Summary of the invention
The problem to be solved in the present invention be design it is a kind of personal from surveying method for establishing model and from surveying system based on disease collectionSystem provides related, accurate personal survey model certainly from various dimensions for sufferer, provides the foundation of science in terms of from survey for sufferer.
In order to achieve the above object, the technical scheme adopted by the invention is as follows: it is a kind of personal from surveying model based on disease collectionMethod for building up includes the following steps:
(1) medical information ontology (Medical Ontology) database, i.e. MO database are established；
(2) there is most frequent symptom in search in MO database, splits data into the disease collection containing the symptom and notDisease collection containing the symptom；
(3) by with the interface that exchanges of sufferer, obtain sufferer whether have the information of the symptom, if so, thening follow the steps(4), if not, thening follow the steps (5)；
(4) symptom is recorded in the personal symptom group of sufferer, and matches the disease collection for containing the symptom as current diseaseDisease collection；Judge whether the personal symptom group of sufferer meets the disease that present illness concentrates all symptom of some disease or 80% or moreSign, if so, (7) are thened follow the steps, if not, thening follow the steps (6)；
(5) disease collection of the matching without containing the symptom is as present illness collection；
(6) it is concentrated in present illness, searches in addition to the symptom for including in personal symptom group, most frequent symptom occur,Present illness collection data are divided into the disease collection containing the symptom and the disease collection without containing the symptom；And return step (3)；
(7) it is sent from the result for surveying model by the interface that exchanges with sufferer using the disease as personal.
Preferably, in the step (1), the method for creation MO database are as follows:
A. disease and its feature are numbered with MO terms, each MO terms represents a vertex, two termsBetween relationship indicated with directed edge, disease and symptom are indicated in a directed acyclic graph in this way；
B. the association between vertex is divided into two types: is_a relationship and part_of relationship；Is_a relationship is a kind of simpleInclusion relation；Part_of relationship indicates the inclusion relation of a part, and often there are many symptom to show for a kind of disease, symptom andIt is the relationship of part_of between disease, is the relationship of is_a between disease and disease, is the pass of is_a between symptom and symptomSystem；
C. for the directed edge between any two terms, if the relationship of part_of, then assign weight；Weight associationProbability d is indicated；Degree of being associated d distributes (0 < d≤1) between the sub- terms that father term is associated；Wherein degree of association d(term1, term2) indicates that the probability of the sub- symptom of term2 occurs in father's symptom term1；
D. for the directed edge between any two terms, if the relationship of is_a, then assign weight；Weight association hundredDivide than indicating；The percentage term all sub- terms associated by father term are associated between father term and sub- termsIn there is ratio, the sum of association percentage of the sub- terms is 1.
Further, after the step (7) further include:
(8), sufferer chooses whether that doctor is needed to check on after result provides, be select the corresponding doctor of the symptom, andReward points, then the result can be sent to letter inquiry in doctor station；Doctor, which gives, to feed back；Disease outcome judges phase with doctorThe buddy list of doctor is added in sufferer by symbol automatically, carries out online exchange；Disease outcome is not inconsistent with doctor's judgement, provides reasonIt explains and directly replys sufferer in a manner of believing in standing, reserve hospital's facial diagnosis；Whether sufferer adopts the judgement of doctor with the side of weightingFormula adjusts MO database.
Further, step (8) is additionally provided with integral mechanism, selects to have selected this while the symptom corresponding doctorThe corresponding integral of doctor；Sufferer associated quad is deducted, sufferer adopts physician feedback, gives doctor's integral, does not adopt, do not giveDoctor's integral.
Further, after the step (7) further include:
(8) sufferer does not select doctor to check on, and the hospital for directly arriving system recommendation carries out facial diagnosis, then gives system results and commentsValence, and reward on total mark is obtained, the evaluation of sufferer adjusts MO database with weighting scheme.
Another aspect of the present invention additionally provides a kind of personal self-measuring system based on disease collection, which is characterized in that packetIt includes:
Medical information ontology (Medical Ontology) database, i.e. MO database；
First search module, there is most frequent symptom in search in MO database, splits data into containing the symptomDisease collection and disease collection without containing the symptom；
Exchange interface module, by with the interface that exchanges of sufferer, obtain whether sufferer has the information of the symptom, if so, thenThe first matching and judgment module are gone to, if not, going to the second matching module；
First matching and judgment module, the symptom are recorded in the personal symptom group of sufferer, and are matched and contained the symptomDisease collection is as present illness collection；Judge whether the personal symptom group of sufferer meets present illness and concentrate all symptom of some diseaseOr 80% or more symptom, if so, (7) are thened follow the steps, if not, going to the second search module；
Second matching module, disease collection of the matching without containing the symptom is as present illness collection；
Second search module is concentrated in present illness, is searched in addition to the symptom for including in personal symptom group, most frequency occursPresent illness collection data are divided into the disease collection containing the symptom and the disease collection without containing the symptom by numerous symptom；And it goes toFirst matching and judgment module；
As a result sending module is sent from the result for surveying model by exchange interface module using the disease as personal.
Preferably, the foundation of medical information ontology (Medical Ontology) database includes:
A. disease and its feature are numbered with MO terms, each MO terms represents a vertex, two termsBetween relationship indicated with directed edge, disease and symptom are indicated in a directed acyclic graph in this way；
B. the association between vertex is divided into two types: is_a relationship and part_of relationship；Is_a relationship is a kind of simpleInclusion relation；Part_of relationship indicates the inclusion relation of a part, and often there are many symptom to show for a kind of disease, symptom andIt is the relationship of part_of between disease, is the relationship of is_a between disease and disease, is the pass of is_a between symptom and symptomSystem；
C. for the directed edge between any two terms, if the relationship of part_of, then assign weight；Weight associationProbability d is indicated；Degree of being associated d distributes (0 < d≤1) between the sub- terms that father term is associated；Wherein degree of association d(term1, term2) indicates that the probability of the sub- symptom of term2 occurs in father's symptom term1；
D. for the directed edge between any two terms, if the relationship of is_a, then assign weight；Weight association hundredDivide than indicating；The percentage term all sub- terms associated by father term are associated between father term and sub- termsIn there is ratio, the sum of association percentage of the sub- terms is 1.
Further, further includes:
Doctor checks on module, and sufferer chooses whether that doctor is needed to check on after result provides, and is to select the symptom correspondingDoctor and reward points, then the result can be sent to the doctor station in letter inquiry；Doctor, which gives, to feed back；Disease outcome withDoctor's judgement is consistent, and sufferer is added to the buddy list of doctor automatically, carries out online exchange；Disease outcome and doctor judge notSymbol provides explanation and directly replys sufferer in a manner of believing in standing, reserves hospital's facial diagnosis, whether sufferer adopts sentencing for doctorIt is disconnected that MO database is adjusted with weighting scheme.
Further, doctor's module of checking on is additionally provided with integral unit, selects to select while the symptom corresponding doctorThe corresponding integral of the doctor is selected；Sufferer associated quad is deducted, sufferer adopts physician feedback, gives doctor's integral, does not adopt thenDoctor's integral is not given.
Further, further includes:
Facial diagnosis evaluation module, sufferer do not select doctor to check on, and the hospital for directly arriving system recommendation carries out facial diagnosis, then giveSystem results evaluation, and reward on total mark is obtained, the evaluation of sufferer adjusts MO database with weighting scheme.
The invention has the benefit that by means of the present invention, individual of the foundation based on disease collection is from survey model and certainlyExamining system provides related, accurate personal survey mould certainly from various dimensions such as database information, physician feedback, hospital's facial diagnosis for suffererType and result verification method provide the foundation of science for sufferer in terms of from survey.
Specific embodiment
The present invention will be further described combined with specific embodiments below.
Method of the invention is as follows:
(1) medical information ontology (Medical Ontology) database, i.e. MO database are established；Method for building up are as follows: willDisease and its feature are numbered with MO terms, and each MO terms represents a vertex, and the relationship between two terms is usedDirected edge indicates, in this way indicates disease and symptom in a directed acyclic graph；
Association between vertex is divided into two types: is_a relationship and part_of relationship；Is_a relationship is a kind of simpleInclusion relation；Part_of relationship indicates the inclusion relation of a part, and often there are many symptom to show for a kind of disease, symptom and diseaseIt is the relationship of part_of between disease, is the relationship of is_a between disease and disease, is the relationship of is_a between symptom and symptom；For the directed edge between any two terms, if the relationship of part_of, then assign weight；Weight is indicated with association probability d；Degree of being associated d distributes (0 < d≤1) between the sub- terms that father term is associated；Wherein degree of association d (term1,Term2) indicate that the probability of the sub- symptom of term2 occurs in father's symptom term1；
For the directed edge between any two terms, if the relationship of is_a, then assign weight；Weight association percentageThan indicating；Percentage is associated with the sub- term in all sub- terms associated by father term between father term and sub- termsThere is ratio, the sum of association percentage of the sub- terms is 1.
(2) there is most frequent symptom in search in MO database, splits data into the disease collection containing the symptom and notDisease collection containing the symptom；
(3) by with the interface that exchanges of sufferer, obtain sufferer whether have the information of the symptom, if so, thening follow the steps(4), if not, thening follow the steps (5)；
(4) symptom is recorded in the personal symptom group of sufferer, and matches the disease collection for containing the symptom as current diseaseDisease collection；Judge whether the personal symptom group of sufferer meets the disease that present illness concentrates all symptom of some disease or 80% or moreSign, if so, (7) are thened follow the steps, if not, thening follow the steps (6)；
(5) disease collection of the matching without containing the symptom is as present illness collection；
(6) it is concentrated in present illness, searches in addition to the symptom for including in personal symptom group, most frequent symptom occur,Present illness collection data are divided into the disease collection containing the symptom and the disease collection without containing the symptom；And return step (3)；
(7) it is sent from the result for surveying model by the interface that exchanges with sufferer using the disease as personal.
According to the above method, increase question-and-answer problem module is seen a doctor from foundation intelligence on platform is surveyed in disease:
Have a fever? --- -- is, concentrates statistics to obtain into all disease collection with fever symptoms, and in the diseaseThere is most symptoms and inquiry；
----no concentrates statistics to show that appearance is most into all disease collection without fever symptoms, and in the diseaseSymptom and inquiry；
And so on, the symptom until meeting all symptoms of some disease or 80% or more gives disease outcome；
Doctor can also be added in this module and participate in link, patient can choose whether to need doctor after final result providesCheck on (after each result) --- it is that patient selects doctor and reward points in the same disease circle to be inquired, then the patient selectsSelecting symptomatic consequence can be sent in doctor station in letter and deduct associated quad.
Doctor, which gives, to feed back；
--- it is that disease outcome is consistent with doctor's judgement, buddy list is added in the patient automatically, can be handed over onlineStream.
--- no, disease outcome is not inconsistent with doctor's judgement, provides explanation and directly replys trouble in a manner of believing in standingPerson reserves hospital's facial diagnosis.Patient, which adopts, gives integral, does not adopt and does not give integral.
Patient can not also select doctor to check on, and the hospital for directly arriving system recommendation gives system results and comment after carrying out facial diagnosisValence simultaneously obtains reward on total mark.
Evaluation of patient and the judgement for whether adopting doctor can make association of the system to symptom and disease all with weighting schemeProperty carry out coupling learning again, until more and more accurate.
The above is only a specific embodiment of the present invention, is not intended to limit the scope of protection of the present invention, it is allWithin the spirit and principles in the present invention, any modification, equivalent substitution, improvement and etc. done should be included in protection of the inventionWithin the scope of.