CN106446549A - Prediction method and prediction system for dyspepsia based on incremental type neural network model - Google Patents
Prediction method and prediction system for dyspepsia based on incremental type neural network model Download PDFInfo
- Publication number
- CN106446549A CN106446549A CN201610860220.0A CN201610860220A CN106446549A CN 106446549 A CN106446549 A CN 106446549A CN 201610860220 A CN201610860220 A CN 201610860220A CN 106446549 A CN106446549 A CN 106446549A
- Authority
- CN
- China
- Prior art keywords
- neuron
- indigestion
- network model
- neural network
- data
- 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
- 201000006549 dyspepsia Diseases 0.000 title claims abstract description 124
- 238000003062 neural network model Methods 0.000 title claims abstract description 86
- 238000000034 method Methods 0.000 title claims abstract description 24
- 239000013598 vector Substances 0.000 claims abstract description 55
- 230000007170 pathology Effects 0.000 claims abstract description 29
- 238000007689 inspection Methods 0.000 claims abstract description 23
- 238000012549 training Methods 0.000 claims abstract description 23
- 238000012937 correction Methods 0.000 claims abstract description 8
- 210000002569 neuron Anatomy 0.000 claims description 111
- 238000012806 monitoring device Methods 0.000 claims description 14
- 238000004891 communication Methods 0.000 claims description 12
- 230000029087 digestion Effects 0.000 claims description 9
- 239000000284 extract Substances 0.000 claims description 9
- 238000013139 quantization Methods 0.000 claims description 9
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 claims description 8
- 230000006870 function Effects 0.000 claims description 7
- 230000008859 change Effects 0.000 claims description 6
- 201000010099 disease Diseases 0.000 claims description 5
- 230000036541 health Effects 0.000 claims description 5
- 238000012544 monitoring process Methods 0.000 claims description 5
- 210000005036 nerve Anatomy 0.000 claims description 5
- WQZGKKKJIJFFOK-GASJEMHNSA-N Glucose Natural products OC[C@H]1OC(O)[C@H](O)[C@@H](O)[C@@H]1O WQZGKKKJIJFFOK-GASJEMHNSA-N 0.000 claims description 3
- 241000190070 Sarracenia purpurea Species 0.000 claims description 3
- 230000033228 biological regulation Effects 0.000 claims description 3
- 230000005540 biological transmission Effects 0.000 claims description 3
- 239000008280 blood Substances 0.000 claims description 3
- 210000004369 blood Anatomy 0.000 claims description 3
- 230000007423 decrease Effects 0.000 claims description 3
- 230000003247 decreasing effect Effects 0.000 claims description 3
- 238000002474 experimental method Methods 0.000 claims description 3
- 239000008103 glucose Substances 0.000 claims description 3
- 230000001537 neural effect Effects 0.000 claims description 3
- 238000011017 operating method Methods 0.000 claims description 3
- 241000894007 species Species 0.000 claims description 3
- 238000010606 normalization Methods 0.000 abstract 1
- 238000005516 engineering process Methods 0.000 description 4
- 230000007935 neutral effect Effects 0.000 description 3
- 208000024891 symptom Diseases 0.000 description 3
- 230000000007 visual effect Effects 0.000 description 3
- 235000013399 edible fruits Nutrition 0.000 description 2
- 230000001575 pathological effect Effects 0.000 description 2
- 201000004569 Blindness Diseases 0.000 description 1
- 210000000577 adipose tissue Anatomy 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000037396 body weight Effects 0.000 description 1
- 238000013479 data entry Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 230000035622 drinking Effects 0.000 description 1
- 239000003651 drinking water Substances 0.000 description 1
- 235000020188 drinking water Nutrition 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003860 sleep quality Effects 0.000 description 1
- 230000000391 smoking effect Effects 0.000 description 1
- 210000002700 urine Anatomy 0.000 description 1
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/70—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for mining of medical data, e.g. analysing previous cases of other patients
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Public Health (AREA)
- Medical Informatics (AREA)
- Data Mining & Analysis (AREA)
- Biomedical Technology (AREA)
- Pathology (AREA)
- Databases & Information Systems (AREA)
- Epidemiology (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Medical Treatment And Welfare Office Work (AREA)
Abstract
The invention discloses a prediction method for dyspepsia based on an incremental type neural network model. The prediction method comprises the steps of establishing a daily data database of dyspepsia; conducting training on the neural network model; collecting daily life data and sending to a server, and storing the daily life data in a user daily data record sheet; extracting the data of the day from the user daily data record sheet to form a n-dimension vector quantity, inputting the n-dimension vector quantity into a dyspepsia pathology neural network model to conduct dyspepsia probability forecasting after the n-dimension vector quantity is subjected to normalization processing; using intelligent home dyspepsia nurse equipment to judge whether a dyspepsia probability value is bigger than 0.5; going to the hospital by oneself to inspect when the user is judged to have the dyspepsia, and transferring an inspection result through the intelligent home dyspepsia nurse equipment back to the server which judges whether the inspection result is correct; executing an incremental type algorithm when the inspection result is incorrect, and conducting a dynamic correction on the neural network model. The prediction method for dyspepsia based on the incremental type neural network model is accurate in prediction, and a tailor made neural network model is made for every user.
Description
Technical field
The invention belongs to field of medical technology, more particularly to a kind of indigestion based on increment type neural network model
Forecasting Methodology and forecasting system.
Background technology
Currently domestic each health management system arranged be respectively provided with indigestion prediction and evaluation, its use prediction mode be data
Join.Its principle is that by system matches fixed data and then personal lifestyle data entry system is shown ill probability.But due to people
The complexity of body and disease, unpredictability, in the form of expression of bio signal and information, Changing Pattern (Self-variation with
Change after medical intervention) on, it is detected and signal representation, the data of acquisition and the analysis of information, decision-making etc. are all multi-party
All there is extremely complex non-linear relationship in face.So using traditional Data Matching can only be blindness data examination it is impossible to
Judge the logic association between data and data and variable, the codomain deviation obtaining is big, causes the specificity ten of system prediction
Point poor, domestic health management system arranged effectively Accurate Prediction cannot be carried out to personal indigestion so current.
Most of before this is all using BP neural network model to indigestion prediction, but when new detection data produces
When it is necessary to train neural network model again, operation efficiency is extremely low.And after system user scale increases, service
Device will be unable to complete in time training mission.
Content of the invention
The purpose of the present invention is that and overcomes the deficiencies in the prior art, there is provided one kind is based on increment type neural network model
Indigestion Forecasting Methodology and forecasting system, the present invention by neural network model train predict a large amount of patient in hospital pathology numbers
According to finding indigestion pathology and indigestion earlier life variations in detail, clinical symptoms, examination criteria value, people at highest risk special
Levy, the logic association between this several causes of disease and variable, ultimately form the digestion to indigestion illness probability Accurate Prediction not
Good pathology neural network model, the present invention passes through to gather user's daily life data, the periodicity of its data of active analysis, rule
Property suffers from indigestion probability eventually through indigestion pathology Neural Network model predictive user, is carried in the way of visual effect
Awake user's instant hospitalizing, constantly revises neural network model when Neural Network model predictive is inaccurate by increasable algorithm,
To set up, for each equipment user, the neural network model training for this user, with the increase of use time, to build
The vertical neural network model that this user is made to measure, accuracy rate is greatly improved.
To achieve these goals, the invention provides a kind of indigestion based on increment type neural network model is predicted
Method, comprises the steps：
Step (1), acquisition hospital's indigestion etiology and pathology data source and the daily monitoring data of patient, thus set up digestion
Bad daily data database；
Step (2), the daily data database of indigestion set up according to step (1) are off-line manner to neutral net
Model is trained, to obtain the indigestion pathology neural network model training；
Step (3), by intelligent monitoring device, the daily life data of user is acquired, and will collection daily life
Live data sends to server, and server preserves the daily life data of user to the daily data logger of user；
Step (4), from the daily data logger of user, extract same day data, form n-dimensional vector, and n-dimensional vector is done
Carry out indigestion probability pre- in the indigestion pathology neural network model training in input step (2) after normalized
Survey, obtain indigestion probability, server sends indigestion probability to wired home indigestion care appliances；
After step (5), the indigestion probability of wired home indigestion care appliances the reception server transmission, judge to disappear
Change whether bad probable value is more than 0.5, if greater than 0.5, be then judged to that this user obtained indigestion, attention device warns to carry
Awake user, if less than 0.5, is then judged to that this user does not obtain indigestion；
Step (6), when user is judged to indigestion, user voluntarily removes examination in hospital, and inspection result is led to
Cross wired home indigestion care appliances and send back server, server judges whether inspection result is correct, if checking knot
Fruit mistake, then explanation indigestion pathology Neural Network model predictive is inaccurate, if inspection result is correct, digestion is described not
Good pathology Neural Network model predictive is accurate；
Step (7), when inspection result mistake, from the daily data logger of user extract m days in record preserve to
In incremental data table, when the record quantity in incremental data table is more than h bar, execute increasable algorithm, to indigestion pathology
Neural network model carries out dynamic corrections；
Step (8), repeat step (3)～(7).
Further, the input layer of neural network model is n node, and hidden layer number is n*2+1, and output layer is 1
Node, extracts k bar record from indigestion daily data database table and is trained, every record is a n-dimensional vector, institute
There is data first before use through normalized so as to numerical value is in [0,1] interval, then execution following steps are to neutral net mould
Type is trained：
1) one n-dimensional vector of input, to neural network model, calculates all of weight vector in neural network model defeated to this
Enter the distance of n-dimensional vector, closest neuron is as won neuron, its computing formula is as follows：
Wherein：WkIt is the weight vector of triumph neuron, | | ... | | for Euclidean distance；
2) weight vector of the neuron in adjustment triumph neuron and triumph neuron field, formula is as follows：
Wherein：WjT () is neuron；Wj(t+1) weight vector before being adjustment and after adjustment；J belongs to triumph neuron neck
Domain；α (t) is learning rate, and it is as the function that the increase of iterations is gradually successively decreased, and span is [01], through multiple
It is 0.62 that Optimal learning efficiency is chosen in experiment；DjIt is the distance of neuron j and triumph neuron；σ (t) is as the letter that the time successively decreases
Number；Iteration all input n-dimensional vectors is input in neural network model and is trained each time, when the iteration reaching regulation
After number of times, neural network model training terminates.
Further, inspection result is sent back the lattice of the object information of server by wired home indigestion care appliances
Formula is：{ checking whether correct, blood glucose value }, server, after receiving object information, judges whether inspection result is correct.
Further, the increasable algorithm carrying out dynamic corrections to indigestion pathology neural network model is：
Vectorial for every in incremental data table V { V1,V2,…,Vn, it is sent in neural network model learning function
Row study, learning procedure is as follows：
1) first to output layer, each weight vector is assigned little random number and is done normalized, then utilizes input mode vector V
Mean value Avg (V), be initialized as the weights of unique neuron in the 0th layer of neural network model, and be set to win nerve
Unit, calculates its quantization error QE；
2) expand out 2 × 2 structures SOM from the 0th layer of neuron, and its level identities Layer is set to 1；
3) for each 2 × 2 structure SOM subnet expanded out in Layer layer, initialize the power of this 4 neurons
Value；The input vector set Ci of i-th neuron is set to sky, main label is set to NULL, the main label ratio r of neuron ii
It is set to 0；The abnormity early warning data vector V of new SOM inherits the triumph input vector set VX of his father's neuron；
4) select a vectorial VX from VXiDo following judgement：
If VXiFor the data of not tape label, then calculate its Euclidean distance with each neuron, chosen distance is the shortest
Neuron is as triumph neuron；
If VXiFor the data of tape label, then select main label and VXiLabel is identical and riThe maximum neuron of value is made
For triumph neuron, update this triumph neuron main label；
If can not find main label and VXiLabel identical neuron, then find and VXiClosest neuron i makees
For triumph neuron；
5) weights of neuron in triumph neuron and its neighborhood are adjusted, update the vectorial set W=W ∪ that wins
{VXi, calculate main label, the main label ratio r of triumph neuroniWith comentropy EiIf. not up to predetermined frequency of training,
Go to step 4)；
6) quantization error QE of each neuron in this neural network model after calculating is adjustedi, neuronal messages entropy Ei
With the average quantization error MQE of subnet, formula is as follows：
Wherein：WiFor the weight vector of neuron i, CiThe set constituting for all input vectors being mapped to neuron i；
Wherein：Ni represents to fall that label is the number of samples of i on neuron, and m represents to fall label data on neuron
Sum, T represents to fall the sample label species set on neuron；
Then judge：
If MQE>QE × threshold value q of father node, wherein q=0.71, then insert a line neuron in this SOM, turn step
Rapid 4)；
If Ei>The E of father nodei× threshold value p, wherein p=0.42, then grow one layer of new subnet from this neuron, will
The subnet newly growing increases in the subnet queue of Layer+1 layer；
If being not inserted into new neuron in SOM also do not grow new subnet, illustrate that the training of this subnet completes；
7) for all 2 × 2 structures SOM of the Layer+1 layer newly expanded out, iteration operating procedure 3)～5) to it again
It is trained, until neural network model no longer produces new neuron and new layering, whole training terminates.
Further, if user includes health check-up by other means and checks oneself, learn that oneself has suffered from indigestion, and intelligence
The attention device of energy family indigestion care appliances does not warn then it represents that wired home indigestion care appliances judge to be forbidden
Really, now execution step (6)～(7), wired home indigestion care appliances are sent to object information on server.
Present invention also offers a kind of forecasting system of described indigestion Forecasting Methodology, including intelligent monitoring device, intelligence
Energy device data acquisition device, server and wired home indigestion care appliances, described intelligent monitoring device and described intelligence
Device data acquisition device is connected, and described smart machine data acquisition unit is passed through communication device one and led to described server network
News, described wired home indigestion care appliances pass through communication device two and described server network communication.
Further, described wired home indigestion care appliances are provided with attention device.
Further, described intelligent monitoring device includes Intelligent worn device, Intelligent water cup, Intelligent weight claim, intelligent horse
Bucket and Intelligent light sensing equipment.
Beneficial effects of the present invention：
1st, the present invention is trained by neural network model and predicts a large amount of patient in hospital pathological data, finds indigestion pathology
With indigestion earlier life variations in detail, clinical symptoms, examination criteria value, people at highest risk's feature, between this several causes of disease
Logic association and variable, ultimately form the indigestion pathology neural network model to indigestion illness probability Accurate Prediction,
The present invention passes through to gather user's daily life data, and the periodicity of its data of active analysis, regularity are eventually through indigestion
Pathology Neural Network model predictive user suffers from indigestion probability, reminds user's instant hospitalizing and pre- in the way of visual effect
Anti-.
2nd, when Neural Network model predictive is inaccurate, neural network model is constantly revised, to be directed to by increasable algorithm
Each equipment user sets up the neural network model training for this user, with the increase of use time, to set up to this
The neural network model that user makes to measure, accuracy rate is greatly improved.
Brief description
In order to be illustrated more clearly that the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing
Have technology description in required use accompanying drawing be briefly described it should be apparent that, drawings in the following description be only this
Some embodiments of invention, for those of ordinary skill in the art, on the premise of not paying creative work, acceptable
Other accompanying drawings are obtained according to these accompanying drawings.
Fig. 1 is the flow chart of the embodiment of the present invention.
Specific embodiment
Below in conjunction with the accompanying drawings invention is further illustrated, but be not limited to the scope of the present invention.
Embodiment
As shown in figure 1, a kind of indigestion Forecasting Methodology based on increment type neural network model that the present invention provides, bag
Include following steps：
Step (1), acquisition hospital's indigestion etiology and pathology data source and the daily monitoring data of patient, thus set up digestion
Bad daily data database；
Wherein daily monitoring data is 16 item data, and its 16 item data is age, sex, body fat, the amount of drinking water and frequency, little
Just number of times, urine color, body weight, BMI index, the length of one's sleep, sleep quality, time for falling asleep, smoking capacity (daily), the amount of drinking
(daily), pursues an occupation, daily travel distance etc. 16 item data, and the present invention sets up 16 dimensional vectors with 16 item data；
Step (2), the daily data database of indigestion set up according to step (1) are off-line manner to neutral net
Model is trained, to obtain the indigestion pathology neural network model training；
Step (3), by intelligent monitoring device, the daily life data of user is acquired, and will collection daily life
Live data sends to server, and server preserves the daily life data of user to the daily data logger of user；
Step (4), from the daily data logger of user, extract same day data, form 16 dimensional vectors, and to 16 dimensional vectors
Carry out indigestion probability in the indigestion pathology neural network model training in input step (2) after doing normalized
Prediction, obtains indigestion probability, server sends indigestion probability to wired home indigestion care appliances；
After step (5), the indigestion probability of wired home indigestion care appliances the reception server transmission, judge to disappear
Change whether bad probable value is more than 0.5, if greater than 0.5, be then judged to that this user obtained indigestion, attention device warns to carry
Awake user, if less than 0.5, is then judged to that this user does not obtain indigestion；
Step (6), when user is judged to indigestion, user voluntarily removes examination in hospital, and inspection result is led to
Cross wired home indigestion care appliances and send back server, server judges whether inspection result is correct, if checking knot
Fruit mistake, then explanation indigestion pathology Neural Network model predictive is inaccurate, if inspection result is correct, digestion is described not
Good pathology Neural Network model predictive is accurate；
Step (7), when inspection result mistake, from the daily data logger of user extract 7 days in record preserve to
In incremental data table, when the record quantity in incremental data table is more than 100, execute increasable algorithm, to indigestion disease
Reason neural network model carries out dynamic corrections；
Step (8), repeat step (3)～(7).
The input layer of the neural network model of the present invention is 16 nodes, and hidden layer number is 33, and output layer is 1 node
(i.e. dyspeptic probability), extracts 400000 records from indigestion daily data database table and is trained, every
Record is 16 dimensional vectors, all data before use first through normalized so as to numerical value is interval in [0,1], then hold
Row following steps are trained to neural network model：
1) one 16 dimensional vector of input, to neural network model, calculate all of weight vector in neural network model defeated to this
Enter the distance of 16 dimensional vectors, closest neuron is as won neuron, its computing formula is as follows：
Wherein：WkIt is the weight vector of triumph neuron, | | ... | | for Euclidean distance；
2) weight vector of the neuron in adjustment triumph neuron and triumph neuron field, formula is as follows：
Wherein：WjT () is neuron；Wj(t+1) weight vector before being adjustment and after adjustment；J belongs to triumph neuron neck
Domain；α (t) is learning rate, and it is as the function that the increase of iterations is gradually successively decreased, and span is [01], through multiple
It is 0.62 that Optimal learning efficiency is chosen in experiment；DjIt is the distance of neuron j and triumph neuron；σ (t) is as the letter that the time successively decreases
Number；Iteration all input n-dimensional vectors is input in neural network model and is trained each time, when the iteration reaching regulation
After number of times, neural network model training terminates.
Inspection result is sent back the lattice of the object information of server by the wired home indigestion care appliances of the present invention
Formula is：{ checking whether correct, blood glucose value }, server, after receiving object information, judges whether inspection result is correct.
The increasable algorithm carrying out dynamic corrections to indigestion pathology neural network model of the present invention is：
Vectorial for every in incremental data table V { V1,V2,…,Vn, it is sent in neural network model learning function
Row study, learning procedure is as follows：
1) first to output layer, each weight vector is assigned little random number and is done normalized, then utilizes input mode vector V
Mean value Avg (V), be initialized as the weights of unique neuron in the 0th layer of neural network model, and be set to win nerve
Unit, calculates its quantization error QE；
2) expand out 2 × 2 structures SOM from the 0th layer of neuron, and its level identities Layer is set to 1；
3) for each 2 × 2 structure SOM subnet expanded out in Layer layer, initialize the power of this 4 neurons
Value；The input vector set Ci of i-th neuron is set to sky, main label is set to NULL, the main label ratio r of neuron ii
It is set to 0；The abnormity early warning data vector V of new SOM inherits the triumph input vector set VX of his father's neuron；
4) select a vectorial VX from VXiDo following judgement：
If VXiFor the data of not tape label, then calculate its Euclidean distance with each neuron, chosen distance is the shortest
Neuron is as triumph neuron；
If VXiFor the data of tape label, then select main label and VXiLabel is identical and riThe maximum neuron of value is made
For triumph neuron, update this triumph neuron main label；
If can not find main label and VXiLabel identical neuron, then find and VXiClosest neuron i makees
For triumph neuron；
5) weights of neuron in triumph neuron and its neighborhood are adjusted, update the vectorial set W=W ∪ that wins
{VXi, calculate main label, the main label ratio r of triumph neuroniWith comentropy EiIf. not up to predetermined frequency of training,
Go to step 4)；
6) quantization error QE of each neuron in this neural network model after calculating is adjustedi, neuronal messages entropy Ei
With the average quantization error MQE of subnet, formula is as follows：
Wherein：WiFor the weight vector of neuron i, CiThe set constituting for all input vectors being mapped to neuron i；
Wherein：niRepresent to fall that label is the number of samples of i on neuron, m represents to fall label data on neuron
Sum, T represents to fall the sample label species set on neuron；
Then judge：
If MQE>QE × threshold value q of father node, wherein q=0.71, then insert a line neuron in this SOM, turn step
Rapid 4)；
If Ei>The E of father nodei× threshold value p, wherein p=0.42, then grow one layer of new subnet from this neuron, will
The subnet newly growing increases in the subnet queue of Layer+1 layer；
If being not inserted into new neuron in SOM also do not grow new subnet, illustrate that the training of this subnet completes；
7) for all 2 × 2 structures SOM of the Layer+1 layer newly expanded out, iteration operating procedure 3)～5) to it again
It is trained, until neural network model no longer produces new neuron and new layering, whole training terminates.
If the user of the present invention includes health check-up by other means and checks oneself, learn that oneself has suffered from indigestion, and intelligence
The attention device of energy family indigestion care appliances does not warn then it represents that wired home indigestion care appliances judge to be forbidden
Really, now execution step (6)～(7), wired home indigestion care appliances are sent to object information on server.
Present invention also offers a kind of forecasting system of described indigestion Forecasting Methodology, including intelligent monitoring device, intelligence
Energy device data acquisition device, server and wired home indigestion care appliances, described intelligent monitoring device and described intelligence
Device data acquisition device is connected, and described smart machine data acquisition unit is passed through communication device one and led to described server network
News, described wired home indigestion care appliances pass through communication device two and described server network communication.
It is provided with attention device on the described wired home indigestion care appliances of the present invention.
The described intelligent monitoring device of the present invention includes Intelligent worn device, Intelligent water cup, Intelligent weight claim, intelligent closestool
With Intelligent light sensing equipment etc..
The present invention by neural network model train predict a large amount of patient in hospital pathological data, find indigestion pathology with
Indigestion earlier life variations in detail, clinical symptoms, examination criteria value, people at highest risk's feature, patrolling between this several causes of disease
Collect association and variable, ultimately form the indigestion pathology neural network model to indigestion illness probability Accurate Prediction, this
By gathering user's daily life data, the periodicity of its data of active analysis, regularity are eventually through indigestion disease for invention
Reason Neural Network model predictive user suffers from indigestion probability, reminds user's instant hospitalizing and pre- in the way of visual effect
Anti-.
All data of the present invention preserve to server, can significantly save calculating cost, hardware configuration is low, thus selling
Valency is also low.
The present invention carries communication device one and communication device two, by wifi from the internet that is dynamically connected, and can protect for a long time
Hold online.Various intelligent monitoring devices can easily access present device by modes such as network or bluetooths, sets in acquisition
The daily life data of the monitoring of intelligent monitoring device, the data that therefore present device obtains can automatically be uploaded after standby mandate
It is real-time, accurate, polynary.
Because everyone physical trait is different, the data characteristics being shown during indigestion morbidity also can be different.
Therefore conventional is not high by neural network prediction dyspeptic method accuracy rate.The present invention is directed to each equipment user and sets up
Train the neural network model for this user, running after a period of time, by producing to measure nerve is being made to this user
Network Prediction Model, accuracy rate is greatly improved.
When neural network model is judged by accident, error message be will be feedbacked to service by wired home indigestion care appliances
Device, for this user's dynamic corrections neural network model, when similar characteristics data in this user next, will not miss again
Sentence.Therefore, with the increase of use time, the judgement of the wired home indigestion care appliances of the present invention will be more and more accurate
Really.
General principle, principal character and the advantages of the present invention of the present invention have been shown and described above.The technology of the industry
, it should be appreciated that the present invention is not restricted to the described embodiments, the simply explanation described in above-described embodiment and specification is originally for personnel
Invention principle, without departing from the spirit and scope of the present invention the present invention also have various changes and modifications, these change
Change and improvement both falls within scope of the claimed invention.Claimed scope by appending claims and its
Equivalent defines.
Claims (8)
1. a kind of indigestion Forecasting Methodology based on increment type neural network model is it is characterised in that comprise the steps：
Step (1), obtain hospital indigestion and cure the disease etiology and pathology data source and the daily monitoring data of patient, thus setting up digestion
Bad daily data database；
Step (2), the daily data database of indigestion set up according to step (1) are off-line manner to neural network model
It is trained, to obtain the indigestion pathology neural network model training；
Step (3), by intelligent monitoring device, the daily life data of user is acquired, and will collection daily life number
According to sending to server, server preserves the daily life data of user to the daily data logger of user；
Step (4), from the daily data logger of user, extract same day data, form n-dimensional vector, and normalizing is done to n-dimensional vector
Carry out indigestion probabilistic forecasting in the indigestion pathology neural network model training in input step (2) after change process,
Obtain indigestion probability, server sends indigestion probability to wired home indigestion care appliances；
After step (5), the indigestion probability of wired home indigestion care appliances the reception server transmission, judge digestion not
Whether good probable value is more than 0.5, if greater than 0.5, is then judged to that this user obtained indigestion, attention device warns to remind use
Family, if less than 0.5, is then judged to that this user does not obtain indigestion；
Step (6), when user is judged to indigestion, user voluntarily removes examination in hospital, and by inspection result pass through intelligence
Server can be sent back by family's indigestion care appliances, server judges whether inspection result is correct, if inspection result is wrong
By mistake, then explanation indigestion pathology Neural Network model predictive is inaccurate, if inspection result is correct, indigestion disease is described
Reason Neural Network model predictive is accurate；
Step (7), when inspection result mistake, from the daily data logger of user extract m days in record preserve to increment
In tables of data, when the record quantity in incremental data table is more than h bar, execute increasable algorithm, to indigestion pathology nerve
Network model carries out dynamic corrections；
Step (8), repeat step (3)～(7).
2. a kind of indigestion Forecasting Methodology based on increment type neural network model according to claim 1, its feature
It is, the input layer of neural network model is n node, and hidden layer number is n*2+1, and output layer is 1 node, from digestion not
Extract k bar record in good daily data database table to be trained, every record is a n-dimensional vector, all data are using
Through normalized so as to numerical value is interval in [0,1], then execution following steps are trained to neural network model for front elder generation：
1) one n-dimensional vector of input, to neural network model, calculates all of weight vector in neural network model and ties up to this input n
The distance of vector, closest neuron is as won neuron, and its computing formula is as follows：
Wherein：WkIt is the weight vector of triumph neuron, | | ... | | for Euclidean distance；
2) weight vector of the neuron in adjustment triumph neuron and triumph neuron field, formula is as follows：
Wherein：WjT () is neuron；Wj(t+1) weight vector before being adjustment and after adjustment；J belongs to triumph neuron field；a
T () is learning rate, it is as the function that the increase of iterations is gradually successively decreased, and span is [0 1], through many experiments
Choosing Optimal learning efficiency is 0.62；DjIt is the distance of neuron j and triumph neuron；σ (t) is as the function that the time successively decreases；
Iteration all input n-dimensional vectors is input in neural network model and is trained each time, when the iteration time reaching regulation
After number, neural network model training terminates.
3. a kind of indigestion Forecasting Methodology based on increment type neural network model according to claim 1, its feature
It is, the form that inspection result is sent back the object information of server by wired home indigestion care appliances is：{ inspection is
No correct, blood glucose value }, server, after receiving object information, judges whether inspection result is correct.
4. a kind of indigestion Forecasting Methodology based on increment type neural network model according to claim 1, its feature
It is, the increasable algorithm carrying out dynamic corrections to indigestion pathology neural network model is：
Vectorial for every in incremental data table V { V1,V2,…,Vn, it is sent in neural network model learning function and learned
Practise, learning procedure is as follows：
1) first to output layer, each weight vector is assigned little random number and is done normalized, then utilizes the flat of input mode vector V
Average Avg (V), is initialized as the weights of unique neuron in the 0th layer of neural network model, and is set to triumph neuron, meter
Calculate its quantization error QE；
2) expand out 2 × 2 structures SOM from the 0th layer of neuron, and its level identities Layer is set to 1；
3) for each 2 × 2 structure SOM subnet expanded out in Layer layer, the weights of this 4 neurons are initialized；Will
The input vector set Ci of i-th neuron is set to sky, and main label is set to NULL, the main label ratio r of neuron iiIt is set to
0；The abnormity early warning data vector V of new SOM inherits the triumph input vector set VX of his father's neuron；
4) select a vectorial VX from VXiDo following judgement：
If VXiFor the data of not tape label, then calculate its Euclidean distance with each neuron, chosen distance nerve the shortest
Unit is as triumph neuron；
If VXiFor the data of tape label, then select main label and VXiLabel is identical and riThe maximum neuron of value is as obtaining
Victory neuron, updates this triumph neuron main label；
If can not find main label and VXiLabel identical neuron, then find and VXiClosest neuron i is as obtaining
Victory neuron；
5) weights of neuron in triumph neuron and its neighborhood are adjusted, update the vectorial set W=W ∪ { VX that winsi,
Calculate main label, the main label ratio r of triumph neuroniWith comentropy EiIf. not up to predetermined frequency of training, go to step
4)；
6) quantization error QE of each neuron in this neural network model after calculating is adjustedi, neuronal messages entropy EiAnd son
The average quantization error MQE of net, formula is as follows：
Wherein：WiFor the weight vector of neuron i, CiThe set constituting for all input vectors being mapped to neuron i；
Wherein：niRepresent to fall that label is the number of samples of i on neuron, m represents to fall the total of label data on neuron
Number, T represents to fall the sample label species set on neuron；
Then judge：
If MQE>QE × threshold value q of father node, wherein q=0.71, then insert a line neuron in this SOM, go to step 4)；
If Ei>The E of father nodei× threshold value p, wherein p=0.42, then grow one layer of new subnet from this neuron, will be new
The subnet growing increases in the subnet queue of Layer+1 layer；
If being not inserted into new neuron in SOM also do not grow new subnet, illustrate that the training of this subnet completes；
7) for all 2 × 2 structures SOM of the Layer+1 layer newly expanded out, iteration operating procedure 3)～5) it is re-started
Training, until neural network model no longer produces new neuron and new layering, whole training terminates.
5. a kind of indigestion Forecasting Methodology based on increment type neural network model according to claim 1, its feature
It is, if user includes health check-up by other means and checks oneself, learns that oneself has suffered from indigestion, and wired home digests not
The attention device of good care appliances does not warn then it represents that wired home indigestion care appliances judge inaccurate, now executes
Step (6)～(7), wired home indigestion care appliances are sent to object information on server.
6. a kind of forecasting system of indigestion Forecasting Methodology described in employing claim 1～6 is it is characterised in that include intelligence
Monitoring device, smart machine data acquisition unit, server and wired home indigestion care appliances, described intelligent monitoring device
It is connected with described smart machine data acquisition unit, described smart machine data acquisition unit passes through communication device one and described service
Device network communication, described wired home indigestion care appliances pass through communication device two and described server network communication.
7. according to claim 7 indigestion Forecasting Methodology forecasting system it is characterised in that described wired home digestion
It is provided with attention device on bad care appliances.
8. according to claim 7 the forecasting system of indigestion Forecasting Methodology it is characterised in that described intelligent monitoring device
Claim including Intelligent worn device, Intelligent water cup, Intelligent weight, intelligent closestool and Intelligent light sensing equipment.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610860220.0A CN106446549A (en) | 2016-09-28 | 2016-09-28 | Prediction method and prediction system for dyspepsia based on incremental type neural network model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610860220.0A CN106446549A (en) | 2016-09-28 | 2016-09-28 | Prediction method and prediction system for dyspepsia based on incremental type neural network model |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106446549A true CN106446549A (en) | 2017-02-22 |
Family
ID=58171174
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610860220.0A Pending CN106446549A (en) | 2016-09-28 | 2016-09-28 | Prediction method and prediction system for dyspepsia based on incremental type neural network model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106446549A (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1477581A (en) * | 2003-07-01 | 2004-02-25 | �Ϻ���ͨ��ѧ | A Predictive Modeling Approach for Computer-Aided Medical Diagnosis |
| US20040054358A1 (en) * | 2002-03-28 | 2004-03-18 | Cox Ian G. | System and method for predictive ophthalmic correction |
| CN102647292A (en) * | 2012-03-20 | 2012-08-22 | 北京大学 | A Method of Intrusion Detection Based on Semi-Supervised Neural Network Model |
| CN105118010A (en) * | 2015-09-30 | 2015-12-02 | 成都信汇聚源科技有限公司 | Chronic disease management method with functions of real-time data processing and real-time information sharing and life style intervention information |
-
2016
- 2016-09-28 CN CN201610860220.0A patent/CN106446549A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040054358A1 (en) * | 2002-03-28 | 2004-03-18 | Cox Ian G. | System and method for predictive ophthalmic correction |
| CN1477581A (en) * | 2003-07-01 | 2004-02-25 | �Ϻ���ͨ��ѧ | A Predictive Modeling Approach for Computer-Aided Medical Diagnosis |
| CN102647292A (en) * | 2012-03-20 | 2012-08-22 | 北京大学 | A Method of Intrusion Detection Based on Semi-Supervised Neural Network Model |
| CN105118010A (en) * | 2015-09-30 | 2015-12-02 | 成都信汇聚源科技有限公司 | Chronic disease management method with functions of real-time data processing and real-time information sharing and life style intervention information |
Non-Patent Citations (1)
| Title |
|---|
| MEHMED KANTARDZIC: "《数据挖掘：概念、模型、方法和算法》", 31 January 2013 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106407699A (en) | Coronary heart disease prediction method and prediction system based on incremental neural network model | |
| CN106250712A (en) | A kind of ureteral calculus Forecasting Methodology based on increment type neural network model and prognoses system | |
| CN106355035A (en) | Pneumonia prediction method and prediction system based on incremental neural network model | |
| CN106446552A (en) | Prediction method and prediction system for sleep disorder based on incremental neural network model | |
| CN106446560A (en) | Hyperlipidemia prediction method and prediction system based on incremental neural network model | |
| CN112037925A (en) | LSTM algorithm-based early warning method for newly-released major infectious diseases | |
| CN106384013A (en) | Incremental neural network model-based type-II diabetes prediction method and prediction system | |
| CN106384012A (en) | Incremental neural network model-based allergic dermatitis prediction method and prediction system | |
| CN106202986A (en) | A kind of tonsillitis Forecasting Methodology based on increment type neural network model and prognoses system | |
| CN106650206A (en) | Prediction method of high blood pressure based on incremental neural network model and prediction system | |
| CN106384005A (en) | Incremental neural network model-based depression prediction method and prediction system | |
| CN106384008A (en) | Incremental neural network model-based allergic rhinitis prediction method and prediction system | |
| CN109657907B (en) | Quality control method and device for geographical national condition monitoring data and terminal equipment | |
| CN106355034A (en) | Sub-health prediction method and prediction system based on incremental neural network model | |
| CN106295238A (en) | A kind of hypertensive nephropathy Forecasting Methodology based on increment type neural network model and prognoses system | |
| Trivedi et al. | CARE: IoT enabled cow health monitoring system | |
| CN106407694A (en) | Neurasthenia prediction method and prediction system based on incremental neural network model | |
| CN106407693A (en) | Hepatitis B prediction method and prediction system based on incremental neural network model | |
| CN106446561A (en) | Incremental neural network model based urticaria prediction method and system | |
| CN106250715A (en) | A kind of chronic pharyngolaryngitis Forecasting Methodology based on increment type neural network model and prognoses system | |
| CN106384009A (en) | Incremental neural network model-based HIV prediction method and prediction system | |
| CN106372442A (en) | Dental ulcer prediction method and system based on incremental neural network model | |
| CN106446550A (en) | Cold prediction method and system based on incremental neutral network model | |
| CN106446551A (en) | Incremental neural network model based chronic gastroenteritis prediction method and system | |
| CN106339605A (en) | Colonitis prediction method and colonitis prediction system based on incremental nerve network model |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20170222 |
|
| WD01 | Invention patent application deemed withdrawn after publication |