fall trend standard, then determine whether the target person has taken preset fall risk medication; If the target person takes the preset potential hazard drug, then obtain the corresponding potential hazard drug; Identify the target potentially harmful drug and obtain the corresponding associated symptoms; Based on the associated symptoms, a corresponding fall risk level is generated as the Level 1 fall warning signal.
4. The fall prediction method according to claim 1, characterized in that, After generating a corresponding Level 1 fall warning signal if the dynamic walking trend meets the preset fall trend standard, the following steps are also included: Based on the Level 1 fall warning signal, obtain the complication symptoms of the target person corresponding to the preset disease type; Identify the aforementioned symptoms and output corresponding emergency remedy prompts.
5. A fall prediction method according to claim 4, characterized in that, The process of identifying the concurrent symptoms and outputting corresponding emergency remedial prompts includes the following steps: If there are multiple complications, then obtain the target person's historical symptom records; If the concurrent symptoms exist in the historical symptom record, then obtain the concurrent occurrence number of the concurrent symptoms for the target person in the historical symptom record; Based on the number of concurrent occurrences, the output priority of the emergency remedial prompt corresponding to the concurrent symptoms is set, and the number of concurrent occurrences is proportional to the output priority; Based on the output priority, the corresponding emergency remedial prompt is output.
6. The fall prediction method according to claim 1, characterized in that, If the target distance between the target person and the coordinates of the potentially dangerous location is within the warning distance threshold, generating a corresponding secondary fall warning signal includes the following steps: If the target distance between the target person and the coordinates of the potential danger location is within the warning distance threshold, then determine whether there are multiple coordinates of the potential danger location; If there are multiple potential hazard location coordinates, the broadcast priority of the potential hazard type corresponding to each potential hazard location coordinate is set according to the target distance between each potential hazard location coordinate and the target person, and the target distance is inversely proportional to the broadcast priority of the hazard type. Based on the broadcast priority of the potential hazard type, the potential hazard type corresponding to the coordinates of the potential hazard location is broadcast, and the corresponding secondary fall warning signal is generated.
7. A fall prediction method according to claim 6, characterized in that, After broadcasting the potential hazard type corresponding to the potential hazard location coordinates according to the broadcasting priority of the potential hazard type, and generating the corresponding secondary fall warning signal, the following steps are also included: The potential hazard types are analyzed based on preset potential hazard occurrence criteria to generate corresponding fall hazard occurrence probabilities; Based on the probability of the fall, the safety hazard level corresponding to the potential hazard type is set.
8. A fall prediction system, characterized in that, include: The first acquisition module (1) is used to acquire the physical examination data of the target personnel; The first judgment module (2) is used to determine whether the target person has a preset disease type based on the physical examination data; If the target person has the preset disease type, the second acquisition module (3) is used to acquire the normal walking posture corresponding to the target person; If the current walking posture of the target person does not conform to the normal walking posture, the third acquisition module (4) is used to acquire the dynamic walking posture trend corresponding to the current walking posture. The first generation module (5) generates a corresponding first-level fall warning signal if the dynamic walking trend meets the preset fall trend standard. The second generation module (6) is used to generate a road surface information coordinate system corresponding to the current walking road of the target person based on the first-level fall warning signal; The third generation module (7) is used to identify the road information coordinate system and generate the corresponding potential danger location coordinates and the walking direction straight line axis corresponding to the target person; If the walking direction straight line axis coincides with the coordinates of the potential danger location, the second judgment module (8) is used to determine whether the target distance between the target person and the coordinates of the potential danger location is within the warning distance threshold. The fourth generation module (9) generates a corresponding secondary fall warning signal if the target distance between the target person and the coordinates of the potential danger location is within the warning distance threshold.
9. A terminal device, comprising a memory and a processor, characterized in that, The memory stores computer instructions that can run on the processor. When the processor loads and executes the computer instructions, it employs a fall prediction method as described in any one of claims 1 to 7.
10. A computer-readable storage medium storing computer instructions, characterized in that, When the computer instructions are loaded and executed by the processor, a fall prediction method as described in any one of claims 1 to 7 is employed.