Elderly Healthcare Support System Based on Multimodal Data Fusion Analysis

Abstract

The present invention belongs to the technical field of health care management. A geriatric health care support system based on multi-modal data fusion analysis is provided, including: a composite receiving port, which is used to identify the type of health care scenario of the monitored elderly person, determine the associated object according to the type of health care scenario; receive the first multi-modal data related to the monitored elderly person and the second multi-modal data related to the associated object; an analysis and processing module, which is used to analyze the first multi-modal data to obtain the first health care demand information; and analyze the second multi-modal data to obtain the second health care demand information related to the first health care demand information; determine a number of matching health care support items according to the first health care demand information and the second health care demand information; a health care support module, which is used to send each health care support item to the health care executor, and the health care executor provides health care support for the monitored elderly person and the associated object on site. The present invention analyzes the health care needs of the elderly based on multi-modal data, with higher accuracy.

Description

Technical Field

[0001] The present invention relates to the technical field of health care management, and more particularly, to an elderly health care support system based on multi-modal data fusion analysis. Background Art

[0002] At present, the management of the health care of the elderly is an important part of improving the living standards of the elderly. For example, analyzing the health care needs of the elderly and then providing relevant health care support. In the prior art, the management of the elderly health care is mainly achieved through a video monitoring system, that is, by shooting the activity videos of the elderly through the video monitoring system, performing image recognition on the activity videos to analyze the health care needs of the elderly, and executing corresponding response strategies to provide health care support for them.

[0003] However, simply analyzing the health care needs of the elderly based on the activity videos captured by the video monitoring system, the accuracy of the analysis results is difficult to be effectively guaranteed, resulting in an easy situation of response errors, which needs to be improved. Summary of the Invention

[0004] In view of the above technical problems, the present invention provides an elderly health care support system, an electronic device, a computer storage medium, and a computer program product based on multi-modal data fusion analysis.

[0005] The present invention discloses an elderly health care support system based on multi-modal data fusion analysis. The elderly health care support system includes a composite receiving port, an analysis and processing module, and a health care support module. Among them, the composite receiving port is used to identify the type of the health care scenario of the monitored elderly, determine the associated object according to the type of the health care scenario, receive the first multi-modal data related to the monitored elderly, and the second multi-modal data related to the associated object. The first multi-modal data and the second multi-modal data both at least include video data, sensing data of wearable devices, and audio data. The analysis and processing module is used to analyze the first health care demand information according to the first multi-modal data, and analyze the second health care demand information related to the first health care demand information according to the second multi-modal data. Determine a number of matching health care support items according to the first health care demand information and the second health care demand information, and send each of the health care support items to the health care support module. The health care support module is used to send each of the health care support items to the health care executor after receiving each of the health care support items, and the health care executor provides health care support for the monitored elderly and the associated object on site.

[0006] As an alternative embodiment, identifying the type of the elderly care scenario of the monitored elderly person and determining the associated object according to the type of the elderly care scenario includes: identifying the type of the elderly care scenario of the monitored elderly person, where the type of the elderly care scenario includes a nursing home scenario and a home scenario; if the type of the elderly care scenario is the nursing home scenario, obtaining the elderly persons in the corresponding nursing home who have a close relationship with the monitored elderly person and whose physical health conditions meet the preset conditions, and taking them as the associated objects; if the type of the elderly care scenario is the home scenario, taking all the persons in the home scenario whose physical health conditions meet the preset conditions as the associated objects; where the physical health conditions meeting the preset conditions means that the incidence probabilities of several specified disease types are higher than the preset probability.

[0007] As an alternative embodiment, obtaining the elderly persons in the corresponding nursing home who have a close relationship with the monitored elderly person and whose physical health conditions meet the preset conditions, and taking them as the associated objects includes: obtaining multiple groups of surveillance videos in the corresponding nursing home, obtaining multiple elderly persons who have accompanied behavior with the monitored elderly person according to each of the surveillance videos, respectively counting the number of times each elderly person has accompanied, and determining several elderly persons with the top-ranked accompanying times as candidate elderly persons; obtaining the medical record information of each of the candidate elderly persons, evaluating the incidence probabilities of several of the specified disease types of each of the candidate elderly persons according to the medical record information, and if any of the incidence probabilities is higher than the preset probability, determining the candidate elderly person as a target elderly person, and taking each target elderly person as the associated object.

[0008] As an alternative embodiment, analyzing the first multi-modal data to obtain the first elderly care demand information includes: analyzing each single-modal data in the first multi-modal data to obtain the third elderly care demand information, and performing a fusion process on each of the third elderly care demand information to obtain the first elderly care demand information.

[0009] As an alternative embodiment, performing a fusion process on each of the third elderly care demand information to obtain the first elderly care demand information includes: determining the weighted weights corresponding to each of the single-modal data, and performing a fusion process on each of the third elderly care demand information according to the weighted weights to obtain the first elderly care demand information.

[0010] As an alternative embodiment, the second elderly care demand information related to the first elderly care demand information obtained according to the second multimodal data analysis includes: obtaining the preliminary emotional stability degree of the associated object according to the analysis of each unimodal data in the second multimodal data, and determining an adjustment value according to the number of disease-causing events experienced by the associated object, and using the adjustment value to adjust the preliminary emotional stability degree to a target emotional stability degree; analyzing the severity level of the consequences of the first elderly care demand information, and inputting the severity level, the target emotional stability degree, and the medical record information of the associated object into an association analysis model, and the association analysis model outputs the second elderly care demand information related to the first elderly care demand information.

[0011] As an alternative embodiment, the sending of each elderly care support project to the elderly care executor includes: retrieving the handling preplans corresponding to each elderly care support project, and sending each handling preplan to the elderly care executor.

[0012] The present invention also discloses an electronic device applied to the elderly care support system based on multimodal data fusion analysis; it includes: at least one processor, a memory, and a computer program stored in the memory and executable on the at least one processor.

[0013] The present invention also discloses a computer storage medium applied to the elderly care support system based on multimodal data fusion analysis; the computer-readable storage medium stores a computer program.

[0014] The present invention also discloses a computer program product applied to the elderly care support system based on multimodal data fusion analysis; the computer program product contains computer code that can be executed by the processor of the electronic device.

[0015] The beneficial effects of the present invention are as follows: 1) The present invention analyzes the elderly care demand of the monitored elderly based on multimodal data of the monitored elderly. The multimodal data includes at least video data, sensing data of wearable devices, and audio data. In this way, the accuracy of the analyzed elderly care demand is higher.

[0016] 2) The present invention also analyzes the multimodal data of the associated object related to the monitored elderly, and analyzes the elderly care demand generated due to the elderly care demand of the monitored elderly, and further provides elderly care support for the associated object. That is, the monitored elderly and the associated object are regarded as a combination, and the elderly care demand is analyzed as a whole, which can further improve the accuracy of elderly care support. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] To more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for use in the embodiments. It should be understood that the following drawings only show certain embodiments of the present invention, and thus should not be regarded as limiting the scope. For those of ordinary skill in the art, without creative efforts, other related drawings can also be obtained based on these drawings.

[0018] Figure 1 It is a schematic structural diagram of an elderly care support system based on multimodal data fusion analysis disclosed in an embodiment of the present invention.

[0019] Figure 2 It is a schematic flowchart of determining associated objects in a pension institution disclosed in an embodiment of the present invention.

[0020] Figure 3 It is a schematic flowchart of determining second elderly care demand information disclosed in an embodiment of the present invention. Detailed implementation manners

[0021] The following specific embodiments illustrate the implementation manners of the present application. Those familiar with this technology can easily understand other advantages and effects of the present application from the content disclosed in this specification. Obviously, the described embodiments are part of the embodiments of the present application, rather than all of them. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope protected by the present application.

[0022] In addition, the technical features involved in different implementation manners of the present application described below can be combined with each other as long as they do not conflict with each other.

[0023] In the prior art, the management of elderly care is mainly achieved through a video monitoring system, that is, by shooting the activity videos of the elderly through the video monitoring system, performing image recognition on the activity videos to analyze the elderly care needs of the elderly, and executing corresponding response strategies to provide elderly care support for them. However, simply analyzing the elderly care needs of the elderly based on the activity videos taken by the video monitoring system, the accuracy of the analysis results is difficult to be effectively guaranteed, resulting in an easy situation of response errors, which needs to be improved.

[0024] Regarding the above technical problems, as Figure 1As shown in the figure, an embodiment of the present invention discloses an elderly care support system based on multi-modal data fusion analysis. The elderly care support system includes a composite receiving port, an analysis and processing module, and a care support module. Among them, the composite receiving port is used to identify the type of care scenario of the monitored elderly person, determine the associated object according to the type of care scenario, receive the first multi-modal data related to the monitored elderly person, and the second multi-modal data related to the associated object. Both the first multi-modal data and the second multi-modal data at least include video data, sensing data of wearable devices, and audio data. The analysis and processing module is used to analyze the first care demand information based on the first multi-modal data, and analyze the second care demand information related to the first care demand information based on the second multi-modal data. Determine a number of matching care support items according to the first care demand information and the second care demand information, and send each care support item to the care support module. The care support module is used to send each care support item to the care executor after receiving each care support item, and the care executor provides care support for the monitored elderly person and the associated object on-site.

[0025] Compared with the elderly care support system based solely on the video surveillance system mentioned in the background art, there are at least two improvements in the present invention: 1) The present invention analyzes the care needs of the monitored elderly person based on multi-modal data, and the multi-modal data at least includes video data, sensing data of wearable devices, and audio data, so the accuracy of the analyzed care needs is higher. 2) The present invention also analyzes the multi-modal data of the associated object related to the monitored elderly person, and analyzes the care needs generated due to the care needs of the monitored elderly person, and then provides care support for the associated object. That is, the monitored elderly person and the associated object are regarded as a combination, and their care needs are analyzed as a whole, which can further improve the accuracy of care support.

[0026] The above-mentioned composite receiving port supports multiple communication protocols, so that it can communicate with video surveillance systems, wearable devices, radio equipment, etc., to obtain video data, sensing data of wearable devices, and audio data respectively.

[0027] As an alternative embodiment, identifying the type of the elderly care scenario of the monitored elderly person and determining the associated object according to the type of the elderly care scenario includes: identifying the type of the elderly care scenario of the monitored elderly person, where the type of the elderly care scenario includes an elderly care institution scenario and a home scenario; if the type of the elderly care scenario is the elderly care institution scenario, obtaining the elderly person in the corresponding elderly care institution who has a close relationship with the monitored elderly person and whose physical health condition meets the preset conditions, and taking him / her as the associated object; if the type of the elderly care scenario is the home scenario, taking all the persons whose physical health condition meets the preset conditions in the home scenario as the associated object; where the physical health condition meeting the preset conditions means that the incidence probability of several specified disease types is higher than the preset probability.

[0028] In this embodiment, the monitored elderly person can choose to receive elderly care in an elderly care institution or at home. Correspondingly, the types of the elderly care scenario include the elderly care institution scenario and the home scenario. The type of the elderly care scenario where the monitored elderly person is located can be realized by arranging a video monitoring system in the corresponding residence, that is, performing image analysis on the monitoring video captured by the video monitoring system to perform semantic analysis on the included scenario type, and then determining the corresponding type of the elderly care scenario.

[0029] The improvement point 2) of the present invention lies in providing elderly care support for both the monitored elderly person and the associated object related to the monitored elderly person at the same time. Among them, the elderly care needs of the monitored elderly person may indirectly cause the associated object to also have elderly care needs. For example, when the monitored elderly person suddenly suffers from a severe disease, the associated object may trigger related severe diseases (mainly cardiovascular and cerebrovascular diseases) due to tension and worry. By predicting the diseases that the associated object may be induced, corresponding elderly care support can be provided in a timely manner, that is, preparatory processing is performed.

[0030] The associated objects in different types of elderly care scenarios are identified in different ways. In the elderly care institution scenario, the elderly person in the elderly care institution who has a close relationship with the monitored elderly person and whose physical health condition meets the preset conditions is taken as the associated object; in the home scenario, almost all family members will be affected by the diseases of the monitored elderly person, so all the persons whose physical health condition meets the preset conditions are taken as the associated objects. The reason for such setting in the present invention is that the persons who are closely related to the monitored elderly person are more likely to be affected by the sudden severe diseases of the monitored elderly person, mainly psychologically, and thus are more likely to induce themselves to have diseases such as cardiovascular and cerebrovascular diseases. The elderly persons in the elderly care institution who have close contact with the monitored elderly person (such as living in the same room, often participating in activities together, etc.) and whose physical health is poor, and the family members whose physical health is poor all belong to the above-mentioned associated objects.

[0031] Among them, several types of diseases can be pre-specified, mainly cardiovascular and cerebrovascular diseases, because these diseases are more easily affected by emotions and cause onset, and the probability of occurrence of the above types of diseases can be evaluated based on the archival information of each person. If the probability of occurrence is higher than the preset probability, it is determined that the physical health condition meets the preset conditions.

[0032] As an optional embodiment, Figure 2 As shown, the method of obtaining the elderly people in the corresponding nursing home who have a close relationship with the monitored elderly person and whose physical health conditions meet the preset conditions and taking them as the associated objects includes: obtaining multiple groups of surveillance videos in the corresponding nursing home, extracting multiple elderly people who have companionship behavior with the monitored elderly person based on each of the surveillance videos, counting the number of companionship behaviors of each elderly person, and determining several elderly people with the highest number of companionship behaviors as candidate elderly people; obtaining the medical record information of each of the candidate elderly people, and evaluating the incidence probability of several of the specified disease types of the candidate elderly people based on the medical record information, if any of the incidence probabilities is higher than the preset probability, the candidate elderly people are determined as target elderly people, and each target elderly person is taken as the associated object.

[0033] In this embodiment, firstly, multiple groups of monitoring videos in a certain period of time in the nursing home are obtained, from which it can be analyzed which elderly people often go out with the monitored elderly people, such as walking together, playing chess, etc., and the elderly people with the highest number of companionship in a certain period (for example, the first 5) are determined as candidate elderly people. At the same time, the medical record information of these candidate elderly people is also obtained, and the medical record information records the type of disease and the degree of the disease (evaluated by the doctor) of each candidate elderly person. Based on these medical record information, the probability of each candidate elderly person being induced to develop various types of cardiovascular and cerebrovascular diseases can be evaluated, wherein the probability of the candidate elderly person who does not involve the specified disease type in the medical record information is 0, and the probability of the candidate elderly person who involves the specified disease type in the medical record information is greater than 0, and the probability of the disease is positively correlated with the number and degree of the specified disease type involved. Finally, if the probability of the onset of any cardiovascular and cerebrovascular disease of the candidate elderly person is higher than the preset probability, the candidate elderly person is determined as a target elderly person, and each target elderly person is an associated object.

[0034] As an optional embodiment, obtaining the first health care needs information based on the analysis of the first multimodal data includes: obtaining the third health care needs information based on the analysis of each single modal data in the first multimodal data, and fusing each of the third health care needs information to obtain the first health care needs information.

[0035] In this embodiment, after receiving the first multimodal data of the monitored elderly person, the analysis and processing module parses out each unimodal data therefrom, and analyzes the elderly care needs of the monitored elderly person based on each unimodal data respectively, that is, multiple pieces of third elderly care demand information are obtained. Then, the multiple pieces of third elderly care demand information are integrated and processed to obtain the first elderly care demand information of the monitored elderly person.

[0036] For example: The monitored elderly person falls to the ground and cannot get up. The third elderly care demand information obtained based on the video data is [heart attack, probability: 75%], [hypoglycemia, probability: 40%]. The third elderly care demand information obtained based on the sensing data of the wearable device is [heart attack, probability: 90%]. The third elderly care demand information obtained based on the audio data is [heart attack, probability: 50%], [hypoglycemia, probability: 50%]. The weighted average value of the probability values of "heart attack" is (65% + 80% + 50%) / 3 = 71.7%, and the weighted average value of the probability value of hypoglycemia is 45%. Therefore, it is determined that "heart attack" is the most likely first elderly care demand information.

[0037] As an optional embodiment, the integrating and processing the third elderly care demand information to obtain the first elderly care demand information includes: determining the weighted weights corresponding to each unimodal data, and integrating and processing each third elderly care demand information according to the weighted weights to obtain the first elderly care demand information.

[0038] In this embodiment, the weighted weights corresponding to each unimodal data are also preset in advance. When specifically setting, it is necessary to consider the recognition accuracy of different unimodal data for elderly care needs. For example, based on the sensing data of the wearable device to identify heart attacks, its accuracy rate is the highest, and based on video data to identify situations such as bumps and falls, its accuracy rate is the highest, and based on audio data to identify the falls of the monitored elderly person in the blind area of the monitor, its accuracy rate is the highest. Therefore, the weighted weights corresponding to each unimodal data are set according to different elderly care needs.

[0039] Then, the weighted average value of each piece of third elderly care demand information obtained is calculated according to the above weighted weights, so as to obtain the first elderly care demand information.

[0040] As an optional embodiment, as Figure 3As shown, the second health care demand information related to the first health care demand information obtained from the second multimodal data analysis includes: obtaining the preliminary emotional stability degree of the associated object according to the analysis of each unimodal data in the second multimodal data, and determining an adjustment value according to the number of disease-causing events experienced by the associated object, and using the adjustment value to adjust the preliminary emotional stability degree to the target emotional stability degree; analyzing the severity level of the consequences of the first health care demand information, and inputting the severity level, the target emotional stability degree, and the medical record information of the associated object into an association analysis model, and the association analysis model outputs the second health care demand information related to the first health care demand information.

[0041] In this embodiment, first, the preliminary emotional stability degree of the associated object is comprehensively analyzed according to each unimodal data in the second multimodal data of the associated object. For example, the preliminary emotional stability degree is analyzed by identifying facial expressions, heart rate, body temperature, speech pitch / tone, etc. At the same time, an adjustment value is also determined according to the number of disease-causing events experienced by the associated object. The more disease-causing events the associated object has experienced, the lower the degree of emotional fluctuation affected by such events, and vice versa. Multiplying the adjustment value by the preliminary emotional stability degree gives the target emotional stability degree.

[0042] Next, analyze the severity level of the consequences of the first health care demand information, that is, the severity of the consequences that the disease type of the monitored elderly may cause, such as disability, death, etc. The higher the severity level of the consequences, the more likely it is to cause a large emotional fluctuation of the associated object, thereby increasing the probability of inducing cardiovascular and cerebrovascular diseases in the associated object.

[0043] Finally, input the severity level of the consequences, the target emotional stability degree, and the medical record information of the associated object into the association analysis model together, and the model predicts the second health care demand information related to the first health care demand information, that is, the disease type that the associated object is most likely to develop due to the disease type of the monitored elderly. The association analysis model is preferably constructed based on Transformer and mainly includes an encoder and a decoder, where the encoder can be a dual-encoder structure. In addition, the association analysis model can also be constructed by a large model (such as the BERT large model), which will not be elaborated here.

[0044] As an alternative embodiment, the sending each of the health care support items to the health care executor includes: retrieving the disposal plans corresponding to each of the health care support items, and sending each of the disposal plans to the health care executor.

[0045] In this embodiment, disposal plans for different disease types are preset in the database, and the disposal plans include the drugs and devices to be used, as well as the specific standard disposal processes.

[0046] An embodiment of the present invention also discloses an electronic device, which is applied to the elderly care support system based on multimodal data fusion analysis; including: at least one processor, a memory, and a computer program stored in the memory and executable on the at least one processor.

[0047] An embodiment of the present invention also discloses a computer storage medium, which is applied to the elderly care support system based on multimodal data fusion analysis; the computer-readable storage medium stores a computer program.

[0048] An embodiment of the present invention also discloses a computer program product, which is applied to the elderly care support system based on multimodal data fusion analysis; the computer program product contains computer code, and the computer code can be executed by the processor of the electronic device.

[0049] The above-mentioned computer-readable storage medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the above. Alternatively, the computer-readable storage medium may be a machine-readable signal medium. More specific examples of the machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0050] In order to provide interaction with the user, the systems and techniques described herein may be implemented on an electronic device that has: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and a pointing device (e.g., a mouse or a trackball), through which the user can provide input to the electronic device. Other kinds of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and the input from the user may be received in any form (including voice input, speech input, or tactile input).

[0051] It should be understood that the various forms of processes shown above can be used, with steps reordered, added or deleted. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and no limitation is made herein.

[0052] The above specific embodiments do not constitute a limitation to the protection scope of the present invention. Those skilled in the art should understand that various modifications, combinations, sub - combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. An elderly care support system based on multimodal data fusion analysis, characterized in that: The elderly care support system includes a composite receiving port, an analysis and processing module, and a care support module; among them, The composite receiving port is used to identify the type of elderly care scenario of the monitored elderly, determine the associated object according to the type of elderly care scenario; receive the first multimodal data related to the monitored elderly and the second multimodal data related to the associated object; among them, both the first multimodal data and the second multimodal data at least include video data, sensing data of wearable devices, and audio data; The analysis and processing module is used to analyze the first care demand information based on the first multimodal data; and analyze the second care demand information related to the first care demand information based on the second multimodal data; determine a number of matching care support items according to the first care demand information and the second care demand information, and send each of the care support items to the care support module; The care support module is used to send each of the care support items to the care executors after receiving each of the care support items, and the care executors provide care support for the monitored elderly and the associated object on-site; Identifying the type of elderly care scenario of the monitored elderly and determining the associated object according to the type of elderly care scenario includes: Identifying the type of elderly care scenario of the monitored elderly, and the type of elderly care scenario includes a nursing home scenario and a home scenario; If the type of elderly care scenario is the nursing home scenario, obtain the elderly in the corresponding nursing home who have a close relationship with the monitored elderly and whose physical health conditions meet the preset conditions, and use them as the associated object; If the type of elderly care scenario is the home scenario, use all the people in the home scenario whose physical health conditions meet the preset conditions as the associated object; Among them, the physical health conditions meeting the preset conditions refer to the incidence probability of various cardiovascular and cerebrovascular diseases being higher than the preset probability; Obtaining the elderly in the corresponding nursing home who have a close relationship with the monitored elderly and whose physical health conditions meet the preset conditions and using them as the associated object includes: Obtain multiple groups of monitoring videos in the corresponding nursing home, obtain multiple elderly people who have accompanied behavior with the monitored elderly according to each of the monitoring videos, respectively count the number of times each elderly person has accompanied, and determine several elderly people with the top number of accompanying times as candidate elderly people; Obtain the medical record information of each of the candidate elderly people, evaluate the incidence probability of various cardiovascular and cerebrovascular diseases of the candidate elderly people according to the medical record information, if any of the incidence probabilities is higher than the preset probability, determine the candidate elderly person as the target elderly person, and use each target elderly person as the associated object.

2. The elderly care support system based on multimodal data fusion analysis according to claim 1, wherein: Analyzing the first care demand information based on the first multimodal data includes: Analyze the third care demand information based on each single-modal data in the first multimodal data, and perform a fusion process on each of the third care demand information to obtain the first care demand information.

3. The elderly care support system based on multimodal data fusion analysis according to claim 2, characterized in that: Performing a fusion process on each of the third care demand information to obtain the first care demand information includes: Determine the weighted weights corresponding to each of the single-modal data, and fuse and process each of the third elderly care demand information according to the weighted weights to obtain the first elderly care demand information.

4. The elderly care support system based on multimodal data fusion analysis according to claim 3, characterized in that: Derive second elderly care demand information related to the first elderly care demand information based on the second multi-modal data analysis, including: Derive the preliminary emotional stability level of the associated object based on the single-modal data in the second multi-modal data, and determine an adjustment value based on the number of disease-causing events experienced by the associated object, and use the adjustment value to adjust the preliminary emotional stability level to the target emotional stability level; Analyze the severity level of the consequences of the first elderly care demand information, and input the severity level, the target emotional stability level, and the medical record information of the associated object into an association analysis model, and the association analysis model outputs the second elderly care demand information related to the first elderly care demand information.

5. The geriatric healthcare support system based on multimodal data fusion analysis according to claim 4, wherein: Send each of the elderly care support items to the elderly care executors, including: Retrieve the disposal plans corresponding to each of the elderly care support items, and send each of the disposal plans to the elderly care executors.

6. An electronic device, characterized in that: Applied to the elderly care support system based on multi-modal data fusion analysis according to any one of claims 1-5; including: at least one processor, a memory, and a computer program stored in the memory and executable on the at least one processor.

7. A computer storage medium, characterized in that: Applied to the elderly care support system based on multi-modal data fusion analysis according to any one of claims 1-5; the computer storage medium stores a computer program.

8. A computer program product, characterized in that: Applied to the elderly care support system based on multi-modal data fusion analysis according to any one of claims 1-5; the computer program product contains computer code that can be executed by a processor of an electronic device.

Images