CN105725991B - health monitoring system based on intelligent agent - Google Patents

health monitoring system based on intelligent agent Download PDF

Info

Publication number
CN105725991B
CN105725991B CN201610055061.7A CN201610055061A CN105725991B CN 105725991 B CN105725991 B CN 105725991B CN 201610055061 A CN201610055061 A CN 201610055061A CN 105725991 B CN105725991 B CN 105725991B
Authority
CN
China
Prior art keywords
knowledge
module
user
data
reasoning
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Expired - Fee Related
Application number
CN201610055061.7A
Other languages
Chinese (zh)
Other versions
CN105725991A (en
Inventor
顾海军
鲁珊珊
黄雅楠
田雅倩
梁辉
赵晓辉
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Jilin University
Original Assignee
Jilin University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Jilin University filed Critical Jilin University
Priority to CN201610055061.7A priority Critical patent/CN105725991B/en
Publication of CN105725991A publication Critical patent/CN105725991A/en
Application granted granted Critical
Publication of CN105725991B publication Critical patent/CN105725991B/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00Measuring for diagnostic purposes; Identification of persons
    • A61B5/02Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
    • A61B5/0205Simultaneously evaluating both cardiovascular conditions and different types of body conditions, e.g. heart and respiratory condition
    • A61B5/02055Simultaneously evaluating both cardiovascular condition and temperature
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00Measuring for diagnostic purposes; Identification of persons
    • A61B5/72Signal processing specially adapted for physiological signals or for diagnostic purposes
    • A61B5/7235Details of waveform analysis
    • A61B5/7264Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00Measuring for diagnostic purposes; Identification of persons
    • A61B5/72Signal processing specially adapted for physiological signals or for diagnostic purposes
    • A61B5/7271Specific aspects of physiological measurement analysis

Landscapes

  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Physiology (AREA)
  • Biophysics (AREA)
  • Public Health (AREA)
  • Cardiology (AREA)
  • Veterinary Medicine (AREA)
  • General Health & Medical Sciences (AREA)
  • Animal Behavior & Ethology (AREA)
  • Surgery (AREA)
  • Artificial Intelligence (AREA)
  • Pathology (AREA)
  • Biomedical Technology (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Medical Informatics (AREA)
  • Molecular Biology (AREA)
  • Signal Processing (AREA)
  • Psychiatry (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Computation (AREA)
  • Mathematical Physics (AREA)
  • Fuzzy Systems (AREA)
  • Pulmonology (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

Health monitoring system based on intelligent agent, it is related to health monitoring system, it solves existing monitor system presence and is difficult to adapt to continually changing user demand, also it is difficult to the problem of combining the comprehensive user in environment space of every field information to guard, data acquisition module is used for temperature collection sensor, pulse transducer, blood pressure sensor data, and the data transmission that acquisition is obtained is to dynamic tier module;Dynamic tier module by data collecting module collected to data be uniformly processed, and treated data as fact knowledge are sent to user agent module;Knowledge base information of the user agent module based on the received inside fact knowledge and user agent module carries out the reasoning of procedural knowledge and rule knowledge, and the reasoning results are passed through transmission of network to user.The present invention is monitored nursing to user using Multi -Agent technology.So that it is with certain universality and practicability, to preferably provide service for monitoring task.

Description

Health monitoring system based on intelligent agent
Technical field
The present invention relates to health monitoring systems, and in particular to a kind of health prison of actively perceive based on intelligent agent technology Protecting system.
Background technique
In monitoring tasks in areas (e.g., smart home, health supervision etc.), traditional method is often simple using wireless Sensor network carries out long distance monitoring, and system is difficult to adapt to continually changing user demand, it is also difficult in conjunction with every field information Comprehensive guards the user in environment space.In order to provide a new solution party to health supervision problem Case needs to construct a kind of service mode of actively perceive formula, to assist guardian to realize the long-range monitoring administration to user.
In order to provide a kind of actively perceive dynamic change service mode, must just cope with continually changing user at any time Demand and each different types of information (e.g., a plurality of types of sensor informations, information of different type user etc.).Therefore, it is The rapid variation for adapting to user intention and the fusion for realizing much information, present invention application intelligent agent technology, in each generation Reason is internal, and knowledge is characterized and made inferences in unified form, to achieve the purpose that multiple fields information mutual communication interconnects, Finally service is provided for user.
Knowledge token and reasoning are one important foundations of artificial intelligence field.Since different field problem usually has difference Attribute and feature, thus expressing for knowledge method is also not quite similar.Currently, logic, structure of the knowledge token method from knowledge Connection between knowledge is set out, and is mainly had: predicate logic method, framework method, Object-Oriented Method, semantic network etc..
It for same knowledge, can be generally indicated with a variety of methods, but since domain knowledge generally all has not Same feature, thus the effect of each representation method is also different.In existing knowledge token method, it is difficult to health Guard this dynamic change, be related to the complication system of many-sided domain knowledge and characterized.
Summary of the invention
The present invention exists for the existing monitor system of solution is difficult to adapt to continually changing user demand, it is also difficult in conjunction with each The problem of comprehensive user in environment space of realm information guards provides a kind of health prison based on intelligent agent Protecting system.
Health monitoring system based on intelligent agent, the system include data acquisition module, dynamic tier module and user Proxy module;
The data acquisition module is used for temperature collection sensor, pulse transducer, blood pressure sensor data, and will acquisition The data transmission of acquisition is to dynamic tier module;
The dynamic tier module by data collecting module collected to data be uniformly processed, and will treated number User agent module is sent to according to as fact knowledge;
The user agent module knowledge base letter inside fact knowledge and the user agent module based on the received Breath carries out the reasoning of procedural knowledge and rule knowledge, and the reasoning results are passed through transmission of network to user.
Beneficial effects of the present invention: the present invention is monitored nursing to user using Multi -Agent technology.In order to establish A kind of health supervision mode of actively perceive formula needs to establish a set of simple and practical representation of knowledge mechanism inside intelligent agent With a set of general reliable inference method so that it is with certain universality and practicability, to preferably be monitoring task Service is provided.
It is using the advantages of above-mentioned inference method:
(1) due to being added to additional process, automatic trigger for the slot of frame in the knowledge token method based on frame Slot movement is filled out in frame, accelerates the speed of knowledge reasoning.
(2) for each domain knowledge frame, reasoning is all carried out in its lower portion, rather than in the complete of unification It is carried out under rule-based knowledge base, in this way, being conducive to improve matched speed, while reducing number regular in domain knowledge base, It enhances Reasoning Efficiency and is conducive to the consistency of knowledge base.
Detailed description of the invention
Fig. 1 is the functional block diagram of the health monitoring system of the present invention based on intelligent agent;
Fig. 2 is the process schematic of knowledge reasoning in the health monitoring system of the present invention based on intelligent agent;
Fig. 3 is the original of user agent module internal reasoning in the health monitoring system of the present invention based on intelligent agent Manage configuration diagram;
Fig. 4 is the schematic diagram of communication module in the health monitoring system of the present invention based on intelligent agent;
Fig. 5 is the flow chart of knowledge reasoning in the health monitoring system of the present invention based on intelligent agent;
Fig. 6 is the flow chart of the health monitoring system of the present invention based on intelligent agent.
Specific embodiment
Specific embodiment one illustrates present embodiment in conjunction with Fig. 1 to Fig. 6, the health monitoring system based on intelligent agent, The system includes data acquisition module, dynamic tier module and user agent module;
The data acquisition module is used for temperature collection sensor, pulse transducer, blood pressure sensor data, and will acquisition The data transmission of acquisition is to dynamic tier module;The dynamic tier module by data collecting module collected to data unite One processing, and treated data as fact knowledge are sent to user agent module;The user agent module is according to connecing The fact that receipts knowledge and the user agent module inside knowledge base information carry out procedural knowledge and rule knowledge reasoning, And the reasoning results are passed through into transmission of network to user.
Specific reasoning process is the fact that inference machine is sent according to communication module knowledge in present embodiment, with knowledge Information in library is compared one by one, more next if identical as the information in knowledge base, if it is different, then updating Knowledge base information.The user agent module includes user agent, old man agency and doctor agency;According to the thing inside each agency Real knowledge, makes inferences, and realizes the wish of processing old man, guardian and doctor.
Specific embodiment two illustrates that present embodiment, present embodiment are specific embodiment one in conjunction with Fig. 1 to Fig. 6 Embodiment:
In present embodiment, based on bottom sensor equipment, using multi-agent technology, an integrated old man is realized More ageng service platforms of agency, doctor agency, monitoring agency and dynamic tier, provide respective service for service object.
Intelligent agent i.e. ageng, function can spontaneously be realized in certain circumstances by being one, and to it is related Act on behalf of the software entity being associated.Continuously, spontaneous requirement derives from the variation of environment, it is desirable that agency can be in the guidance of nobody It is responded in real time under interference by demand flexible, in a manner of intelligence to user.
The fact that described in present embodiment knowledge: the knowledge in relation to some facts, such as the classification of things, attribute, science The fact, objective fact etc..It emphasizes the static state of knowledge, that is, describes the attribute and its correlation of things.Such as, the current body temperature of old man is 37.5 degree are a fact knowledge.
Procedural knowledge: procedural knowledge be in relation to " what if " knowledge.It emphasizes the dynamic of knowledge, i.e., expression reasoning and Relevant fact etc. is searched for the process of knowledge.The solution procedure of its relevant issues, skill sex knowledge tell how to make one Thing.
Rule knowledge: usually there is causal knowledge.Such as, if current time is the morning, the behavior of old man is anticipated It is willing to trot out.
In present embodiment, it is related to all kinds of knowledge such as true, process and rule, needs to use different types of knowledge A kind of unified representation.The form of expression for using frame, fact knowledge, procedural knowledge and rule knowledge are all fused to In frame, knowledge representation method that there is satisfactory texture, complete is formed.The knowledge table based on frame is used in present embodiment Show that method, basic framework consist of three parts:
One, regular troughs (Slot): the place that frame stores information is mainly slot (Slot), and each slot again can be by several sides (facet) is formed in face.One slot indicates an attribute of object, and a side is used to describe the one aspect of respective attributes. Value possessed by slot and side is referred to as slot value and side value.In a knowledge system with frame representation, generally all contain There are multiple frames, in order to distinguish the different slots in different frame and a frame, not ipsilateral, needs to assign difference respectively Name, be referred to as frame name, slot name and facet name.
In general slot value can there are several types of types:
Occurrence, the value are given by actual conditions;
Process values, the value are a calculating process, it utilizes other slot values of the frame, carry out by given calculating process based on Obtained occurrence (such as a certain method);
Another frame name just constitutes frame calling, material is thus formed a frames when slot value is another frame name Chain, related frame aggregate, which is got up, just forms frame system.
The slot of frame can be with additional process, referred to as process attachment, including subprogram and certain reasoning process.It is common attached Process is added to have:
If-needed: once affiliated slot value will be used (inquiry), then start subsequent additional process;
If-added: once affiliated slot value will be assigned (modification), then start subsequent additional process;
Two, method slot (Methods): method slot is used to store the method in object, it is a kind of special in frame Dynamic process.The structure of method is defined by method name, method local variable and method procedure body is constituted.When method be triggered with Afterwards, required information is filtered out from the message of transmission to use for method procedure body, then executes method procedure body.
Three, regular slot (Rules): regular slot is used to store production rule collection, and each rule set is placed on as a value In regular slot.Each rule leaves several rules concentratedly.
Static attribute of frame etc. is indicated with regular troughs in present embodiment.Each regular troughs are with " facet name side value " Form is indicated, meanwhile, two processes, i.e. if-needed and if-added are added for each side value.I.e. when side value quilt When using and being assigned, corresponding method in frame is triggered respectively.Herein, additional process is option.
For regular slot, we use the form of " rule number, rule body " to be indicated;For method slot, we are used The form of " Method method name " is indicated.Wherein the definition procedure of specific method is completed outside frame, according to Method Method name, local variable define, the mode of method procedure body is expressed.
Be using the advantages of above-mentioned knowledge representation method based on frame: frame knowledge representation method uses " slot name side The unified structure of name side value ", it is concise, it can be readily appreciated that meeting the thinking habit of people.Frame knowledge representation method will be more The single knowledge representation method of kind (rule, frame and process) is fused into a frame, adequately expresses true, process and rule All kinds of static and dynamic and the knowledge such as then, compensate for the defect of single knowledge token method.
Using a unified, pervasive knowledge representation method based on frame in present embodiment, is integrated and united with this The representation of knowledge form of one different agencies.Knowledge can thus be showed with certain mode and be stored in computer It goes, but to enable a computer to infer unknown knowledge with the knowledge grasped, it is necessary to which knowledge is made inferences.
Reasoning, which refers to the process of, releases conclusion from the existing fact according to certain rule.Data are based on using a kind of herein The mode of the forward reasoning of driving.The agency from external received message is needed, adding for slot (is passed through using method call first Journey) mode fill the slot value of knowledge frame;Then, reasoning process is just carried out in lower portion, and traversal rule slot is found The rule to match, and make inferences and produce corresponding the reasoning results.Reasoning process is as shown in Figure 5.
Specific step is as follows:
(1) oneself the initial knowledge for providing user is sent into knowledge base KB (Knowledge Base);
(2) according to the known fact in KB, the additional process of slot is triggered, spontaneously carries out slot value filling;
(3) scan the Rlue keyword in the knowledge frame in corresponding field, check wherein whether have it is applicable (can be with Knowledge matches in the IF slot of Rule) knowledge otherwise turn (6) if so, then turning (4);
(4) knowledge (i.e. knowledge in the Then slot of Rule) that is applicable in all in Rule is all elected, is constituted applicable Result set RS (Result Set), turn (5);
(5) it if RS is not sky, is exported RS as final result;If RS is sky, turn (6);
(6) it asks the user whether further supplement the new fact, if so, then supplementing new fact, then turns (2);Otherwise Expression can not seek solution, unsuccessfully exit.
The communication module: it according to the form of the communication information of agreement, outwardly sends message or receives extraneous send The information come.Wherein, the structure of communication information is described as follows in conjunction with Fig. 4.
(1) message header contains system-level the characteristic information, such as the sending time of message.
(2) sender of the message: sender of the message's mark, such as system, dynamic tier, old man agency.
(3) message recipient: message recipient's mark, such as old man agency, monitoring agency, doctor agency.
(4) type of message: the mark of type of message, if the message is class signal message or informative messages.
(5) message body: the content in message body is exactly the data to be transmitted of message.According to knowledge frame described herein Frame, we arrange, and message body is transmitted according to the form of " slot name facet name side value ".
Knowledge base described in present embodiment: the knowledge and rule needed in storage knowledge processing procedure.Currently, database Technical application is very extensive, and database interface is powerful, and also very convenient for the management of database.Thus, select number Carrying out storage to the knowledge in expert system according to library has powerful advantage.It is chatted in synthesis, the storage for knowledge can use The mode of database, and can choose the commercial product of comparative maturity, such as the MySQL database of Oracle company.
It is made inferences with all kinds of knowledge in knowledge base, obtains reasonable the reasoning results.
Internal module and communication information based on agency, are illustrated the Booting sequence of whole system.The starting of system Process is illustrated in fig. 6 shown below.It is described as follows:
(1) user inputs: user is inputted by man-machine interface, and system sends disappearing for task start to each agency Breath starts whole system;
(2) request of data: user agent (old man agency/monitoring agency/doctor agency) passes through communication module to environment generation Request of data is sent in haircut, requests the data of dynamic tier;
(3) reasoning is guarded: under the scheduling of reasoning module, made inferences according to all types of knowledge in knowledge base: if Corresponding the reasoning results are obtained, then turn to result output;If not obtaining the reasoning results, continue to look for novelty to dynamic tier Data.
(4) result exports: final the reasoning results are exported.Result may finally be sent to user via mobile network Mobile phone.

Claims (2)

1. the health monitoring system based on intelligent agent, characterized in that the system includes data acquisition module, dynamic tier module And user agent module, the user agent module include inference machine and communication module;
The data acquisition module is used for temperature collection sensor, pulse transducer, blood pressure sensor data, and acquisition is obtained Data transmission to dynamic tier module;
The dynamic tier module by data collecting module collected to data be uniformly processed, and treated data are made User agent module is sent to for fact knowledge;
The user agent module based on the received the knowledge base information inside fact knowledge and the user agent module into The reasoning of row procedural knowledge and rule knowledge, and the reasoning results are passed through into transmission of network to user;
The fact that inference machine is sent according to communication module knowledge is compared one by one with the information in knowledge base, and if knowledge Information in library is identical, then more next, if it is different, then more new knowledge base information;
By the way of a kind of forward reasoning based on data-driven, from the agency of external received message, method tune is used first Mode fills the slot value of knowledge frame;Then, reasoning process is just carried out in lower portion, and traversal rule slot is found The rule to match, and make inferences and produce corresponding the reasoning results;
Specific reasoning process are as follows:
(1) oneself the initial knowledge for providing user is sent into knowledge base KB;
(2) according to the known fact in KB, the additional process of slot is triggered, spontaneously carries out slot value filling;
(3) the Rule keyword in the knowledge frame in corresponding field is scanned, checks wherein whether there is knowledge applicatory, if so, Then turn (4), otherwise turns (6);
(4) applicable knowledge all in Rule is all elected, constitutes result set RS applicatory, turned (5);
(5) it if RS is not sky, is exported RS as final result;If RS is sky, turn (6);
(6) it asks the user whether further supplement the new fact, if so, then supplementing new fact, then turns (2);Otherwise it indicates Solution can not be sought, is unsuccessfully exited.
2. the health monitoring system according to claim 1 based on intelligent agent, which is characterized in that user agent's mould Block includes user agent, old man agency and doctor agency;It according to the true knowledge inside each agency, makes inferences, realization processing The wish of old man, guardian and doctor.
CN201610055061.7A 2016-01-27 2016-01-27 health monitoring system based on intelligent agent Expired - Fee Related CN105725991B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201610055061.7A CN105725991B (en) 2016-01-27 2016-01-27 health monitoring system based on intelligent agent

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201610055061.7A CN105725991B (en) 2016-01-27 2016-01-27 health monitoring system based on intelligent agent

Publications (2)

Publication Number Publication Date
CN105725991A CN105725991A (en) 2016-07-06
CN105725991B true CN105725991B (en) 2019-01-01

Family

ID=56246641

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201610055061.7A Expired - Fee Related CN105725991B (en) 2016-01-27 2016-01-27 health monitoring system based on intelligent agent

Country Status (1)

Country Link
CN (1) CN105725991B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109885010A (en) * 2019-03-20 2019-06-14 中南大学 The health combined based on multi-Agent and Internet of Things sees maintaining method, device and storage medium

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1480884A (en) * 2003-07-16 2004-03-10 中南大学 Swarm intelligence man-machine decision method based on Internet structure
US8126731B2 (en) * 2006-10-24 2012-02-28 Medapps, Inc. Systems and methods for medical data interchange activation
CN102090930B (en) * 2009-12-10 2013-01-09 中兴保全股份有限公司 Home healthcare service management device
JP6354144B2 (en) * 2013-12-10 2018-07-11 Tdk株式会社 Electronic device, method and program

Also Published As

Publication number Publication date
CN105725991A (en) 2016-07-06

Similar Documents

Publication Publication Date Title
Lan et al. What is semantic communication? A view on conveying meaning in the era of machine intelligence
Sezer et al. Context-aware computing, learning, and big data in internet of things: a survey
JP4620721B2 (en) Method and apparatus for automatically determining semantic classification of context data
US10163420B2 (en) System, apparatus and methods for adaptive data transport and optimization of application execution
De Maio et al. A knowledge-based framework for emergency DSS
Mahmood et al. Data mining techniques for wireless sensor networks: A survey
CN106663143B (en) M2M ontology management and semantic interoperability
Wang et al. A brief review of machine learning and its application
CN103336813B (en) A kind of Internet of Things data integrated management scheme based on middleware framework
US20140324747A1 (en) Artificial continuously recombinant neural fiber network
Saleem et al. Data analytics in the Internet of Things: A survey
Cristea et al. Context-aware environments for the internet of things
CN109074802A (en) The modulation of packetizing audio signal
Wickramarathne et al. Belief theoretic methods for soft and hard data fusion
KR20200124267A (en) Semantic actions and inference support through distributed semantic data
CN108108743A (en) Abnormal user recognition methods and the device for identifying abnormal user
CN106302680A (en) A kind of data based on Internet of Things display background system
Santipantakis et al. OBDAIR: Ontology-Based Distributed framework for Accessing, Integrating and Reasoning with data in disparate data sources
Strobbe et al. Hybrid reasoning technique for improving context-aware applications
CN105725991B (en) health monitoring system based on intelligent agent
Ali A social Internet of Things application architecture: applying semantic web technologies for achieving interoperability and automation between the cyber, physical and social worlds
Wang et al. Context-aware complex event processing for event cloud in internet of things
CN114048328B (en) Knowledge-graph link prediction method and system based on conversion hypothesis and message transmission
Zhang et al. Extended dempster-shafer theory in context reasoning for ubiquitous computing environments
Akanbi et al. Semantic interoperability middleware architecture for heterogeneous environmental data sources

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20190101

Termination date: 20210127

CF01 Termination of patent right due to non-payment of annual fee