Intelligent family doctor information distributed interconnection communication method and system

Abstract

The invention provides an intelligent family doctor information distributed interconnection communication method and system, and relates to the technical field of communication, and the method comprises the steps: obtaining multiple physiological index data and position information of a plurality of monitoring personnel, and transmitting the data and the information to a background server; determining a medical demand score according to the multiple physiological index data and the registration information; determining a medical demand level according to the medical demand score; determining a first target monitoring person according to the medical demand level; and according to the position information, the medical demand score and the physiological index data of the first target monitoring personnel, generating treatment notification information, and sending the treatment notification information to a mobile device of a family doctor. According to the invention, whether each monitoring person has an emergency or needs treatment can be accurately judged, and treatment notification information can be generated based on information such as position information, so that family doctors can be helped to determine the positions and disease types of the monitoring persons needing to be rescued, and the family doctors can conveniently treat patients in time.

Description

Technical Field

[0001] The present invention relates to the field of communication technology, and in particular to an intelligent family doctor information distributed interconnection communication method and system. Background Art

[0002] In the related art, family doctors can usually be invited to diagnose and treat patients by appointment. However, this method usually cannot provide timely treatment for patients with acute diseases, resulting in poor treatment results of family doctors. Although the physiological indicators of patients can be monitored by wearable physiological indicator monitoring devices, as the number of patients increases, it is difficult for family doctors to monitor too much data at the same time, and it is also difficult to determine whether each patient has an acute disease in a timely manner, resulting in difficulty for family doctors to effectively and timely treat patients.

[0003] The information disclosed in the background technology section of this application is only intended to deepen the understanding of the general background technology of this application, and should not be regarded as an admission or any form of suggestion that the information constitutes the prior art already known to those skilled in the art. Summary of the invention

[0004] The present invention provides an intelligent family doctor information distributed interconnection communication method and system, which can solve the technical problems that related technologies are difficult to monitor too much data at the same time and difficult to timely determine whether each patient has an emergency.

[0005] According to a first aspect of the present invention, there is provided an intelligent family doctor information distributed interconnection communication method, comprising: Through the wearable physiological index monitoring device, multiple physiological index data of multiple monitoring persons are obtained at multiple moments in the current monitoring cycle; Through the positioning device of the wearable physiological index monitoring device, the location information of each monitoring person is obtained at the end of the current monitoring cycle; Send multiple physiological index data and location information of each monitored person to the backend server; Determine the medical needs score of each monitored person according to the multiple physiological index data of each monitored person and the registration information of each monitored person through the backend server; Determine the medical needs level of each monitored person according to the medical needs score of each monitored person; Determine the first target monitored person who needs treatment according to the medical need level of each monitored person; Generate rescue notification information based on the location information, medical needs score and physiological index data of the first target monitored person; The rescue notification information is sent to the family doctor's mobile device.

[0006] According to a second aspect of the present invention, there is provided an intelligent family doctor information distributed interconnected communication system, comprising: A first acquisition module is used to respectively acquire multiple physiological index data of multiple monitored persons at multiple moments in a current monitoring cycle through a wearable physiological index monitoring device; The second acquisition module is used to acquire the location information of each monitoring person at the end of the current monitoring cycle through the positioning device of the wearable physiological index monitoring device; The first sending module is used to send multiple physiological index data and location information of each monitored person to the background server; A medical needs scoring module, used to determine the medical needs score of each monitored person according to multiple physiological indicator data of each monitored person and the registration information of each monitored person through the background server; A medical needs level module is used to determine the medical needs level of each monitored person according to the medical needs score of each monitored person; A first target monitoring personnel module, used to determine a first target monitoring personnel who needs treatment according to the medical need level of each monitoring personnel; A rescue notification information module, used to generate rescue notification information according to the location information, medical needs score and physiological index data of the first target monitored person; The second sending module is used to send the rescue notification information to the mobile device of the family doctor.

[0007] Technical effect: According to the present invention, multiple physiological indicator data of multiple monitoring personnel can be collected through the background server, and the medical needs score of each monitoring personnel can be determined, so as to determine the needs of each monitoring personnel for treatment, and accurately judge whether each monitoring personnel has an emergency and needs treatment, and can generate treatment notification information based on location information and other information, so as to help family doctors determine the location and type of symptoms of the monitoring personnel who need help, so that family doctors can provide timely treatment to patients. When training the disease type neural network model, the weighted average of the average relative gap between the physiological indicator data and the physiological indicator reference value can be used as the probability that the training personnel need treatment, and the probability is combined with the medical needs annotation information, and the loss function of the disease type neural network model is obtained based on the form of the cross entropy loss function. The form of the loss function makes it only necessary to judge whether the training personnel need treatment when obtaining the medical needs annotation information, and there is no need to score the severity of the disease of the training personnel, which reduces the cost of labeling acquisition, makes the training samples of the disease type neural network model easier to obtain, and effectively improves the training intensity and accuracy of the disease type neural network model. When determining the medical needs score, the disease type weights obtained by the trained disease type neural network model can be used to determine the weighted average of the average relative gaps between various physiological indicator data and physiological indicator reference values, thereby determining the deviation between the physiological indicators that the monitoring personnel need to focus on observing and the reference values, and then determining the severity of the monitored personnel's disease, which can be used to accurately describe the monitored personnel's need for treatment. When determining the list of treatment notification sequences, the treatment efficiency of each neighbor target monitoring personnel can be determined in each round of neighbor sorting, and based on the treatment efficiency, the first target monitoring personnel ranked next in the treatment notification sequence list can be determined, thereby improving the treatment efficiency.

[0008] It should be understood that the above general description and the following detailed description are exemplary and explanatory only and do not limit the present invention. Other features and aspects of the present invention will become more apparent from the following detailed description of exemplary embodiments with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other embodiments can be obtained based on these drawings without creative work. Figure 1 A schematic diagram exemplarily shows a flow chart of an intelligent family doctor information distributed interconnection communication method according to an embodiment of the present invention; Figure 2A schematic diagram of an intelligent family doctor information distributed interconnected communication system according to an embodiment of the present invention is exemplarily shown. DETAILED DESCRIPTION

[0010] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0011] The technical solution of the present invention is described in detail with specific embodiments below. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments.

[0012] Figure 1 The following is a flow chart showing an intelligent distributed interconnected communication method for family doctor information according to an embodiment of the present invention. The method includes: Step S101, obtaining multiple physiological index data of multiple monitored persons at multiple moments in the current monitoring cycle through a wearable physiological index monitoring device; Step S102, obtaining the location information of each monitored person at the end of the current monitoring cycle through the positioning device of the wearable physiological index monitoring device; Step S103, sending multiple physiological index data and location information of each monitored person to the backend server; Step S104, determining the medical needs score of each monitored person according to the multiple physiological indicator data of each monitored person and the registration information of each monitored person through the server; Step S105, determining the medical need level of each monitored person according to the medical need score of each monitored person; Step S106, determining a first target monitored person who needs treatment according to the medical need level of each monitored person; Step S107, generating treatment notification information according to the location information, medical needs score and physiological index data of the first target monitored person; Step S108: sending the rescue notification information to the family doctor's mobile device.

[0013] According to the intelligent family doctor information distributed interconnection communication method of the embodiment of the present invention, multiple physiological indicator data of multiple monitored persons can be collected through the background server, and the medical needs score of each monitored person can be determined, thereby determining the needs of each monitored person for treatment, and accurately judging whether each monitored person has an emergency and whether treatment is required, and generating treatment notification information based on location information and other information, thereby helping family doctors determine the location and type of symptoms of monitored persons who need help, so that family doctors can provide timely treatment to patients.

[0014] According to one embodiment of the present invention, in step S101, the wearable physiological monitoring device can be worn by the monitored person, and can obtain multiple physiological indicator data of the monitored person in real time, such as blood pressure data, blood sugar data, heart rate data, blood oxygen data, etc. In the example, each monitoring cycle is one hour, and the time interval between adjacent moments is 1 minute. The physiological indicator data can be detected once a minute and sent to the background server.

[0015] According to an embodiment of the present invention, in step S102, a positioning device may be provided in the wearable physiological index detection device, and the position information of each monitored person may be determined at the end time of the current monitoring cycle.

[0016] According to one embodiment of the present invention, in step S103, multiple physiological indicator data at multiple moments obtained in the current monitoring cycle and location information obtained at the end moment of the current monitoring cycle can be sent to the background server, wherein the multiple physiological indicator data at multiple moments can be sent to the background server immediately when measured, or can be packaged and sent to the background server at the end of the current monitoring cycle, and the present invention does not impose any restrictions on this.

[0017] According to one embodiment of the present invention, in step S104, the background server may determine whether the physiological indicators of the monitored person are normal based on multiple physiological indicator data of the monitored person and the registration information of the monitored person on his own diseases. If the physiological indicators are abnormal, a higher medical needs score is obtained, which may indicate that the monitored person needs treatment.

[0018] According to one embodiment of the present invention, the medical needs score of each monitored person is determined through the background server according to multiple physiological indicator data of each monitored person and the registration information of each monitored person, including: determining the basic disease type and physiological indicator reference value of the monitored person according to the registration information of the monitored person; determining the disease type weight of the monitored person according to the basic disease type vector of the monitored person and the trained disease type neural network model; determining the medical needs score of each monitored person according to the disease type weight, the physiological indicator reference value and the physiological indicator data.

[0019] According to one embodiment of the present invention, the registration information of the monitored person can be filled in by the monitored person himself or obtained through a physical examination of the monitored person, and mainly records the type and severity of the monitored person's underlying disease. In addition, the underlying disease type vector can be determined based on the type and severity of the underlying disease. For example, if a monitored person suffers from hypertension, the first component of the underlying disease type vector is not 0, and the more severe the disease is, the larger the value of the first component is. If a monitored person suffers from coronary heart disease, the second component of the underlying disease type vector is not 0, and the more severe the disease is, the larger the value of the second component is, etc. In addition, the reference value of the physiological index can also be determined based on the underlying disease type of the monitored person. For example, if a monitored person suffers from hypertension, the reference value of his blood pressure index is 130 mmHg for systolic pressure and 90 mmHg for diastolic pressure, while the reference value of the blood pressure index of a monitored person who does not suffer from hypertension is 120 mmHg for systolic pressure and 80 mmHg for diastolic pressure.

[0020] According to one embodiment of the present invention, monitored persons suffering from different types of underlying diseases need to focus on monitoring different physiological indicators. For example, monitored persons suffering from hypertension need to focus on monitoring blood pressure indicators, and monitored persons suffering from diabetes need to focus on monitoring blood sugar indicators, etc. Therefore, reasonable weights (i.e., disease type weights) can be assigned to various physiological indicators of the monitored persons based on the underlying disease type vectors of the monitored persons, so as to use the disease type weights and physiological indicator data to comprehensively judge whether the monitored persons need treatment.

[0021] According to one embodiment of the present invention, the disease type weight can be determined by a disease type neural network model, and the training method of the disease type neural network model includes: obtaining multiple physiological indicator data and physiological indicator reference values ​​of multiple trainees at multiple times during a training cycle; obtaining basic disease type vectors of multiple trainees; processing the basic disease type vectors of the trainees through the disease type neural network model to obtain sample disease type weights of the trainees; determining the loss function of the disease type neural network model based on the sample disease type weights, multiple physiological indicator data, the physiological indicator reference values ​​and medical needs labeling information of each trainee; training the disease type neural network model based on the loss function of the disease type neural network model to obtain a trained disease type neural network model.

[0022] According to one embodiment of the present invention, when training a disease type neural network model, physiological indicator data of a trainee with an underlying disease may be used. For example, multiple physiological indicator data of the trainee may be collected at multiple times during the training cycle, and the physiological indicator reference value of each physiological indicator may be determined. Furthermore, the underlying disease type vector of the trainee may be determined based on the type and severity of the underlying disease suffered by the trainee, so that the underlying disease type vector of the trainee may be processed through the disease type neural network model to obtain the sample disease type weight of the trainee. The method for obtaining the sample disease type weight is similar to the above-mentioned disease type weight, which will not be repeated here.

[0023] According to one embodiment of the present invention, there may be certain errors in the sample disease type weights of trainable personnel. The loss function of the disease type neural network model can be determined based on the sample disease type weights, multiple physiological indicator data, physiological indicator reference values ​​and medical needs annotation information of each trainee, so as to train the disease type neural network model through the loss function to improve the accuracy of the disease type neural network model and reduce the errors of the sample disease type weights.

[0024] According to an embodiment of the present invention, the loss function of the disease type neural network model is determined according to the sample disease type weight, multiple physiological indicator data, the physiological indicator reference value and the medical demand annotation information of each trainee, including: determining the loss function LOSS of the disease type neural network model according to formula (1): (1) in, is the physiological index data of the kth physiological index of the jth trainee at the i-th moment of the training cycle, is the physiological index reference value of the kth physiological index of the jth trainee, is the sample disease type weight of the kth physiological indicator of the jth trainee, is the medical needs labeling information of the jth trainee, n is the number of moments in the training cycle, m is the number of types of physiological indicators, N is the number of trainees in a batch, i≤n, k≤m, j≤N, and i, n, k, m, j and N are all positive integers.

[0025] According to one embodiment of the present invention, in formula (1), is the relative gap between the physiological index data at a single moment and the reference value of the physiological index. is the average value of the relative gap between the physiological index data and the physiological index reference value within the training cycle, that is, the average relative gap within a training cycle. The average value of the above average relative gap is weighted by the sample disease type weight obtained by the disease type neural network model, and then averaged to obtain , that is, the weighted average of the average relative gaps between various physiological index data and physiological index reference values. The closer the weighted average is to 1, the greater the relative deviation between the measured physiological index data and the physiological index reference value, and the higher the need for treatment. Conversely, the closer the weighted average is to 0, the smaller the relative deviation between the measured physiological index data and the physiological index reference value, and the lower the need for treatment. The weighted average can be used as the probability that the jth trainee needs treatment. Based on the probability and the medical demand labeling information of the jth trainee, the loss function of the disease type neural network model can be determined.

[0026] According to one embodiment of the present invention, the medical needs labeling information of the j-th trainee may be a labeling performed by a doctor based on the physiological indicator data of the j-th trainee, and is used to describe whether the j-th trainee needs treatment. For example, when the j-th trainee needs treatment, it may be labeled as 1, otherwise, it may be labeled as 0. This labeling method may reduce the labeling difficulty for doctors, who only need to determine whether the trainee needs treatment, without having to score the severity of the trainee's disease.

[0027] According to one embodiment of the present invention, based on the probability that the jth trainee needs treatment and the medical demand labeling information of the jth trainee, the error of the sample disease type weight of the jth trainee can be obtained in the form of a cross entropy loss function, that is, Furthermore, the error of the sample disease type weights of a batch of trainees is averaged to obtain the loss function of the disease type neural network model. Based on this loss function, doctors only need to judge whether the trainees need treatment during labeling, without scoring the severity of the trainees' diseases, thereby reducing the difficulty of the standard and the cost of labeling, making it easier to obtain training samples of the disease type neural network model, effectively improving the training intensity and accuracy of the disease type neural network model, and improving the accuracy of the sample disease type weights.

[0028] According to one embodiment of the present invention, after obtaining the above-mentioned loss function, the loss function can be used for back propagation to adjust the parameters of the disease type neural network model, thereby improving the accuracy of the disease type neural network model. After multiple trainings, a trained disease type neural network model can be obtained.

[0029] In this way, the weighted average of the average relative gap between the physiological indicator data and the physiological indicator reference value can be used as the probability that the trainee needs treatment, and the probability is combined with the medical demand annotation information. Based on the form of the cross-entropy loss function, the loss function of the disease type neural network model is obtained. The form of this loss function makes it necessary to judge whether the trainee needs treatment when obtaining the medical demand annotation information, and there is no need to score the severity of the trainee's disease, which reduces the cost of labeling acquisition, makes the training samples of the disease type neural network model easier to obtain, and effectively improves the training intensity and accuracy of the disease type neural network model.

[0030] According to one embodiment of the present invention, a trained disease type neural network model can be used to determine the disease type weight of the monitored person based on the basic disease type vector of the monitored person, and the disease type weight can be used to obtain a medical needs score, thereby accurately describing the monitored person's need for treatment.

[0031] According to an embodiment of the present invention, the medical need score of each monitored person is determined according to the disease type weight, the physiological index reference value and the physiological index data, including: determining the medical need score of the xth monitored person according to formula (2): , (2) in, is the physiological index data of the yth physiological index of the xth monitored person at the ith moment of the current monitoring cycle, is the physiological index reference value of the yth physiological index of the xth monitored person, is the disease type weight of the yth physiological indicator of the xth monitored person, n is the number of moments in the current monitoring cycle, m is the number of types of physiological indicators, i≤n, y≤m, and i, x, y, m and n are all positive integers.

[0032] According to one embodiment of the present invention, is the relative gap between the physiological index data at a single moment and the reference value of the physiological index. It is the average value of the relative gap between the physiological indicator data and the physiological indicator reference value within the monitoring period, that is, the average relative gap within a monitoring period. The average value of the above average relative gap is weighted by the disease type weight obtained by the trained disease type neural network model, and then averaged to obtain the weighted average value of the average relative gap between various physiological indicator data and physiological indicator reference values, that is, the medical demand score of the xth monitored person, which can be used to indicate the probability that the xth monitored person needs treatment, that is, the monitored person's need for treatment.

[0033] In this way, the disease type weights obtained by the trained disease type neural network model can be used to determine the weighted average of the average relative gaps between various physiological indicator data and physiological indicator reference values, thereby determining the deviations between the physiological indicators that the monitoring personnel need to focus on observing and the reference values, and then determining the severity of the monitored personnel's disease, which can be used to accurately describe the monitored personnel's need for treatment.

[0034] According to one embodiment of the present invention, in step S105, the medical needs level of each monitored person can be determined based on the medical needs score. For example, multiple value intervals can be divided. If the medical needs score of a monitored person belongs to [0,0.3), it is considered that the medical needs level of the monitored person is low, that is, no medical treatment is required. If the medical needs score of a monitored person belongs to [0.3,0.5), it is considered that the medical needs level of the monitored person is intermediate, and a medication reminder can be sent to the monitored person. If the medical needs score of a monitored person belongs to [0.5,0.8), it is considered that the medical needs level of the monitored person is relatively high, and a medical advice can be sent to the monitored person. If the medical needs score of a monitored person belongs to [0.8,+∞), it is considered that the medical needs level of the monitored person is high, and a family doctor can be dispatched to the location of the monitored person for treatment.

[0035] According to one embodiment of the present invention, in step S106, the first target monitored person who needs treatment may be determined based on the medical need level of each monitored person. For example, a monitored person with a high medical need level may be determined as the first target monitored person.

[0036] According to one embodiment of the present invention, in step S107, treatment notification information can be generated based on the location information, medical needs score and physiological indicator data of the first target monitored person, so as to notify the family doctor to go to the location of the first target monitored person for treatment, and in the treatment notification information, the treatment order of each first target monitored person can be determined based on the location information and medical needs score of the first target monitored person, thereby improving the family doctor's treatment efficiency for multiple first target monitored persons.

[0037] According to one embodiment of the present invention, based on the location information, medical needs score and physiological indicator data of the first target monitoring person, rescue notification information is generated, including: arranging the first target monitoring persons according to the medical needs score of each first target monitoring person to obtain a first list; obtaining the first geographical block range of the first target monitoring person ranked first in the first list according to the location information of the first target monitoring person ranked first in the first list; in the first list, based on the location information, screening the first target monitoring persons within the first geographical block range as the neighboring target monitoring persons of the first target monitoring person ranked first in the first list; determining the medical needs score of each neighboring target monitoring person according to the location information of the neighboring target monitoring persons, the medical needs score of the neighboring target monitoring persons, and the medical needs score of the first target monitoring person ranked first in the first list The neighbor ranking scores and the rescue notification sequence list of the monitored personnel are determined; according to the rescue notification sequence list, the first target monitored personnel ranked No. 1 in the first list and the location information and physiological index data of each neighbor target monitored personnel, the rescue notification information of each first target monitored personnel in the rescue notification sequence list is generated; the first target monitored personnel ranked No. 1 and the neighbor target monitored personnel are deleted from the first list to obtain an updated first list; iterative execution determines the first target monitored personnel ranked No. 1 and his neighbor target monitored personnel in the first list, determines the neighbor ranking scores of the neighbor target monitored personnel and determines the rescue notification sequence list, as well as determines the rescue notification information of each first target monitored personnel in the rescue notification sequence list, deletes the first target monitored personnel ranked No. 1 and his neighbor target monitored personnel, and obtains the updated first list, until the first target monitored personnel no longer exists in the first list.

[0038] According to one embodiment of the present invention, the first target monitoring persons may be first sorted according to the medical need scores to obtain a first list, that is, in the first list, the first target monitoring persons are sorted in descending order of the medical need scores.

[0039] According to one embodiment of the present invention, the first target monitored person ranked first can be selected from the first list, that is, the first target monitored person with the highest medical needs score and the one who is most in need of treatment, and the location information of the first target monitored person can be determined. Furthermore, in order to improve the efficiency of family doctors in treating multiple first target monitored persons, after treating the first target monitored person with the highest medical needs score, the first target monitored persons near him can be treated, so as to achieve the purpose of treating more first target monitored persons in the shortest time and with the least moving distance. After the first target monitored person with the highest medical needs score and the first target monitored persons near him are treated, other first target monitored persons are selected from the first list for treatment.

[0040] According to one embodiment of the present invention, after finding the first target monitoring person ranked first in the first list, the first geographical block range where the first target monitoring person is located can be determined, for example, the geographical range of the community or cell where the first target monitoring person is located, and the first target monitoring persons whose location information is within the geographical block range can be found from the first list. These first target monitoring persons within the same geographical block range are the first target monitoring persons near the first target monitoring person ranked first in the first list, that is, the neighboring target monitoring persons.

[0041] According to one embodiment of the present invention, if there are multiple neighboring target monitoring personnel, the neighboring target monitoring personnel can be ranked to determine the order of treatment of the neighboring target monitoring personnel after the first target monitoring personnel ranked first is treated.

[0042] According to one embodiment of the present invention, the neighbor ranking score of each neighbor target monitoring person is determined based on the location information of the neighbor target monitoring person, the medical need score of the neighbor target monitoring person, and the medical need score of the first target monitoring person ranked first in the first list, including: determining the neighbor ranking score of the zth neighbor target monitoring person in the first round of neighbor sorting according to formula (3): , (3) in, is the location information of the first target monitoring person ranked first in the rescue notification sequence list, and the first target monitoring person ranked first in the rescue notification sequence list is the first target monitoring person ranked first in the first list, When sorting the first round of neighbors, the medical needs score of the zth neighbor target monitoring person is, The location information of the zth nearest neighbor target monitoring personnel in the first round of nearest neighbor sorting; The neighbor target monitoring person with the highest neighbor ranking score in the first round of neighbor ranking is determined as the first target monitoring person ranked second in the rescue notification sequence list; According to formula (4), when determining the neighbor ranking score of the zth neighbor target monitoring personnel in the xth round of neighbor ranking, , (4) in, is the medical needs score of the zth neighbor target monitoring person in the xth round of neighbor sorting, is the location information of the zth neighbor target monitoring personnel in the xth round of neighbor sorting, The location information of the first target monitoring personnel ranked x in the rescue notification sequence list, wherein, in the xth round of neighbor sorting, the neighbor target monitoring personnel participating in the sorting are other neighbor target monitoring personnel except the first x first target monitoring personnel in the rescue notification sequence list; The neighbor target monitoring person with the highest neighbor sorting score during the xth round of neighbor sorting is determined as the first target monitoring person ranked xth in the rescue notification sequence list.

[0043] According to one embodiment of the present invention, in formula (3), It can represent the distance between the zth nearest target monitoring person and the first target monitoring person ranked first in the first list, so, It represents the treatment efficiency of the zth neighbor target monitoring personnel in the first round of neighbor sorting, that is, the treatment needs that can be solved by the family doctor per unit distance moved, which can be used as the neighbor sorting score of the zth neighbor target monitoring personnel in the first round of neighbor sorting. Therefore, among multiple neighbor target monitoring personnel, the neighbor target monitoring personnel with the highest treatment efficiency can be determined as the first target monitoring personnel ranked second in the treatment notification order list. Similarly, after determining the first target monitoring person ranked x in the treatment notification sequence list, the xth round of neighbor sorting can be performed, and the neighbor sorting scores of the neighbor target monitoring persons participating in the sorting can be determined according to formula (4), that is, after the treatment of the first target monitoring person ranked x in the treatment notification sequence list is completed, the medical demand scores of the remaining neighbor target monitoring persons are determined by the ratio of the distance between the remaining neighbor target monitoring persons and the first target monitoring person ranked x in the treatment notification sequence list, and the treatment efficiency of continuing to treat other neighbor target monitoring persons after the treatment of the first target monitoring person ranked x in the treatment notification sequence list is completed, so as to select the neighbor target monitoring person with the highest treatment efficiency as the first target monitoring person ranked x+1 in the treatment notification sequence list. In the above manner, the first target monitoring person ranked 1 in the first list and its neighbor target monitoring persons can be sorted into the treatment notification sequence list.

[0044] In this way, the rescue efficiency of each neighbor target monitoring personnel can be determined in each round of neighbor sorting, and based on the rescue efficiency, the first target monitoring personnel ranked next in the rescue notification order list can be determined, thereby improving the rescue efficiency.

[0045] According to one embodiment of the present invention, based on the rescue notification sequence list, the first target monitoring person ranked first in the first list, and the location information and physiological indicator data of each neighboring target monitoring person, rescue notification information of each first target monitoring person in the rescue notification sequence list can be generated and sent to the family doctor's mobile device in step S108. That is, the first target monitoring person ranked first in the first list and each neighboring target monitoring person are sorted in the order of the notification sequence list, and the location information and physiological indicator data of each first target monitoring person are attached to the sorted list, so that the family doctor can be notified by sending it to the family doctor's mobile device to go to each location information in order, and treat each first target monitoring person with reference to the physiological indicator data.

[0046] According to one embodiment of the present invention, after the first target monitoring person ranked No. 1 and each neighboring target monitoring person in the above-mentioned first list have been treated, the first target monitoring person ranked No. 1 and the neighboring target monitoring persons may be deleted from the first list to obtain an updated first list. In the updated first list, the first target monitoring person ranked No. 1 may be selected, and the neighboring target monitoring persons may be searched based on the geographic block range where the person is located, and the neighbor ranking score and the treatment notification sequence list may be determined in a manner similar to the above, thereby generating treatment notification information for multiple first target monitoring persons within the geographic block range, and sending the information to the family doctor's mobile device in step S108. The above process may be iteratively performed until the treatment of all first target monitoring persons is completed.

[0047] According to the intelligent family doctor information distributed interconnection communication method of the embodiment of the present invention, multiple physiological index data of multiple monitoring personnel can be collected through the background server, and the medical needs score of each monitoring personnel can be determined, so as to determine the needs of each monitoring personnel for treatment, and it can accurately judge whether each monitoring personnel has an emergency and whether they need treatment, and can generate treatment notification information based on location information and other information, so as to help family doctors determine the location and type of symptoms of the monitoring personnel who need help, so that family doctors can provide timely treatment to patients. When training the disease type neural network model, the weighted average of the average relative gap between the physiological index data and the physiological index reference value can be used as the probability that the training personnel need treatment, and the probability is combined with the medical needs annotation information, and the loss function of the disease type neural network model is obtained based on the form of the cross entropy loss function. The form of the loss function makes it only necessary to judge whether the training personnel need treatment when obtaining the medical needs annotation information, and there is no need to score the severity of the disease of the training personnel to reduce the annotation acquisition cost, so that the training samples of the disease type neural network model are easier to obtain, and the training intensity and accuracy of the disease type neural network model are effectively improved. When determining the medical needs score, the disease type weights obtained by the trained disease type neural network model can be used to determine the weighted average of the average relative gaps between various physiological indicator data and physiological indicator reference values, thereby determining the deviation between the physiological indicators that the monitoring personnel need to focus on observing and the reference values, and then determining the severity of the monitored personnel's disease, which can be used to accurately describe the monitored personnel's need for treatment. When determining the list of treatment notification sequences, the treatment efficiency of each neighbor target monitoring personnel can be determined in each round of neighbor sorting, and based on the treatment efficiency, the first target monitoring personnel ranked next in the treatment notification sequence list can be determined, thereby improving the treatment efficiency.

[0048] Figure 2 A schematic diagram of an intelligent family doctor information distributed interconnected communication system according to an embodiment of the present invention is exemplarily shown, and the system includes: A first acquisition module is used to respectively acquire multiple physiological index data of multiple monitored persons at multiple moments in a current monitoring cycle through a wearable physiological index monitoring device; The second acquisition module is used to acquire the location information of each monitoring person at the end of the current monitoring cycle through the positioning device of the wearable physiological index monitoring device; The first sending module is used to send multiple physiological index data and location information of each monitored person to the background server; A medical needs scoring module, used to determine the medical needs score of each monitored person according to multiple physiological indicator data of each monitored person and the registration information of each monitored person through the background server; A medical needs level module is used to determine the medical needs level of each monitored person according to the medical needs score of each monitored person; A first target monitoring personnel module, used to determine a first target monitoring personnel who needs treatment according to the medical need level of each monitoring personnel; A rescue notification information module, used to generate rescue notification information according to the location information, medical needs score and physiological index data of the first target monitored person; The second sending module is used to send the rescue notification information to the mobile device of the family doctor.

[0049] It should be understood by those skilled in the art that the embodiments of the present invention described above and shown in the accompanying drawings are only examples and do not limit the present invention. The purpose of the present invention has been fully and effectively achieved. The functional and structural principles of the present invention have been demonstrated and explained in the embodiments, and the embodiments of the present invention may be deformed or modified in any way without departing from the principles.

[0050] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. An intelligent family doctor information distributed interconnection communication method, characterized in that: include: Through the wearable physiological index monitoring device, multiple physiological index data of multiple monitoring persons are obtained at multiple moments in the current monitoring cycle; Through the positioning device of the wearable physiological index monitoring device, the location information of each monitoring person is obtained at the end of the current monitoring cycle; Send multiple physiological index data and location information of each monitored person to the backend server; Determine the medical needs score of each monitored person according to the multiple physiological index data of each monitored person and the registration information of each monitored person through the backend server; Determine the medical needs level of each monitored person according to the medical needs score of each monitored person; Determine the first target monitored person who needs treatment according to the medical need level of each monitored person; Generate rescue notification information based on the location information, medical needs score and physiological index data of the first target monitored person; The rescue notification information is sent to the family doctor's mobile device.

2. The intelligent family doctor information distributed interconnection communication method according to claim 1 is characterized in that: The backend server determines the medical needs score of each monitored person according to multiple physiological index data of each monitored person and the registration information of each monitored person, including: Determine the basic disease type vector and physiological index reference value of the monitored person according to the registered information of the monitored person; Determine the disease type weight of the monitored person according to the basic disease type vector of the monitored person and the trained disease type neural network model; The medical needs score of each monitored person is determined according to the disease type weight, the physiological indicator reference value and the physiological indicator data.

3. The intelligent family doctor information distributed interconnection communication method according to claim 2 is characterized in that: The training methods of the disease type neural network model include: Obtain multiple physiological index data and physiological index reference values ​​of multiple trainees at multiple times during a training cycle; Obtain basic disease type vectors of multiple trainees; The basic disease type vector of the trainee is processed through the disease type neural network model to obtain the sample disease type weight of the trainee; Determine the loss function of the disease type neural network model according to the sample disease type weight, multiple physiological indicator data, the physiological indicator reference value and the medical needs annotation information of each trainee; According to the loss function of the disease type neural network model, the disease type neural network model is trained to obtain a trained disease type neural network model.

4. The intelligent family doctor information distributed interconnection communication method according to claim 3 is characterized in that: According to the sample disease type weight, multiple physiological index data, the physiological index reference value and the medical needs labeling information of each training personnel, the loss function of the disease type neural network model is determined, including: According to the formula ； Determine the loss function LOSS of the disease type neural network model, where is the physiological index data of the kth physiological index of the jth trainee at the i-th moment of the training cycle, is the physiological index reference value of the kth physiological index of the jth trainee, is the sample disease type weight of the kth physiological indicator of the jth trainee, is the medical needs labeling information of the jth trainee, n is the number of moments in the training cycle, m is the number of types of physiological indicators, N is the number of trainees in a batch, i≤n, k≤m, j≤N, and i, n, k, m, j and N are all positive integers.

5. The intelligent family doctor information distributed interconnection communication method according to claim 2 is characterized in that: Determine the medical needs score of each monitored person according to the disease type weight, the physiological index reference value and the physiological index data, including: According to the formula ； Determine the medical needs score for the xth monitored person ,in, is the physiological index data of the yth physiological index of the xth monitored person at the ith moment of the current monitoring cycle, is the physiological index reference value of the yth physiological index of the xth monitored person, is the disease type weight of the yth physiological indicator of the xth monitored person, n is the number of moments in the current monitoring cycle, m is the number of types of physiological indicators, i≤n, y≤m, and i, x, y, m and n are all positive integers.

6. The intelligent family doctor information distributed interconnection communication method according to claim 1 is characterized in that: Generate rescue notification information based on the location information, medical needs score and physiological index data of the first target monitored person, including: Arrange the first target monitoring persons according to the medical needs score of each first target monitoring person to obtain a first list; According to the location information of the first target monitoring person ranked first in the first list, obtaining the first geographical block range where the first target monitoring person ranked first in the first list is located; In the first list, according to the location information, a first target monitoring person within the first geographical block is selected as a neighboring target monitoring person of the first target monitoring person ranked first in the first list; Determine the neighbor ranking score and treatment notification order list of each neighbor target monitoring person according to the location information of the neighbor target monitoring person, the medical need score of the neighbor target monitoring person, and the medical need score of the first target monitoring person ranked first in the first list; Generate rescue notification information for each first target monitoring person in the rescue notification sequence list according to the rescue notification sequence list, the first target monitoring person ranked first in the first list, and the location information and physiological index data of each neighboring target monitoring person; Deleting the first target monitoring person ranked first and the neighboring target monitoring persons in the first list to obtain an updated first list; Iterative execution determines the first target monitoring person ranked first and his neighboring target monitoring persons in the first list, determines the neighbor ranking scores of the neighboring target monitoring persons and determines the rescue notification sequence list, and determines the rescue notification information of each first target monitoring person in the rescue notification sequence list, and deletes the first target monitoring person ranked first and his neighboring target monitoring persons to obtain the updated first list until the first target monitoring person no longer exists in the first list.

7. The intelligent family doctor information distributed interconnection communication method according to claim 6 is characterized in that: Determining the neighbor ranking score of each neighbor target monitoring person according to the location information of the neighbor target monitoring person, the medical need score of the neighbor target monitoring person, and the medical need score of the first target monitoring person ranked first in the first list, including: According to the formula ； Determine the neighbor ranking score of the zth neighbor target monitoring personnel in the first round of neighbor ranking ,in, is the location information of the first target monitoring person ranked first in the rescue notification sequence list, and the first target monitoring person ranked first in the rescue notification sequence list is the first target monitoring person ranked first in the first list, When sorting the first round of neighbors, the medical needs score of the zth neighbor target monitoring person is, The location information of the zth nearest neighbor target monitoring personnel in the first round of nearest neighbor sorting; The neighbor target monitoring person with the highest neighbor ranking score in the first round of neighbor ranking is determined as the first target monitoring person ranked second in the rescue notification sequence list; According to the formula ； Determine the neighbor ranking score of the zth neighbor target monitoring personnel when the xth round of neighbor ranking is determined ,in, is the medical needs score of the zth neighbor target monitoring person in the xth round of neighbor sorting, is the location information of the zth neighbor target monitoring personnel in the xth round of neighbor sorting, The location information of the first target monitoring personnel ranked x in the rescue notification sequence list, wherein, in the xth round of neighbor sorting, the neighbor target monitoring personnel participating in the sorting are other neighbor target monitoring personnel except the first x first target monitoring personnel in the rescue notification sequence list; The neighbor target monitoring person with the highest neighbor sorting score in the xth round of neighbor sorting is determined as the first target monitoring person ranked at position x+1 in the rescue notification order list.

8. An intelligent family doctor information distributed interconnected communication system for executing the method according to any one of claims 1 to 7, characterized in that: include: A first acquisition module is used to respectively acquire multiple physiological index data of multiple monitored persons at multiple moments in a current monitoring cycle through a wearable physiological index monitoring device; The second acquisition module is used to acquire the location information of each monitoring person at the end of the current monitoring cycle through the positioning device of the wearable physiological index monitoring device; The first sending module is used to send multiple physiological index data and location information of each monitored person to the background server; A medical needs scoring module, used to determine the medical needs score of each monitored person according to multiple physiological indicator data of each monitored person and the registration information of each monitored person through the background server; A medical needs level module is used to determine the medical needs level of each monitored person according to the medical needs score of each monitored person; A first target monitoring personnel module, used to determine a first target monitoring personnel who needs treatment according to the medical need level of each monitoring personnel; A rescue notification information module, used to generate rescue notification information according to the location information, medical needs score and physiological index data of the first target monitored person; The second sending module is used to send the rescue notification information to the mobile device of the family doctor.

Images