CN108665975A - Clinical path matching process and system - Google Patents
Clinical path matching process and system Download PDFInfo
- Publication number
- CN108665975A CN108665975A CN201710205898.XA CN201710205898A CN108665975A CN 108665975 A CN108665975 A CN 108665975A CN 201710205898 A CN201710205898 A CN 201710205898A CN 108665975 A CN108665975 A CN 108665975A
- Authority
- CN
- China
- Prior art keywords
- clinical
- clinical path
- clinical picture
- diagnosis
- path
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Measuring And Recording Apparatus For Diagnosis (AREA)
Abstract
The present invention provides a kind of clinical path matching process, includes the following steps：Obtain patient information；According to the patient information of acquisition, structure clinical picture vector；It is concentrated in the clinical picture built in advance, k nearest neighbors of the clinical picture vector, wherein k >=1 is calculated and found by k arest neighbors sorting algorithms；And by the element in the diagnosis and treatment active set built in advance corresponding to the k recently neighbors, the diagnosis and treatment activity corresponding to the clinical picture vector is obtained, to form clinical path.Clinical path matching process according to the present invention can accurately generate clinical path.The present invention also provides a kind of clinical path matching systems.
Description
Technical field
The present invention relates to medical information technical fields, and in particular to a kind of clinical path matching process and system.
Background technology
Clinical path is to be proposed by the U.S. nineties in last century, for the diagnosis and treatment standard of single illness quality and cost management
Change pattern.Clinical path refers to the standardized workflow finished writing in advance, is cured according to evidence-based by the professional of each subject
Principle reaches certain disease or the critical therapies of operation, inspection, the sequence of nursing activity and the arrangement of time as far as possible
To optimization so that most of patients for suffering from this disease or implementing this operation flow can be medically treated according to this by being admitted to hospital to discharge
And nursing.
Currently, China is wideling popularize the implementation of clinical path.But the medical act being related to due to medical path
More with diagnosis and treatment campaign item, patient sees a doctor, and the conditions occurred are complex, and the coupling between more illnesss is stronger, symptom
The case where occurring in progradation is more, it is difficult to form the diagnosis and treatment standard mode of single illness quality and cost management.If
It is stiff to indiscriminately imitate fixed clinical path, Relapse rate may be caused, therapeutic process is stiff, is susceptible to and delays treatment, excessively doctor
Situations such as treatment.
Although in addition, country defend planning commission publication clinical path guidance standard have universal directive significance, due to
The otherness and disease itself of hospital in all parts of the country have the characteristics that personalization in different patients so that applying clinical path
When, it is necessary to it is combined with the concrete condition of area, hospital.
Therefore, how to provide it is a kind of accurate, flexibly, be conveniently adjusted and effective clinical path matching process and be
System, becomes urgent problem.
Invention content
A kind of clinical path matching process of present invention offer and system, can accurately and rapidly generate clinical path.
An aspect of of the present present invention is related to a kind of clinical path matching process, includes the following steps：
Obtain patient information；
According to the patient information of acquisition, structure clinical picture vector；
Concentrated in the clinical picture that builds in advance, calculated by k arest neighbors sorting algorithms and find the clinical picture to
K nearest neighbors of amount, wherein k >=1；And
By the element in the diagnosis and treatment active set built in advance corresponding to the k recently neighbors, face described in acquisition
Diagnosis and treatment activity corresponding to bed phenomenon vector, to form clinical path.
As a result, in clinical path matching process according to the present invention, k arest neighbors (knn, k-nearest
Neighbor) this data digging method of sorting algorithm is applied in the matching process of clinical path, so as to according to patient
Information accurately determines diagnosis and treatment activity, it is thus possible to accurately obtain clinical path
Preferably, clinical path matching process according to the present invention is further comprising the steps of：According to the patient obtained in real time
Information corrects the clinical path.
As a result, clinical path matching process according to the present invention can according to the concrete condition of patient, to clinical path into
The rational dynamic self-adapting of row, so that clinical path can be more flexible, more accurately adaptable with patient symptom.
Preferably, the patient information that the basis obtains in real time, the step of correcting the clinical path include：Patient's
In therapeutic process, if therapeutic scheme effect is not notable or new burst clinical picture occurs, need to reacquire patient
Information, to build new clinical picture vector；And concentrated in the clinical picture built in advance, classified by k arest neighbors and is calculated
Method calculates and finds k nearest neighbors of the clinical picture vector, to correct the clinical path, wherein k >=1.
As a result, according to the concrete condition of patient, clinical path can obtain reasonable, accurate and quantization dynamic self-adapting tune
It is whole, so that clinical path can be more flexible, more accurately adaptable with patient symptom.
Preferably, clinical path matching process according to the present invention is further comprising the steps of：It is fed back according to clinic, corrects institute
State clinical picture collection and/or the diagnosis and treatment active set.
As a result, clinical path matching process according to the present invention can closer to the current situation of medicine, avoid generate with
The clinical path that the illness of patient is not consistent.
Preferably, described to be concentrated in the clinical picture built in advance, it is calculated and is found described by k arest neighbors sorting algorithms
The step of k of clinical picture vector nearest neighbors includes：The clinical picture vector is calculated to concentrate with the clinical picture
Each element distance, and the result for calculating distance is arranged from small to large, the element of k distance minimum before taking, to obtain
K nearest neighbors of the clinical picture vector.
The number that the determination process of clinical path can be passed through science by clinical path matching process according to the present invention as a result,
Further quantified according to method for digging, so as to more accurately determine clinical path.
Another aspect of the present invention relates to a kind of clinical path matching system, including patient information acquisition module, clinic are existing
As vector structure module, nearest neighbor algorithm execution module and clinical path generation module, wherein
The patient information acquisition module is for obtaining patient information；
The clinical picture vector structure module is used for the patient information according to acquisition, structure clinical picture vector；
The nearest neighbor algorithm execution module is used to concentrate in the clinical picture, is calculated simultaneously by k arest neighbors sorting algorithms
Find k nearest neighbors of the clinical picture vector, wherein k >=1；And
The clinical path generation module is used for through the member in the diagnosis and treatment active set corresponding to the k recently neighbors
Element obtains the diagnosis and treatment activity corresponding to the clinical picture vector, to form clinical path.
As a result, in clinical path matching system according to the present invention, this data mining side of k arest neighbors sorting algorithms
Method is applied in the matching process of clinical path, so as to accurately determine diagnosis and treatment activity according to patient information, it is thus possible to
Accurately obtain clinical path.
Preferably, clinical path matching system according to the present invention further includes clinical path correcting module, the clinic road
Diameter correcting module is used to, according to the patient information obtained in real time, correct the clinical path.
As a result, clinical path matching system according to the present invention can according to the concrete condition of patient, to clinical path into
The rational dynamic self-adapting of row, so that clinical path can be more flexible, more accurately adaptable with patient symptom.
Preferably, when therapeutic scheme effect is not notable or new burst clinical picture occurs, the clinical path is repaiied
Positive module can start the patient information acquisition module to reacquire patient information, and can start successively described clinical existing
As vector structure module, the nearest neighbor algorithm execution module and the clinical path generation module, to correct the clinical path.
As a result, according to the concrete condition of patient, clinical path can obtain reasonable, accurate and quantization dynamic self-adapting tune
It is whole, so that clinical path can be more flexible, more accurately adaptable with patient symptom.
Preferably, clinical path matching system according to the present invention further includes that clinical picture collection and diagnosis and treatment active set correct mould
Block, the clinical picture collection and diagnosis and treatment active set correcting module be used for according to clinic feed back, correct the clinical picture collection and/or
The diagnosis and treatment active set.
As a result, clinical path matching process according to the present invention can closer to the current situation of medicine, avoid generate with
The clinical path that the illness of patient is not consistent.
Preferably, the nearest neighbor algorithm execution module is concentrated for calculating the clinical picture vector with the clinical picture
Each element distance, and the result for calculating distance is arranged from small to large, the element of k distance minimum before taking, to obtain
K nearest neighbors of the clinical picture vector.
The number that the determination process of clinical path can be passed through science by clinical path matching process according to the present invention as a result,
Further quantified according to method for digging, so as to more accurately determine clinical path.
Compared with the prior art, in clinical path matching process according to the present invention, according to the patient information of acquisition,
By this data digging method of k arest neighbors sorting algorithm, constructed clinical picture vector passes through the meter with clinical picture collection
It calculates, so as to obtain required diagnosis and treatment activity, to form clinical path, so as to be accurately determined according to patient information
Diagnosis and treatment activity, it is thus possible to accurately obtain clinical path.
Description of the drawings
To describe the technical solutions in the embodiments of the present invention more clearly, make required in being described below to embodiment
Attached drawing is briefly described, it should be apparent that, drawings in the following description are only some embodiments of the invention, for
For those skilled in the art, without creative efforts, it can also be obtained according to these attached drawings other attached
Figure.
Fig. 1 is the flow diagram of the clinical path matching process of first embodiment according to the present invention；
Fig. 2 is the flow diagram of clinical path matching process according to the second embodiment of the present invention；
Fig. 3 is the flow diagram of clinical path matching process according to the third embodiment of the invention；
Fig. 4 is the functional block diagram of the clinical path matching system of first embodiment according to the present invention；
Fig. 5 is the functional block diagram of the clinical path matching system of second embodiment according to the present invention；
Fig. 6 is the functional block diagram of the clinical path matching system of 3rd embodiment according to the present invention.
Specific implementation mode
In order to make the purpose , technical scheme and advantage of the present invention be clearer, with reference to the accompanying drawings and embodiments, right
The present invention is further elaborated.It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, and
It is not used in the restriction present invention.
Referring to Fig.1, the clinical path matching process of first embodiment according to the present invention is shown, the clinical path match party
Method includes the following steps：S1：Obtain patient information；S2：According to the patient information of acquisition, structure clinical picture vector；S3：Pre-
The clinical picture first built is concentrated, and k arest neighbors of the clinical picture vector is calculated and found by k arest neighbors sorting algorithms
Element, wherein k >=1；And S4：Pass through the member in the diagnosis and treatment active set built in advance corresponding to the k recently neighbors
Element obtains the diagnosis and treatment activity corresponding to the clinical picture vector, to form clinical path.
The clinical matching process of first embodiment according to the present invention is specifically described below in conjunction with Fig. 1, but not
As limitation.
In the clinical matching process of first embodiment according to the present invention, clinical picture collection and diagnosis and treatment active set usually by
The clinical path of medical institutions instructs evaluation group to organize and builds in advance, and wherein clinical picture collection is by clinical picture vector structure
At set.
During clinic diagnosis, doctor to the interrogation of patient and by medical apparatus and instruments detect about people
The summation of the physical and chemical index of body, referred to as clinical picture.Specifically, clinical picture vector p is a series of combination of indexs, i.e. p=
{e1,e2,...,en, wherein eiReferred to as i-th of index, these usual indexs all have stronger intuitive meaning.These parameters packet
The physical and chemical index detected by Medical Devices is included, such as：Total cholesterol, total bilirubin, uric acid, granulocyte percentage and blood
The quantizating index such as platelet, and the semantic phenomena that are obtained by doctor's interrogation, such as dizziness, nausea, diarrhea number etc..These refer to
Mark can be Boolean quantity, for example whether there is phenomena such as dizzy, nausea, or discrete magnitude, such as diarrhea number etc., or
Can be continuous quantity, such as total cholesterol, granulocyte percentage etc..Diagnosis and treatment active set in step S1 is made of diagnosis and treatment activity
Set, and diagnosis and treatment activity refer to by it is various check, using drug, using the methods of instrument and operation, disease is made and is sentenced
Disconnected and elimination disease, alleviates the state of an illness, palliates the agonizing sufferings, and improves function, and extending life helps the restorative activity of patient.And it is narrow
The diagnosis and treatment activity of justice refers to doctor according to diagnosis and treatment phenomenon, the therapeutic scheme made, used here as narrow sense meaning.In clinical practice,
Clinical picture is corresponding with diagnosis and treatment activity, constitutes clinical picture-two tuple of diagnosis and treatment activity so that clinical picture can be determined uniquely
Diagnosis and treatment activity, that is, the element that clinical picture is concentrated are corresponding with the element in diagnosis and treatment active set, so as to according to clinic
The clinical picture that phenomenon is concentrated, determines the diagnosis and treatment activity in corresponding diagnosis and treatment active set.
It is to be appreciated that the correspondence for the element and the element in diagnosis and treatment active set that clinical picture is concentrated is not limited to one by one
Correspondence can also be one-to-many or many-to-one correspondence.
Specifically, referring again to Fig. 1, in step sl, healthcare givers can for the patient for medical treatment of being admitted to hospital carry out interrogation and
Conventional equipment detects, to obtain patient information；In step s 2, believed according to interrogation and the detected patient of conventional equipment
Breath, structure clinical picture vector；In step s3, in the clinical picture vector set p built in advance, classified by k arest neighbors
Algorithm calculates and finds k nearest neighbors of foregoing clinical phenomenon vector, wherein k >=1；In step s 4, pass through aforementioned k
The element in diagnosis and treatment active set corresponding to nearest neighbors, obtains the diagnosis and treatment activity corresponding to clinical picture vector, to shape
At clinical path.Usually, step S1, step S2, step S3 and step S4 are real by the Application of clinical pathway group of medical institutions
It applies.
The core concept of k arest neighbors sorting algorithms used in clinical path matching process according to the present invention is if one
Most of in k of a sample in feature space most adjacent samples belong to some classification, then the sample also belongs to this
A classification, and with the characteristic of sample in this classification.This method on determining categorised decision only according to closest one or
The classifications of the several samples of person determines the classification belonging to sample to be divided.Knn algorithms in classification decision, only with it is minimal amount of adjacent
Sample is related, so as to reduce operand, improves arithmetic speed.
In clinical path matching process according to the present invention, this data digging method application of k arest neighbors sorting algorithm
Into the matching process of clinical path, so as to accurately determine diagnosis and treatment activity according to patient information, it is thus possible to accurately
Clinical path is obtained, compared with the prior art, clinical path matching process according to the present invention can be more acurrate and promptly complete
At the determination of the diagnosis and treatment scheme of patient.
Further, step S4 is preferably included each element in clinical picture vector and clinical picture collection p into row distance
It calculates, and result of calculation is arranged from small to large, the minimum element of k distance before taking, to obtain the clinical picture vector
K nearest neighbors.It is to be appreciated that distance described herein is intended only as a kind of metric form of similitude between vector,
In clinical path matching process according to the present invention, any distance for meeting metric requirements may be used, which includes
But it is not limited to：Euclidean distance (Euclidean distance), mahalanobis distance (Mahalanobis distance).
The number that the determination process of clinical path can be passed through science by clinical path matching process according to the present invention as a result,
Further quantified according to method for digging, so as to more accurately determine clinical path, avoids clinical path determination process
In uncertainty.
Further, in step s 4, k is preferably taken as 1, it is thus possible to clinical path is more directly obtained, to
It provides and more specifically instructs for the Application of clinical pathway group of medical institutions.
Hereinafter, with reference to Fig. 2, illustrate clinical path matching process involved according to the second embodiment of the present invention.
For convenience of description, the step different from clinical path matching process according to first embodiment is only specifically described here
Suddenly.In fig. 2, clinical path matching process according to second embodiment further includes step S5, in step s 5, in controlling for patient
During treatment, according to the patient information obtained in real time, clinical path is corrected.
Specifically, in the therapeutic process of patient, if therapeutic effect is not notable or new burst clinical picture occurs,
The Application of clinical pathway group of medical institutions needs to re-start interrogation and deep detection, to obtain new patient information, to
New clinical path is built, and then rational dynamic self-adapting tune can be carried out to clinical path according to the concrete condition of patient
It is whole, so that clinical path can be more flexible, more accurately adaptable with patient symptom, it is thus possible to instruct medical staff
Optimal completion Rehabilitation process obtains better medical effect, improves medical service quality, distributes medical resource rationally,
Realize that the comprehensive income of medical resource maximizes.
Further, the step S5 of clinical path matching process according to second embodiment includes the following steps：In patient
Therapeutic process in, if therapeutic scheme effect is not notable or new burst clinical picture occurs, needs to reacquire and suffer from
Person's information, to build new clinical picture vector；And concentrated in the clinical picture built in advance, classified by k arest neighbors
Algorithm calculates and finds k nearest neighbors of the clinical picture vector, to correct the clinical path, wherein k >=1.
Specifically, in the therapeutic process of patient, if therapeutic scheme effect is not notable or new burst clinic occurs
Phenomenon then needs to reacquire patient information, to build new clinical picture vector, when new clinical picture vector has been built
It after, is concentrated in clinical picture, is calculated by k arest neighbors sorting algorithms and the k for finding new clinical picture vector is a recently
Neighbors, wherein k >=1；By the element in the diagnosis and treatment active set corresponding to aforementioned k recently neighbors, the clinic is obtained
Diagnosis and treatment activity corresponding to phenomenon vector, to form clinical path.
As a result, according to the concrete condition of patient, clinical path can obtain reasonable, accurate and quantization dynamic self-adapting tune
It is whole, so that clinical path can be more flexible, more accurately adaptable with patient symptom, it is thus possible to instruct medical staff
Optimal completion Rehabilitation process obtains better medical effect, improves medical service quality, distributes medical resource rationally,
Realize that the comprehensive income of medical resource maximizes.
Hereinafter, with reference to Fig. 3, illustrate according to the third embodiment of the invention involved clinical path matching process.
For convenience of description, the step different from clinical path matching process according to second embodiment is only specifically described here
Suddenly.In figure 3, clinical path matching process according to the third embodiment of the invention further includes step S6, i.e., according to clinical anti-
Clinical picture collection p and/or diagnosis and treatment active set are corrected in feedback.
Specifically, in medical practice, the Application of clinical pathway group of medical institutions feeds back according to clinic to clinical path
Evaluation group is instructed to propose to correct the request of clinical picture collection p and diagnosis and treatment active set, wherein clinic feedback refers to therapeutic process
In, there is not corresponding situation in clinical picture collection p and diagnosis and treatment active set, such as the symptom of patient occurs in existing clinical picture
The situation that do not list.Later, clinical path instructs evaluation group to submit a report asking for clinical path administration committee, after approved, clinical road
Diameter instructs the amendment that evaluation group organizes each section's experts' evaluation that can carry out clinical picture collection p and diagnosis and treatment active set, and result is anti-
It is fed to clinical path administration committee.It is existing that clinical path instructs evaluation group that Application of clinical pathway group is organized to carry out new clinic
As collection p and diagnosis and treatment active set Training and Learning, it is ensured that Application of clinical pathway group in time, effectively carries out clinical path practice and lives
It is dynamic.
In the present embodiment, clinical picture collection p and diagnosis and treatment active set can be adjusted according to clinic feedback, so as to root
The disease symptoms for factually trampling middle appearance to generate corresponding diagnosis and treatment activity, and then determine clinical path, it is thus possible to medicine
Development be adapted, avoid medical procedure ossify.
Further, in step s 6, it corrects the clinical picture collection and/or the diagnosis and treatment active set preferably includes：Institute
It states the element additions and deletions of clinical picture collection, the element additions and deletions of the diagnosis and treatment active set, and/or corrects the element of clinical picture collection and examine
The correspondence between the element of active set is treated, so as to the current situation closer to medicine, avoids generating the disease with patient
The clinical path that disease is not consistent.
It is to be appreciated that it can also includes the clinical picture collection to correct the clinical picture collection and the diagnosis and treatment active set
Element amendment and/or the element amendment of the diagnosis and treatment active set, such as some clinical picture vector that clinical picture is concentrated is corrected,
Such as total cholesterol, total bilirubin, uric acid, granulocyte percentage and blood platelet quantizating index.
With reference to Fig. 4, the clinical path matching system 1 of first embodiment according to the present invention, clinical path matching are shown
System includes patient information acquisition module 11, clinical picture vector structure module 12, nearest neighbor algorithm execution module 13 and clinical road
Diameter generation module 14, wherein patient information acquisition module 11 is for obtaining patient information；Clinical picture vector builds module 12 and uses
In the patient information according to acquisition, structure clinical picture vector；Nearest neighbor algorithm execution module 13 is used to concentrate in clinical picture, leads to
Cross the k nearest neighbors that k arest neighbors sorting algorithms calculate and find the clinical picture vector, wherein k >=1；Clinical path
Generation module 14 is used to, by the element in the diagnosis and treatment active set corresponding to aforementioned k recently neighbors, obtain described clinical existing
Diagnosis and treatment activity as corresponding to vector, to form clinical path.
It is to be appreciated that patient information acquisition module 11 as described herein, clinical picture vector structure module 12, nearest neighbor algorithm
The modules such as execution module 13 and clinical path generation module 14 can be the hardware or software module of terminal device, can also be each
Kind can be into the terminal device of row data communication, such as desktop computer, laptop, personal digital assistant (Personal
Digital Assistant, PDA), mobile phone (Mobile Phone, MP) etc..
The clinical path matching system 1 of first embodiment according to the present invention is specifically described below in conjunction with Fig. 4,
But without limitation.
In the clinical matching system 1 of first embodiment according to the present invention, clinical picture collection and diagnosis and treatment active set are usual
It is instructed evaluation group to organize by the clinical path of medical institutions and is built in advance, the clinical picture collection built in advance and diagnosis and treatment activity
Collection is stored in data memory module (not shown), which is arranged in first embodiment according to the present invention
In clinical matching system 1, wherein clinical picture collection is the set being made of clinical picture vector.The data memory module can be
A kind of read-only memory unit ROM, electrically-erasable storage unit EEPROM, flash memory cell FLASH or solid hard disk etc..
During clinic diagnosis, doctor to the interrogation of patient and by medical apparatus and instruments detect about people
The summation of the physical and chemical index of body, referred to as clinical picture.Specifically, clinical picture vector p is a series of combination of indexs, i.e. p=
{e1,e2,...,en, wherein eiReferred to as i-th of index, these usual indexs all have stronger intuitive meaning.These parameters packet
The quantizations such as the total cholesterol detected by Medical Devices, total bilirubin, uric acid, granulocyte percentage and blood platelet are included to refer to
It marks, and the semantic phenomena obtained by doctor's interrogation, such as dizziness, nausea, diarrhea number etc..These indexs can be boolean
Amount, for example whether there is phenomena such as dizzy, nausea, or discrete magnitude, such as diarrhea number etc., or can be continuous
Amount, such as total cholesterol, granulocyte percentage etc..Diagnosis and treatment active set in step S1 is the set being made of diagnosis and treatment activity, and
Diagnosis and treatment activity refer to by it is various check, using drug, using the methods of instrument and operation, disease is judged and eliminated to disease
Disease alleviates the state of an illness, palliates the agonizing sufferings, and improves function, and extending life helps the restorative activity of patient.And the diagnosis and treatment of narrow sense are lived
Dynamic finger doctor is according to diagnosis and treatment phenomenon, the therapeutic scheme made, used here as narrow sense meaning.In clinical practice, clinical picture with
Diagnosis and treatment activity is corresponding, constitutes clinical picture-two tuple of diagnosis and treatment activity so that and clinical picture can uniquely determine diagnosis and treatment activity,
The element that namely clinical picture is concentrated is corresponding with the element in diagnosis and treatment active set, so as to what is concentrated according to clinical picture
Clinical picture determines the diagnosis and treatment activity in corresponding diagnosis and treatment active set.
It is to be appreciated that the correspondence for the element and the element in diagnosis and treatment active set that clinical picture is concentrated is not limited to one by one
Correspondence can also be one-to-many or many-to-one correspondence.
Specifically, referring again to Fig. 4, in the clinical path matching system 1 of first embodiment according to the present invention, patient
Data obtaining module 11 carries out interrogation by the patient to medical treatment of being admitted to hospital and conventional equipment detects, to obtain patient information；Face
Bed phenomenon vector structure module 12 is according to interrogation and the detected patient information of conventional equipment, structure clinical picture vector；It is adjacent
Nearly algorithm performs module 13 is led to for accessing above-mentioned data memory module, and in the clinical picture vector set p built in advance
Cross the k nearest neighbors that k arest neighbors sorting algorithms calculate and find foregoing clinical phenomenon vector, wherein k >=1；Clinical path
Generation module 14 is used to access data memory module and by the diagnosis and treatment active set corresponding to aforementioned k recently neighbors
Element obtains the diagnosis and treatment activity corresponding to clinical picture vector, to form clinical path.
The core of k arest neighbors sorting algorithms used in the clinical path matching system 1 of first embodiment according to the present invention
Thought is thought if most of in k of the sample in feature space most adjacent samples belong to some classification,
The sample also belongs to this classification, and with the characteristic of sample in this classification.This method is determining categorised decision on foundation
The classification of one or several closest samples determines the classification belonging to sample to be divided.Knn algorithms are in classification decision, only
It is related with minimal amount of adjacent sample, so as to reduce operand, improve arithmetic speed.
In the clinical path matching system 1 of first embodiment according to the present invention, this data of k arest neighbors sorting algorithm
Method for digging is applied in the matching process of clinical path, so as to accurately determine diagnosis and treatment activity according to patient information, because
And clinical path can be accurately obtained, compared with the prior art, clinical path matching system according to the present invention can be more accurate
The determination of diagnosis and treatment scheme that is true and being quickly accomplished patient.
Further, nearest neighbor algorithm execution module 13 is preferably by each element in clinical picture vector and clinical picture collection p
It is calculated into row distance, and result of calculation is arranged from small to large, the minimum element of k distance before taking, to obtain foregoing clinical
K nearest neighbors of phenomenon vector.It is to be appreciated that distance described herein is intended only as a kind of degree of similitude between vector
Amount mode, in clinical path matching process according to the present invention, any distance for meeting metric requirements may be used, should
Distance includes but not limited to：Euclidean distance (Euclidean distance), mahalanobis distance (Mahalanobis distance).
Therefore, clinical path matching system according to the present invention can be by the data digging method of science by clinical path
Determination process quantified, so as to more accurately determine clinical path, avoid in clinical path determination process not
Certainty.
Further, in the course of work of nearest neighbor algorithm execution module 13, k is preferably taken as 1, it is thus possible to more straight
Ground connection obtains clinical path, is more specifically instructed to be provided for the Application of clinical pathway group of medical institutions.
Hereinafter, with reference to Fig. 5, illustrate clinical path matching system 2 involved according to the second embodiment of the present invention.
For convenience of description, it only specifically describes here different from clinical path matching system 1 according to first embodiment
Part.In Figure 5, clinical path matching system 2 according to the present invention further includes clinical path correcting module 25, the clinical path
Correcting module 25 is used in the therapeutic process of patient, according to the patient information newly obtained, builds new clinical path.Specifically
Ground, if therapeutic effect is not notable or new burst clinical picture occurs, needs to re-start in the therapeutic process of patient
Interrogation and deep detection, to obtain new patient information, clinical path correcting module 25 is according to the patient information obtained come structure
New clinical path is built, and then rational dynamic self-adapting can be carried out to clinical path according to the concrete condition of patient,
So that clinical path can be more flexible, more accurately adaptable with patient symptom, it is thus possible to instruct medical staff most
Excellent completion Rehabilitation process obtains better medical effect, improves medical service quality, distributes medical resource rationally, real
The comprehensive income of existing medical resource maximizes.
Further, when therapeutic scheme effect is not notable or new burst clinical picture occurs, clinical path amendment
Module 25 can start patient information acquisition module to reacquire patient information, and can start clinical picture vector structure successively
Block, nearest neighbor algorithm execution module and clinical path generation module are modeled, to correct the clinical path.Clinical picture vector is built
Module, the principle of nearest neighbor algorithm execution module and clinical path generation module and working mechanism have hereinbefore been described, therefore
This is not repeated.
As a result, according to the concrete condition of patient, clinical path can obtain reasonable, accurate and quantization dynamic self-adapting tune
It is whole, so that clinical path can be more flexible, more accurately adaptable with patient symptom, it is thus possible to instruct medical staff
Optimal completion Rehabilitation process obtains better medical effect, improves medical service quality, distributes medical resource rationally,
Realize that the comprehensive income of medical resource maximizes.
Hereinafter, with reference to Fig. 6, illustrate according to the third embodiment of the invention involved clinical path matching system 3.
For convenience of description, it only specifically describes here different from clinical path matching system 2 according to second embodiment
Part.In figure 6, clinical path matching system 3 according to the present invention further includes that clinical picture collection and diagnosis and treatment active set correct mould
Block 36, the clinical picture collection and diagnosis and treatment active set correcting module 36 be used for according to clinic feed back, correct clinical picture collection p and/or
Diagnosis and treatment active set.
Specifically, in medical practice, the Application of clinical pathway group of medical institutions feeds back according to clinic to clinical path
Evaluation group is instructed to propose to correct the request of clinical picture collection p and diagnosis and treatment active set.Later, clinical path instructs evaluation group to report
Please clinical path administration committee, after approved, it is clinical existing that clinical path instructs evaluation group that each section's experts' evaluation is organized to pass through
Clinical picture collection p and diagnosis and treatment active set are modified with diagnosis and treatment active set correcting module 36 as collecting, revised clinical picture
Collection p and diagnosis and treatment active set can be stored in clinical picture collection and diagnosis and treatment active set correcting module 36, and result is fed back to clinic
The path management committee.Clinical path instruct evaluation group organize Application of clinical pathway group carry out new clinical picture collection p and
Diagnosis and treatment active set Training and Learning, it is ensured that Application of clinical pathway group is timely, effectively carries out clinical path practical activity.
In the present embodiment, clinical picture collection and diagnosis and treatment active set correcting module 36 can be according to clinic feedbacks to clinical picture
Collection p and diagnosis and treatment active set are adjusted, so as to according to the disease symptoms occurred in practice, be lived with generating corresponding diagnosis and treatment
It is dynamic, and then determine clinical path, it is thus possible to the development with medicine is adapted, and avoids ossifing for medical procedure.
Further, in clinical picture collection and diagnosis and treatment active set correcting module 36, clinical picture collection and/or diagnosis and treatment are corrected
Active set preferably includes：The element additions and deletions of clinical picture collection, the element additions and deletions of diagnosis and treatment active set, and/or correct clinical picture collection
Element and the element of diagnosis and treatment active set between correspondence avoid generating so as to the current situation closer to medicine
The clinical path not being consistent with the illness of patient.
It is to be appreciated that the element amendment of clinical picture collection can also be included by correcting clinical picture collection and the diagnosis and treatment active set
It is vectorial with the element amendment of diagnosis and treatment active set, such as some clinical picture that amendment clinical picture is concentrated, such as total cholesterol, total courage
The quantizating index such as red pigment, uric acid, granulocyte percentage and blood platelet.
The above is merely preferred embodiments of the present invention, be not intended to limit the invention, it is all the present invention spirit and
Any amendment, equivalent replacement and improvement etc., should all be included in the protection scope of the present invention made by within principle.
Claims (10)
1. a kind of clinical path matching process, includes the following steps：
Obtain patient information；
According to the patient information of acquisition, structure clinical picture vector；
It is concentrated in the clinical picture built in advance, the k of the clinical picture vector is calculated and found by k arest neighbors sorting algorithms
A nearest neighbors, wherein k >=1；And
By the element in the diagnosis and treatment active set built in advance corresponding to the k recently neighbors, obtain described clinical existing
Diagnosis and treatment activity as corresponding to vector, to form clinical path.
2. clinical path matching process as described in claim 1, which is characterized in that further comprising the steps of：According to obtaining in real time
The patient information taken corrects the clinical path.
3. clinical path matching process as claimed in claim 2, which is characterized in that patient's letter that the basis obtains in real time
Breath, the step of correcting the clinical path include：
In the therapeutic process of patient, if therapeutic scheme effect is not notable or new burst clinical picture occurs, need
Patient information is reacquired, to build new clinical picture vector；And
It is concentrated in the clinical picture built in advance, the k of the clinical picture vector is calculated and found by k arest neighbors sorting algorithms
A nearest neighbors, to correct the clinical path, wherein k >=1.
4. clinical path matching process as described in any one of claims 1 to 3, which is characterized in that further comprising the steps of：Root
It is fed back according to clinic, corrects the clinical picture collection and/or the diagnosis and treatment active set.
5. clinical path matching process as described in any one of claims 1 to 3, which is characterized in that described to build in advance
Clinical picture is concentrated, and the step of k nearest neighbors of the clinical picture vector is calculated and found by k arest neighbors sorting algorithms
Suddenly include：The clinical picture vector is calculated at a distance from each element that the clinical picture is concentrated, and the knot that distance will be calculated
Fruit arranges from small to large, the minimum element of k distance before taking, to obtain k nearest neighbors of the clinical picture vector.
6. a kind of clinical path matching system, which is characterized in that build mould including patient information acquisition module, clinical picture vector
Block, nearest neighbor algorithm execution module and clinical path generation module, wherein
The patient information acquisition module is for obtaining patient information；
The clinical picture vector structure module is used for the patient information according to acquisition, structure clinical picture vector；
The nearest neighbor algorithm execution module is used to concentrate in the clinical picture, calculates and finds by k arest neighbors sorting algorithms
K nearest neighbors of the clinical picture vector, wherein k >=1；And
The clinical path generation module is used for through the element in the diagnosis and treatment active set corresponding to the k recently neighbors,
The diagnosis and treatment activity corresponding to the clinical picture vector is obtained, to form clinical path.
7. clinical path matching system as claimed in claim 6, which is characterized in that further include clinical path correcting module, institute
Clinical path correcting module is stated for according to the patient information that obtains in real time, correcting the clinical path.
8. clinical path matching system as claimed in claim 7, which is characterized in that when therapeutic scheme effect is not notable or goes out
When now new burst clinical picture, the clinical path correcting module can start the patient information acquisition module to obtain again
Patient information is taken, and clinical picture vector structure module, the nearest neighbor algorithm execution module and described can be started successively
Clinical path generation module, to correct the clinical path.
9. such as claim 6 to 8 any one of them clinical path matching system, which is characterized in that further include clinical picture collection
With diagnosis and treatment active set correcting module, the clinical picture collection and diagnosis and treatment active set correcting module are used to be fed back according to clinic, correct
The clinical picture collection and/or the diagnosis and treatment active set.
10. such as claim 6 to 8 any one of them clinical path matching system, which is characterized in that the nearest neighbor algorithm executes
Module is for calculating the clinical picture vector at a distance from each element that the clinical picture is concentrated, and the knot that will calculate distance
Fruit arranges from small to large, the minimum element of k distance before taking, to obtain k nearest neighbors of the clinical picture vector.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710205898.XA CN108665975A (en) | 2017-03-30 | 2017-03-30 | Clinical path matching process and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710205898.XA CN108665975A (en) | 2017-03-30 | 2017-03-30 | Clinical path matching process and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108665975A true CN108665975A (en) | 2018-10-16 |
Family
ID=63786732
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710205898.XA Pending CN108665975A (en) | 2017-03-30 | 2017-03-30 | Clinical path matching process and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108665975A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109545383A (en) * | 2018-11-12 | 2019-03-29 | 北京懿医云科技有限公司 | Actual clinical path mutation detection method and device, storage medium, electronic equipment |
| CN111739634A (en) * | 2020-05-14 | 2020-10-02 | 平安科技（深圳）有限公司 | Method, device and equipment for intelligently grouping similar patients and storage medium |
| CN112489791A (en) * | 2020-12-09 | 2021-03-12 | 镇江市第四人民医院（镇江市妇幼保健院） | Clinical path-based decision-making assisting system and method |
| CN117542535A (en) * | 2024-01-10 | 2024-02-09 | 电子科技大学 | Clinical path correction method and device |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102567607A (en) * | 2010-12-28 | 2012-07-11 | 广州中医药大学第二附属医院 | Clinic path information system |
| CN104866979A (en) * | 2015-06-08 | 2015-08-26 | 苏芮 | Traditional Chinese medicine case data processing method and system of emergent acute infectious disease |
| CN106096286A (en) * | 2016-06-15 | 2016-11-09 | 北京千安哲信息技术有限公司 | Clinical path formulating method and device |
| CN106156497A (en) * | 2016-07-01 | 2016-11-23 | 深圳市前海安测信息技术有限公司 | Medical auxiliary system based on clinical path and method |
| CN106228023A (en) * | 2016-08-01 | 2016-12-14 | 清华大学 | A kind of clinical path method for digging based on body and topic model |
-
2017
- 2017-03-30 CN CN201710205898.XA patent/CN108665975A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102567607A (en) * | 2010-12-28 | 2012-07-11 | 广州中医药大学第二附属医院 | Clinic path information system |
| CN104866979A (en) * | 2015-06-08 | 2015-08-26 | 苏芮 | Traditional Chinese medicine case data processing method and system of emergent acute infectious disease |
| CN106096286A (en) * | 2016-06-15 | 2016-11-09 | 北京千安哲信息技术有限公司 | Clinical path formulating method and device |
| CN106156497A (en) * | 2016-07-01 | 2016-11-23 | 深圳市前海安测信息技术有限公司 | Medical auxiliary system based on clinical path and method |
| CN106228023A (en) * | 2016-08-01 | 2016-12-14 | 清华大学 | A kind of clinical path method for digging based on body and topic model |
Non-Patent Citations (2)
| Title |
|---|
| HUANG ZHENGXING 等: "Similarity Measure Between Patient Traces for Clinical Pathway Analysis: Problem, Method, and Applications", 《IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS》 * |
| 章虎林: "案例推理和关联规则在中风智能诊疗中的应用研究", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109545383A (en) * | 2018-11-12 | 2019-03-29 | 北京懿医云科技有限公司 | Actual clinical path mutation detection method and device, storage medium, electronic equipment |
| CN111739634A (en) * | 2020-05-14 | 2020-10-02 | 平安科技（深圳）有限公司 | Method, device and equipment for intelligently grouping similar patients and storage medium |
| WO2021139116A1 (en) * | 2020-05-14 | 2021-07-15 | 平安科技（深圳）有限公司 | Method, apparatus and device for intelligently grouping similar patients, and storage medium |
| CN112489791A (en) * | 2020-12-09 | 2021-03-12 | 镇江市第四人民医院（镇江市妇幼保健院） | Clinical path-based decision-making assisting system and method |
| CN112489791B (en) * | 2020-12-09 | 2024-03-05 | 镇江市第四人民医院（镇江市妇幼保健院） | Clinical path-based auxiliary decision making system and method |
| CN117542535A (en) * | 2024-01-10 | 2024-02-09 | 电子科技大学 | Clinical path correction method and device |
| CN117542535B (en) * | 2024-01-10 | 2024-03-22 | 电子科技大学 | Clinical path correction method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Yin et al. | A health decision support system for disease diagnosis based on wearable medical sensors and machine learning ensembles | |
| US11217349B2 (en) | System and method for processing human related data including physiological signals to make context aware decisions with distributed machine learning at edge and cloud | |
| Newaz et al. | Adversarial attacks to machine learning-based smart healthcare systems | |
| US20190374160A1 (en) | Hierarchical health decision support system and method | |
| CN108665975A (en) | Clinical path matching process and system | |
| Bellos et al. | Identification of COPD patients’ health status using an intelligent system in the CHRONIOUS wearable platform | |
| Ho et al. | Septic shock prediction for patients with missing data | |
| Wang et al. | Overview of the BioCreative/OHNLP challenge 2018 task 2: clinical semantic textual similarity | |
| Nagireddi et al. | The analysis of pain research through the lens of artificial intelligence and machine learning | |
| Zhao et al. | Detecting adverse drug events with multiple representations of clinical measurements | |
| CN111798954A (en) | Drug combination recommendation method based on time attention mechanism and graph convolution network | |
| US20220122735A1 (en) | System and method for processing human related data including physiological signals to make context aware decisions with distributed machine learning at edge and cloud | |
| JP2022527174A (en) | Intraoperative clinical decision support system | |
| US20160188832A1 (en) | Physiologic Severity of Illness Score for Acute Care Patients | |
| WO2021073255A1 (en) | Time series clustering-based medication reminder method and related device | |
| Zai et al. | Assessing hospital readmission risk factors in heart failure patients enrolled in a telemonitoring program | |
| Maali et al. | A novel partially connected cooperative parallel PSO-SVM algorithm: Study based on sleep apnea detection | |
| Torres-Espín et al. | Topological network analysis of patient similarity for precision management of acute blood pressure in spinal cord injury | |
| CN111627515B (en) | Medicine recommendation method, device, electronic equipment and medium | |
| Al-Mualemi et al. | A deep learning-based sepsis estimation scheme | |
| KR20210142238A (en) | Clinical support system based on deep learning and decision support method using the system | |
| Tarle et al. | An artificial neural network based pattern classification algorithm for diagnosis of heart disease | |
| Chen et al. | Artificial Intelligence‐Based Medical Sensors for Healthcare System | |
| US20240321447A1 (en) | Method and System for Personalized Prediction of Infection and Sepsis | |
| TWI510948B (en) | Health activity plan establishment system, establishment method and computer program product |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181016 |