CN105708443A - Electrocardio perception diagnosis method based on body area network - Google Patents
Electrocardio perception diagnosis method based on body area network Download PDFInfo
- Publication number
- CN105708443A CN105708443A CN201610053617.9A CN201610053617A CN105708443A CN 105708443 A CN105708443 A CN 105708443A CN 201610053617 A CN201610053617 A CN 201610053617A CN 105708443 A CN105708443 A CN 105708443A
- Authority
- CN
- China
- Prior art keywords
- data
- ecg
- file
- hypercube
- electrocardiogram
- 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
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
- A61B5/316—Modalities, i.e. specific diagnostic methods
- A61B5/318—Heart-related electrical modalities, e.g. electrocardiography [ECG]
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/0002—Remote monitoring of patients using telemetry, e.g. transmission of vital signals via a communication network
- A61B5/0004—Remote monitoring of patients using telemetry, e.g. transmission of vital signals via a communication network characterised by the type of physiological signal transmitted
- A61B5/0006—ECG or EEG signals
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/0002—Remote monitoring of patients using telemetry, e.g. transmission of vital signals via a communication network
- A61B5/0015—Remote monitoring of patients using telemetry, e.g. transmission of vital signals via a communication network characterised by features of the telemetry system
- A61B5/002—Monitoring the patient using a local or closed circuit, e.g. in a room or building
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
- A61B5/30—Input circuits therefor
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Medical Informatics (AREA)
- Surgery (AREA)
- Veterinary Medicine (AREA)
- Biophysics (AREA)
- Pathology (AREA)
- Biomedical Technology (AREA)
- Heart & Thoracic Surgery (AREA)
- Public Health (AREA)
- Molecular Biology (AREA)
- Physics & Mathematics (AREA)
- Animal Behavior & Ethology (AREA)
- General Health & Medical Sciences (AREA)
- Computer Networks & Wireless Communication (AREA)
- Physiology (AREA)
- Cardiology (AREA)
- Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
- Medical Treatment And Welfare Office Work (AREA)
Abstract
The invention relates to an electrocardio perception diagnosis method based on a body area network. The method comprises the following steps that one terminal of a module is connected with the lower part of the left chest, and the other terminal is connected with the upper part of the right chest; a USB (Universal Serial Bus) interface of USB electrocardio collecting equipment is connected with a mobile phone or a PC (Personal Computer); a system zip file is decompressed into a specified document by system operation, electrocardio ECG.exe is clicked, and a system operates; START equipment is clicked; CONNECT is clicked, and external equipment is connected; after a collected electrocardio data graph is stable, RECORD is clicked, and the collected data is recorded and stored; after the data collection is completed, SSTOP is clicked, and recording data is customized; the collected data is stored in a specified position; DIAGNOSE is clicked, and a diagnosis result is given out; and a file is input/output for providing relevant information of a data file built, updated or accessed by the operation.
Description
Technical field
The present invention relates to electrocardio sensory perceptual system field, particularly to a kind of heart based on body area network
Inductance knows diagnostic method.
Background technology
Along with progress and the development of human civilization of science and technology, people increasingly pay close attention to the strong of self
Health and condition, based on this, increasing physical sign sensing detection equipment is manufactured,
Not only in hospital, people have been able to the electrocardio awareness apparatus by wearing, by mobile phone or
Other display devices, understand the heart state etc. of oneself in real time.
But although electrocardio perception field product is various at present, but because acquisition precision and later stage
Processing method problem, is commonly present signal disturbing, distortion, and fuzzy etc. various shortcomings, result is past
Toward accurate not, also there is the biggest room for promotion.
Therefore it provides a kind of acquisition precision is high, reducing signal disturbing, distortion, fuzzy lacks
The electrocardio sensory perceptual system of point just becomes particularly important.
Summary of the invention
For solving above-mentioned problem of the prior art, the present invention provides a kind of heart based on body area network
Inductance knows diagnostic method, it is possible to realize detecting in real time the electrocardio related data of user, to evaluate and test visitor
The health status at family, fineness is high, and precise and high efficiency has good commercial application value.
For reaching above-mentioned purpose, the technical scheme is that
A kind of electrocardio perception diagnostic method based on body area network, comprises the steps:
Step one: the connection of hardware device
Module one terminal is connected on below left breast, and another terminal is connected on above right breast；And by USB
The USB interface of electrocardiogram acquisition equipment is connected with mobile phone or pc；
Step 2: system operates
1) system compresses bag is unziped in the document specified, and click on electrocardio-ECG.exe, run
System；
2) Start equipment is clicked on；
3) click on Connect, connect external equipment；
4) after the electrocardiogram (ECG) data figure arrived to be collected is stable, click on Record, record and preserve and collect
Data；
5) SStop, customization record data are clicked on after gathering data；
6) data collected are saved in the position specified；
7) click on " Diagnose ", provide diagnostic result；
Step 3: input/output file
There is provided the data file set up by this operation, update or access for information about, such as:
1) E: Data MIT-BIH-Data this be case file folder
2) E: Data BMD101-Data this be electrocardiogram acquisition log file folder
3) E: Data Data-Sink this be the interim buffer culture of data base
Output result literary composition section.
Further, in described step 2, the acquisition function of electrocardiogram (ECG) data is real by following step
Existing:
Amplified the electrocardiosignal after filtration module amplifies by front end electrocardiosignal to be directly fed to
The P0.23 pin of LPC2478 chip carries out A/D conversion, converted after data be without symbol
32 bit data format, put it in caching array and give USB and store, and remain to
LCD shows output, and the process of implementing is:
First, need to create two data buffer area " GcWritFileData [DATA_N] " and
" GcReadFileData [DATA_N] " is as " written document relief area " and " reads file buffering
District ", initialize mC/OS II operating system, create the task task 0 for processing A/D conversion
And it is used for the task task 1 that LCD shows, start multitask environment, first enter in Task0
The initialization of row hardware platform, arranging P0.23 is AIN0 [0] function, as A/D conversion
Input pin, carries out ADC setting, arranges change over clock etc., uses and directly initiates ADC
Conversion, the 2.5V voltage that the reference voltage carrying out changing provides for accurate constant pressure source, finally turn
Change result to preserve to " written document relief area ".Then initialize USB HOST, and create literary composition
Part system task " OSFileTask ", by " OSFileOpen " function creation and open one
The file of named " ECD.dat ", by " OSFileWrite " function by " GcWritFileData "
The data of written document relief area are written in disk, and whether judge written document by its return value
Success, re-uses " OSFileRead " function by " GcWritFileData " after completing written document
The reading and writing data of written document relief area reads filebuf to " GcReadFileData ", then leads to
Cross write, read filebuf data relatively determine that write file data is whether correct, so far,
Complete to receive data from A/D conversion and be transferred to the process of USB.Then by our reading
Electrocardiogram (ECG) data in disk is read by program fetch.
Further, in described step 2, the diagnostic function of electrocardiogram (ECG) data is real by following method
Existing:
Data-interface: from electrocardio-data collection sample data；Software further processes,
By the signal gathered is filtered noise reduction process, finally obtain electrocardiosignal；
Low-pass filtering:
Feature extraction algorithm:
PCA is principal component analysis, is mainly used in Data Dimensionality Reduction, for the spy of a series of examples
Levying the multi-C vector of composition, some element in multi-C vector itself does not has distinction, such as certain
Individual element is all 1 in all of example, or little with 1 gap, then this element is originally
Body does not just have distinction, does feature with it and distinguishes, and contribution can be the least；So our mesh
Be those dimensions looking for those the big elements of change, i.e. variance big, and get rid of those changes not
Big dimension, so that what feature stayed is all " fine work ", and amount of calculation have also been smaller；
For the feature of a k dimension, the every one-dimensional characteristic being equivalent to it with other dimensions is all
Orthogonal: being equivalent in multidimensional coordinate system, coordinate axes is all vertical, by changing these
The coordinate system of dimension, so that this feature is big in some dimension upside deviation, and ties up upside deviation at some
The least；One the 45 degree ellipses tilted, in the first coordinate system, if according to x, y-coordinate comes
Projection, the attribute of x and y of these points is difficult to distinguish them, because they are in x, y-axis
The variance of upper changes in coordinates is all similar, and we cannot sentence according to this certain x attribute put
Which this point disconnected is, and if coordinate axes is rotated, with transverse as x-axis, then oval
Distribution on major axis is long, and variance is big, and the distribution on short axle is short, and variance is little, institute
Can consider only to retain the major axis attribute of these points, distinguish the point on ellipse, so, district
Dividing property ratio x, the method for y-axis to be got well；
So our way is exactly to try to achieve the projection matrix of a k dimensional feature, this projects square
Feature can be dropped to low-dimensional from higher-dimension by battle array；Projection matrix can also be called transformation matrix；New
The necessary each dimension of low dimensional feature is orthogonal, and characteristic vector is all orthogonal；By seeking sample matrix
Covariance matrix, then obtain the characteristic vector of covariance matrix, these characteristic vectors just may be used
To constitute this projection matrix；The eigenvalue of covariance matrix is depended in the selection of characteristic vector
Size.
First, during system initialization, we are to MIT-BIH-Data case file
Carrying out pretreatment, method is as follows:
The first step: read MIT-BIH-Data case literary composition, set up the patient chart of following structure:
Second step: utilize previously described feature extraction algorithm to extract every electrocardio in patient chart
The characteristic information of data, obtains characteristic information patient chart, the structure of table following (n is much smaller than M):
3rd step: electrocardiogram (ECG) data space is divided into multiple non-overlapping hypercube by us,
And distribute a unique numbering (ID) to each data hypercube；Described hypercube is
Following structure:
Assuming that the characteristic information of electrocardiogram (ECG) data has n data, we are by this n characteristic information
It is considered as n dimension data, then a hypercube then includes n interval [a1,b1),
[a2,b2),…[an,bn) and an ID, wherein interval [ai,bi) represent that this hypercube only deposits i-th
Individual characteristic diMeet aidi<biCharacteristic；
4th step: if a characteristic D=(d1,d2,…,dn) meet d1[a1,b1)and
d2[a2,b2)and…and dn[an,bn), then ecg characteristics data D are integrated into this hypercube,
And set up a mapping table, the structure of described mapping table is:
5th step: remove empty hypercube, calculate the center-of-mass coordinate of each non-NULL hypercube,
And set up data hypercube concordance list based on center-of-mass coordinate, shown center-of-mass coordinate is asked exactly
The meansigma methods of the ecg characteristics data in same hypercube；
When the electrocardiogram (ECG) data collected is sent to after noise reduction filtering by data acquisition module
Characteristic extracting module, characteristic extracting module extracts the characteristic information of patient's electrocardiogram (ECG) data；Work as user
When clicking on " Diagnose " button, these characteristic informations as querying condition, are sent by system
To long-range database server, server is examined by following method after receiving querying condition
Disconnected:
The first step: server by utilizing similarity measurement formula, in data hypercube concordance list
Find hypercube corresponding to K nearest center-of-mass coordinate of Distance query condition No. ID, and
Return these No. ID；
Second step: find out k the hypercube determined in the first step in the mapping table No. ID
Corresponding ecg characteristics data；
3rd step: utilize similarity measurement formula to look in the ecg characteristics data that second step finds
To the ecg characteristics data most like with querying condition, and by the disease corresponding to this feature data
Title and clinical manifestation return to user as diagnostic result；
Described similarity measurement formula be conventional Euclidean distance measure formulas, geneva away from
From formula, Minkowski away from, from the similarity measurements such as Pearson's correlation coefficient and cosine similarity
One in amount formula.
Further, the data storage employing MIT-BIH data base of the present invention:
MIT-BIH is the ARR data of the research provided by Massachusetts Institute Technology
Storehouse；That generally acknowledges the most in the world can have three as the ecg database of standard, is the U.S. respectively
The MIT-BIH data base that the Massachusetts Institute of Technology provides, the AHA data of American heart association
Storehouse and Europe AT-T ecg database；Wherein MIT-BIH data base Application comparison in recent years
Extensively；
The data form of MIT-BIH:
MIT-BIH, in order to save file size and memory space, employs self-defining form；
One electrocardiographic recording is made up of three parts:
(1) header file [.hea], storage mode ASCII character character；
(2) data file [.dat], by binary storage, every three bytes store two numbers,
One number 12bit；
(3) comment file [.art], by binary storage；
MIT-BIH data base as reference database, existing electrocardiogram (ECG) data is carried out point
Analysis, sets up the data base belonging to this software；
As long as the electrocardiogram (ECG) data gathered is processed by data, entered quick similar search algorithm and looked into
Ask result；
Further, output file is provided that
Export corresponding symptom, and the cardiac-related diseases that may cause.
Relative to prior art, the invention have the benefit that
Based on history case storehouse, the function of cardioelectric monitor and diagnosis automatically can be realized, greatly simultaneously
Today of data age, utilizing history case data to carry out medical diagnosis will be following becoming
Gesture；, register difficulty busy in hospital, the today of difficulty of seeking medical advice, portable cardiac monitoring sets with diagnosis
The standby family that can spread to is with individual, it is possible to achieve detects ubiquitously and diagnoses, especially
Health care is provided for middle-aged and elderly people.
The diagnosis capability of native system is more accurate along with improving of case storehouse, current various big hospital
Set up the data center of oneself, utilize the auxiliary diagnosis of big data to become being total to of each hospital
Knowing, this product meets the trend of modern hospital development.
The present invention can realize detecting ubiquitously and diagnosing, it is not necessary to goes to hospital to queue up and seeks medical advice,
Have only to the information according to hospital journals data base, it is achieved care diagnostic.
The hardware device of the present invention uses BMD101, is NeuroSky (god reads science and technology)
Equipment on the sheet of the detection of third generation bio signal and process.BMD101 can gather from uV to
The bio signal of mV.
The electrocardiosignal the collected interference signal that meeting is more in living environment, causes signal mode
Stick with paste, the shortcoming such as distortion.So this product further processes on software, by collection
Signal is filtered noise reduction process, finally obtains electrocardiosignal.
Accompanying drawing explanation
Fig. 1 is the theory diagram of present system.
Fig. 2 is that present system electrocardiosignal amplifies filtration module figure.
Fig. 3: for electrocardio-data collection figure；
Fig. 3 provides effect curve: as it can be seen, the heart that electrocardiogram gathers with large medical equipment
Electricity curve is the same, waveform stabilization.
Fig. 4: for EGC parameter
The EGC parameter data that Fig. 4 is given: as shown in the figure, it is possible to provide concrete EGC parameter
Data, to make medical treatment reference.
Fig. 5 a: detect Fig. 1 for heart patient
Fig. 5 a is the detection figure of heart patient: as it can be seen, this product can detect heart disease
The heart disease that patient is suffered from, and provide corresponding supplemental characteristic.The result of upper figure detection is room property
Premature beat, easily causes angina pectoris, myocardial infarction and coronary heart disease.Average heart rate in 30 seconds is
72。
Fig. 5 b: detect figure for normal person；
Fig. 5 b is the ECG detecting figure of normal person: show according to the result of detection, this diagnostic result
For normal beats, the average heart rate in 30 seconds is 79.Note: normal person's heart rate per minute exists
60---100。
Fig. 6 is the operation step map of present system.
Detailed description of the invention
With detailed description of the invention, technical solution of the present invention is done the most below in conjunction with the accompanying drawings
Describe:
As shown in figures 1 to 6: a kind of electrocardio perception diagnostic method based on body area network, by Pulse Rate
Receiving original pulse data according to processing module, pulse data processing module, electrocardiosignal amplify filter
Mode block, ARM controller are sequentially connected with, and electrocardiosignal is amplified filtration module and amplification filtered
Rear data are transferred to ARM controller, and ARM controller connects display respectively and may move
Memorizer, realizes data by removable memory and stores, show electrocardiogram (ECG) data by display.
Further, described electrocardiosignal amplifies filtration module by preamplifier, low-pass filtering
Device, high pass filter, main amplifier, 50Hz wave trap, level boost, A/D converter
Being sequentially connected with, the electrocardiosignal that piezoelectric transducer extracts is successively through preamplifier, low-pass filtering
Device, high pass filter, main amplifier, 50Hz wave trap, level boost, A/D converter
Carry out signal and amplify filtering.
Further, the electrocardiosignal after described electrocardiosignal amplifies filtration module amplification is directly sent
A/D conversion is carried out to the P0.23 pin of LPC2478 chip；Data after converted are nothing
Symbol 32 bit data format, puts it in caching array and gives USB and store, and stay
Treat to LCD display output.
The method of work of present system includes:
The operation module interfaces of system is as shown in the table:
According to operational objective prompting operation.
Table 1
Step one: the connection of hardware device
Module redness terminal is connected on below left breast, and white terminal is connected on above right breast.And by the USB heart
The USB interface of electricity collecting device is connected with computer or mobile phone, pc etc..
Step 2: system operates
2) system compresses bag is unziped in the document specified, and click on electrocardio-ECG.exe, run
System.
2) Start equipment is clicked on
3) click on Connect, connect external equipment
4) after the electrocardiogram (ECG) data figure arrived to be collected is stable, click on Record, record and preserve and collect
Data
5) SStop, customization record data are clicked on after gathering data
6) data collected are saved in the position specified
7) click on " Diagnose ", provide diagnostic result
Step 3: input/output file
There is provided the data file set up by this operation, update or access for information about, such as:
1) E: Data MIT-BIH-Data this be case file folder
2) E: Data BMD101-Data this be electrocardiogram acquisition log file folder
3) E: Data Data-Sink this be the interim buffer culture of data base
Output literary composition section
Output result literary composition section.
In said method, the acquisition function of electrocardiogram (ECG) data is realized by following step
Amplified the electrocardiosignal after filtration module amplifies by front end electrocardiosignal to be directly fed to
The P0.23 pin of LPC2478 chip carries out A/D conversion, converted after data be without symbol
32 bit data format, put it in caching array and give USB and store, and remain to
LCD shows output, and the process of implementing is:
First, need to create two data buffer area " GcWritFileData [DATA_N] " and
" GcReadFileData [DATA_N] " is as " written document relief area " and " reads file buffering
District ", initialize mC/OS II operating system, create the task task 0 for processing A/D conversion
And it is used for the task task 1 that LCD shows, start multitask environment, first enter in Task0
The initialization of row hardware platform, arranging P0.23 is AIN0 [0] function, as A/D conversion
Input pin, carries out ADC setting, arranges change over clock etc., uses and directly initiates ADC
Conversion, the 2.5V voltage that the reference voltage carrying out changing provides for accurate constant pressure source, finally turn
Change result to preserve to " written document relief area ".Then initialize USB HOST, and create literary composition
Part system task " OSFileTask ", by " OSFileOpen " function creation and open one
The file of named " ECD.dat ", by " OSFileWrite " function by " GcWritFileData "
The data of written document relief area are written in disk, and whether judge written document by its return value
Success, re-uses " OSFileRead " function by " GcWritFileData " after completing written document
The reading and writing data of written document relief area reads filebuf to " GcReadFileData ", then leads to
Cross write, read filebuf data relatively determine that write file data is whether correct, so far,
Complete to receive data from A/D conversion and be transferred to the process of USB.Then by our reading
Electrocardiogram (ECG) data in disk is read by program fetch.
The feature of human ecg signal
Electrocardiosignal belongs to biomedicine signals, has a characteristic that
(1) signal has the feature of near-field detection, leaves the distance that people's body surface is small, the most basic
On can't detect signal；
(2) electrocardiosignal is the faintest, is at most mV magnitude；
(3) belong to low frequency signal, and energy is mainly below hundreds of hertz；
(4) interference is the strongest.Interference is both from organism, such as myoelectricity interference, respiration interference
Deng；Also from vitro, as introduce because of imperfect earth etc. when Hz noise, picking up signal its
His alien cross-talk etc.；
(5) interference signal and the own band overlapping of electrocardiosignal (such as Hz noise etc.).
Based on These characteristics, the design requirement of Acquisition Circuit is:
For the These characteristics of electrocardiosignal, the design to Acquisition Circuit system is analyzed as follows:
(1) signal amplifies is indispensable link, and should be by the width of signal boost to A/D input port
Degree requirement, the i.e., at least magnitude of " V "；
(2) impact of Hz noise should be weakened as far as possible；
(3) the baseline drift problem caused because of breathing etc. it is considered as；
(4) signal frequency is the highest, and passband is typically to meet requirement, it is contemplated that input resistance
The factors such as anti-, linear, low noise.
In said method, the diagnostic function of electrocardiogram (ECG) data is realized by following method:
Data-interface: from electrocardio-data collection sample data.
1. the electrocardiosignal the collected interference signal that meeting is more in living environment, causes
Signal ambiguity, the shortcoming such as distortion.So this product further processes on software, logical
Cross the signal to gathering and be filtered noise reduction process, finally obtain electrocardiosignal.
Low-pass filtering: low-pass filtering (Low-pass filter) is a kind of filter type, and rule is
Low frequency signal can be normal through, exceedes and sets the high-frequency signal of marginal value and be then blocked, weaken.
But the amplitude intercept, weakened then can be according to different frequencies and different filter (mesh
) and change.It is also designated as sometimes high frequency remove filter (high-cut filter) or
The highest removal of person filters (treble-cut filter).Feature extraction algorithm (pca):
PCA (Principal Component Analysis, PCA) is principal component analysis,
It is mainly used in Data Dimensionality Reduction, for the multi-C vector of the feature composition of a series of examples, multidimensional
Some element in vector itself does not has distinction, such as certain element in all of example all
It is 1, or little with 1 gap, then and this element does not inherently have distinction, makes of it
Feature is distinguished, and contribution can be the least.So our purpose is to look for those change elements greatly,
Those dimensions that i.e. variance is big, and get rid of the dimension that those changes are little, so that what feature stayed
It is all " fine work ", and amount of calculation have also been smaller.
For the feature of a k dimension, the every one-dimensional characteristic being equivalent to it with other dimensions is all
Orthogonal (being equivalent in multidimensional coordinate system, coordinate axes is all vertical), then Wo Menke
To change the coordinate system of these dimensions, so that this feature is big in some dimension upside deviation, and at certain
A little dimension upside deviations are the least.Such as, one the 45 degree ellipses tilted, in the first coordinate system, as
Fruit is according to x, and y-coordinate projects, and the attribute of x and y of these points is difficult to distinguish them,
Because their variance of changes in coordinates on x, y-axis is all similar, we cannot be according to this point
Certain x attribute judge this point which is, and if by coordinate axes rotate, long with ellipse
Axle is x-axis, then oval distribution on major axis is long, and variance is big, and dividing on short axle
Cloth is short, and variance is little, it is possible to considers only to retain the major axis attribute of these points, distinguishes ellipse
On point, so, distinction to be got well than x, the method for y-axis！
So our way is exactly to try to achieve the projection matrix of a k dimensional feature, this projects square
Feature can be dropped to low-dimensional from higher-dimension by battle array.Projection matrix can also be called transformation matrix.New
The necessary each dimension of low dimensional feature is orthogonal, and characteristic vector is all orthogonal.By seeking sample matrix
Covariance matrix, then obtain the characteristic vector of covariance matrix, these characteristic vectors just may be used
To constitute this projection matrix.The eigenvalue of covariance matrix is depended in the selection of characteristic vector
Size.
First, during system initialization, we are to MIT-BIH-Data case file
Carrying out pretreatment, method is as follows:
The first step: read MIT-BIH-Data case literary composition, set up the patient chart of following structure:
Second step: utilize previously described feature extraction algorithm to extract every electrocardio in patient chart
The characteristic information of data, obtains characteristic information patient chart, the structure of table following (n is much smaller than M):
3rd step: electrocardiogram (ECG) data space is divided into multiple non-overlapping hypercube by us,
And distribute a unique numbering (ID) to each data hypercube；Described hypercube is
Following structure:
Assuming that the characteristic information of electrocardiogram (ECG) data has n data, we are by this n characteristic information
It is considered as n dimension data, then a hypercube then includes n interval [a1,b1),
[a2,b2),…[an,bn) and an ID, wherein interval [ai,bi) represent that this hypercube only deposits i-th
Individual characteristic diMeet ai<di<biCharacteristic；
4th step: if a characteristic D=(d1,d2,…,dn) meet d1[a1,b1)and
d2[a2,b2)and…and dn[an,bn), then ecg characteristics data D are integrated into this hypercube,
And set up a mapping table, the structure of described mapping table is:
5th step: remove empty hypercube, calculate the center-of-mass coordinate of each non-NULL hypercube,
And set up data hypercube concordance list based on center-of-mass coordinate, shown center-of-mass coordinate is asked exactly
The meansigma methods of the ecg characteristics data in same hypercube；
When the electrocardiogram (ECG) data collected is sent to after noise reduction filtering by data acquisition module
Characteristic extracting module, characteristic extracting module extracts the characteristic information of patient's electrocardiogram (ECG) data；Work as user
When clicking on " Diagnose " button, these characteristic informations as querying condition, are sent by system
To long-range database server, server is examined by following method after receiving querying condition
Disconnected:
The first step: server by utilizing similarity measurement formula, in data hypercube concordance list
Find hypercube corresponding to K nearest center-of-mass coordinate of Distance query condition No. ID, and
Return these No. ID；
Second step: find out k the hypercube determined in the first step in the mapping table No. ID
Corresponding ecg characteristics data；
3rd step: utilize similarity measurement formula to look in the ecg characteristics data that second step finds
To the ecg characteristics data most like with querying condition, and by the disease corresponding to this feature data
Title and clinical manifestation return to user as diagnostic result；
Described similarity measurement formula be conventional Euclidean distance measure formulas, geneva away from
From formula, Minkowski away from, from the similarity measurements such as Pearson's correlation coefficient and cosine similarity
One in amount formula.
Benefit: first, the sample value data volume of electrocardio-data collection reaches thousands of up to ten thousand, so
One section of electrocardiogram (ECG) data is had only to the data that can represent ecg characteristics to extract, for
Electrocardiogram (ECG) data affects little data, and we will filter out.By PCA algorithm, contribution rate
The data (including 85%) more than 85% extract, it is to avoid the data of redundancy calculate.
Secondly, we using the characteristic information that puts forward as querying condition, in inquiry case database with
The most similar case of given querying condition, as diagnostic result, not only improves diagnostic result essence
Standard, and diagnostic function can be more powerful with constantly improving of case database.Thirdly,
Use the diagnostic method energy real-time response of the present invention, the phase because case database is too big can be avoided
Answer the deficiency of overlong time.
The present invention uses MIT-BIH data base:
MIT-BIH is the ARR data of the research provided by Massachusetts Institute Technology
Storehouse.That generally acknowledges the most in the world can have three as the ecg database of standard, is the U.S. respectively
The MIT-BIH data base that the Massachusetts Institute of Technology provides, the AHA data of American heart association
Storehouse and Europe AT-T ecg database.Wherein MIT-BIH data base Application comparison in recent years
Extensively.
The data form of MIT-BIH:
MIT-BIH, in order to save file size and memory space, employs self-defining form.
One electrocardiographic recording is made up of three parts:
(1) header file [.hea], storage mode ASCII character character.
(2) data file [.dat], by binary storage, every three bytes store two numbers,
One number 12bit.
(3) comment file [.art], by binary storage.
MIT-BIH data base as reference database, existing electrocardiogram (ECG) data is carried out point
Analysis, sets up the data base belonging to this software.
As long as the electrocardiogram (ECG) data gathered is processed by data, entered quick similar search algorithm and looked into
Ask result.
Output file is provided that
Export corresponding symptom, the cardiac-related diseases that simultaneously can cause.
Based on history case storehouse, the function of cardioelectric monitor and diagnosis automatically can be realized, greatly simultaneously
Today of data age, utilizing history case data to carry out medical diagnosis will be following becoming
Gesture；, register difficulty busy in hospital, the today of difficulty of seeking medical advice, portable cardiac monitoring sets with diagnosis
The standby family that can spread to is with individual, it is possible to achieve detects ubiquitously and diagnoses, especially
Health care is provided for middle-aged and elderly people.
The diagnosis capability of native system is more accurate along with improving of case storehouse, current various big hospital
Set up the data center of oneself, utilize the auxiliary diagnosis of big data to become being total to of each hospital
Knowing, this product meets the trend of modern hospital development.
The present invention can realize detecting ubiquitously and diagnosing, it is not necessary to goes to hospital to queue up and seeks medical advice,
Have only to the information according to hospital journals data base, it is achieved care diagnostic.
The above, the only detailed description of the invention of the present invention, but protection scope of the present invention is also
It is not limited to this, any change expected without creative work or replacement, all should contain
Within protection scope of the present invention.Therefore, protection scope of the present invention should be with claims
The protection domain limited is as the criterion.
Claims (5)
1. an electrocardio perception diagnostic method based on body area network, it is characterised in that comprise the steps:
Step one: the connection of hardware device
Module one terminal is connected on below left breast, and another terminal is connected on above right breast；And the USB interface of USB electrocardiogram acquisition equipment is connected with mobile phone or pc；
Step 2: system operates
1) system compresses bag is unziped in the document specified, and click on electrocardio-ECG.exe, run system；
2) Start equipment is clicked on；
3) click on Connect, connect external equipment；
4), after the electrocardiogram (ECG) data figure arrived to be collected is stable, clicks on Record, record and preserve the data collected；
5) SStop, customization record data are clicked on after gathering data；
6) data collected are saved in the position specified；
7) click on " Diagnose ", provide diagnostic result；
Step 3: input/output file
There is provided the data file set up by this operation, update or access for information about, such as:
1) E: Data MIT-BIH-Data this be case file folder
2) E: Data BMD101-Data this be electrocardiogram acquisition log file folder
3) E: Data Data-Sink this be the interim buffer culture of data base
Output result literary composition section.
Method the most according to claim 1, it is characterised in that in described step 2, the acquisition function of electrocardiogram (ECG) data is realized by following step:
Electrocardiosignal after being amplified filtration module amplification by front end electrocardiosignal is directly fed to the P0.23 pin of LPC2478 chip and carries out A/D conversion, data after converted are without symbol 32 bit data format, put it in caching array and give USB and store, and remain to LCD display output, the process of implementing is:
nullFirst，Two data buffer area " GcWritFileData [DATA_N] " and " GcReadFileData [DATA_N] " need to be created as " written document relief area " and " reading filebuf "，Initialize mC/OS II operating system，Create the task task 0 for processing A/D conversion and the task task 1 shown for LCD，Start multitask environment，First the initialization of hardware platform is carried out in Task0，Arranging P0.23 is AIN0 [0] function，Input pin as A/D conversion，Carry out ADC setting，Change over clock etc. is set，Employing directly initiates ADC conversion，The 2.5V voltage that the reference voltage carrying out changing provides for accurate constant pressure source，Last transformation result preserves to " written document relief area ".nullThen USB HOST is initialized，And create file system tasks " OSFileTask "，With " OSFileOpen " function creation the file of opening one named " ECD.dat "，By " OSFileWrite " function, the data of " GcWritFileData " written document relief area are written in disk，And judge that written document is the most successful by its return value，Re-use " OSFileRead " function after completing written document and the reading and writing data of " GcWritFileData " written document relief area to " GcReadFileData " is read filebuf，Again by writing、That reads filebuf data relatively determines that write file data is the most correct，So far，Complete to receive data from A/D conversion and be transferred to the process of USB.Then by our reading program, the electrocardiogram (ECG) data in disk is read.
Method the most according to claim 1, it is characterised in that in described step 2, the diagnostic function of electrocardiogram (ECG) data is realized by following method:
First, during system initialization, we carry out pretreatment to MIT-BIH-Data case file, and method is as follows:
The first step: read MIT-BIH-Data case literary composition, set up patient chart；
Second step: utilize previously described feature extraction algorithm to extract the characteristic information of every electrocardiogram (ECG) data in patient chart, obtains characteristic information patient chart, and wherein n is much smaller than M；
3rd step: electrocardiogram (ECG) data space is divided into multiple non-overlapping hypercube by us, and distributes a unique numbering (ID) to each data hypercube.Described hypercube is following structure:
Assuming that the characteristic information of electrocardiogram (ECG) data has n data, this n characteristic information is considered as n dimension data by us, then a hypercube then includes n interval [a1,b1),[a2,b2),…[an,bn) and an ID, wherein interval [ai,bi) represent that this hypercube only deposits ith feature data diMeet ai<di<biCharacteristic；
4th step: if a characteristic D=(d1,d2,…,dn) meet d1[a1,b1)and d2[a2,b2)and…and dn[an,bn), then ecg characteristics data D are integrated into this hypercube, and set up a mapping table；
5th step: removing empty hypercube, calculate the center-of-mass coordinate of each non-NULL hypercube, and set up data hypercube concordance list based on center-of-mass coordinate, shown center-of-mass coordinate is exactly to seek the meansigma methods of the ecg characteristics data in same hypercube；
When the electrocardiogram (ECG) data collected is sent to characteristic extracting module after noise reduction filtering by data acquisition module, and characteristic extracting module extracts the characteristic information of patient's electrocardiogram (ECG) data；When user clicks on " Diagnose " button, these characteristic informations as querying condition, are sent to long-range database server by system, and server is diagnosed by following method after receiving querying condition:
The first step: server by utilizing similarity measurement formula, find hypercube corresponding to K nearest center-of-mass coordinate of Distance query condition in data hypercube concordance list No. ID, and return these No. ID；
Second step: find out No. ID corresponding ecg characteristics data of k the hypercube determined in the first step in the mapping table；
3rd step: utilize similarity measurement formula to find the ecg characteristics data most like with querying condition in the ecg characteristics data that second step finds, and the disease name corresponding to this feature data and clinical manifestation are returned to user as diagnostic result；
Described similarity measurement formula is that conventional Euclidean distance measure formulas, mahalanobis distance formula, Minkowski are away from, one in the similarity measurement formula such as Pearson's correlation coefficient and cosine similarity.
Method the most according to claim 1, it is characterised in that the data storage employing MIT-BIH data base of the present invention:
MIT-BIH is the ARR data base of the research provided by Massachusetts Institute Technology；That generally acknowledges the most in the world can have three as the ecg database of standard, is the MIT-BIH data base of Massachusetts Institute Technology's offer respectively, the AHA data base of American heart association and Europe AT-T ecg database；Wherein MIT-BIH data base Application comparison in recent years is extensive；
The data form of MIT-BIH:
MIT-BIH, in order to save file size and memory space, employs self-defining form；One electrocardiographic recording is made up of three parts:
(1) header file [.hea], storage mode ASCII character character；
(2) data file [.dat], by binary storage, every three bytes store two numbers, number 12bit；
(3) comment file [.art], by binary storage；
MIT-BIH data base as reference database, existing electrocardiogram (ECG) data is analyzed, sets up the data base belonging to this software；
As long as the electrocardiogram (ECG) data gathered is processed by data, entered quick similar search algorithm Query Result.
Method the most according to claim 1, it is characterised in that output file is provided that
Export corresponding symptom, and the cardiac-related diseases that may cause.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610053617.9A CN105708443A (en) | 2016-01-27 | 2016-01-27 | Electrocardio perception diagnosis method based on body area network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610053617.9A CN105708443A (en) | 2016-01-27 | 2016-01-27 | Electrocardio perception diagnosis method based on body area network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105708443A true CN105708443A (en) | 2016-06-29 |
Family
ID=56155060
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610053617.9A Pending CN105708443A (en) | 2016-01-27 | 2016-01-27 | Electrocardio perception diagnosis method based on body area network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105708443A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112494044A (en) * | 2020-11-09 | 2021-03-16 | 沈阳东软智能医疗科技研究院有限公司 | Fatigue driving detection method and device, readable storage medium and electronic equipment |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1907214A (en) * | 2006-08-18 | 2007-02-07 | 方祖祥 | Portable remote real-time monitor with first-aid and locate function |
| CN102281359A (en) * | 2011-08-31 | 2011-12-14 | 惠州Tcl移动通信有限公司 | Mobile phone with heartbeat detection function and implementation method thereof |
| CN103259859A (en) * | 2013-05-08 | 2013-08-21 | 王天一 | Body area network based on Android and intelligent terminal connection method thereof |
| CN103393418A (en) * | 2013-03-25 | 2013-11-20 | 西安电子科技大学 | Portable electrocardiograph monitoring system |
| CN105125202A (en) * | 2015-07-31 | 2015-12-09 | 苏州玄禾物联网科技有限公司 | Electrocardiogram monitoring system based on low-noise amplifier |
-
2016
- 2016-01-27 CN CN201610053617.9A patent/CN105708443A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1907214A (en) * | 2006-08-18 | 2007-02-07 | 方祖祥 | Portable remote real-time monitor with first-aid and locate function |
| CN102281359A (en) * | 2011-08-31 | 2011-12-14 | 惠州Tcl移动通信有限公司 | Mobile phone with heartbeat detection function and implementation method thereof |
| CN103393418A (en) * | 2013-03-25 | 2013-11-20 | 西安电子科技大学 | Portable electrocardiograph monitoring system |
| CN103259859A (en) * | 2013-05-08 | 2013-08-21 | 王天一 | Body area network based on Android and intelligent terminal connection method thereof |
| CN105125202A (en) * | 2015-07-31 | 2015-12-09 | 苏州玄禾物联网科技有限公司 | Electrocardiogram monitoring system based on low-noise amplifier |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112494044A (en) * | 2020-11-09 | 2021-03-16 | 沈阳东软智能医疗科技研究院有限公司 | Fatigue driving detection method and device, readable storage medium and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| de Albuquerque et al. | Robust automated cardiac arrhythmia detection in ECG beat signals | |
| Wang et al. | An improved convolutional neural network based approach for automated heartbeat classification | |
| CN107788976A (en) | Sleep monitor system based on Amplitude integrated electroencephalogram | |
| Li | Wavelets for electrocardiogram: overview and taxonomy | |
| CN103027675B (en) | Novel portable three-lead real-time wireless electrocardiogram monitoring system and analyzing method | |
| CN102274020A (en) | Low-power consumption portable electrocardiograph monitor and control method thereof | |
| CN103584852A (en) | Personalized electrocardiogram intelligent auxiliary diagnosis device and method | |
| CN109009073A (en) | Atrial fibrillation detection device and storage medium | |
| CN113876335B (en) | Electrocardiogram and heart beat classification method, device, electronic equipment and medium | |
| Wang et al. | A wearable ECG monitor for deep learning based real-time cardiovascular disease detection | |
| CN108175382B (en) | CPC-based non-contact sleep evaluation method and device | |
| Gupta | Wavelet transform and vector machines as emerging tools for computational medicine | |
| CN109044340B (en) | Electrocardiogram data classification method and device, electronic equipment and storage medium | |
| CN109222964A (en) | atrial fibrillation detection device and storage medium | |
| CN113080991A (en) | Method, system, diagnosis device and storage medium for predicting and diagnosing heart failure based on CNN model and LSTM model | |
| TWI288875B (en) | Multiple long term auto-processing system and method thereof | |
| Martinho et al. | Towards continuous user recognition by exploring physiological multimodality: An electrocardiogram (ECG) and blood volume pulse (BVP) approach | |
| Gupta et al. | A simplistic and novel technique for ECG signal pre-processing | |
| Gupta et al. | Pre-processing based ECG signal analysis using emerging tools | |
| CN105708443A (en) | Electrocardio perception diagnosis method based on body area network | |
| CN105496399A (en) | Electrocardio-sensing diagnosis system based on body area network | |
| CN206852594U (en) | A kind of device that user characteristics is obtained according to human-body biological electromagnetic wave | |
| WO2021148044A1 (en) | Alarm warning method and device for electrocardiograph, electrocardiograph, and storage apparatus | |
| CN107616783A (en) | A kind of cardiac monitoring platform | |
| Deotale et al. | Identification of arrhythmia using ECG signal patterns |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20160629 |
|
| RJ01 | Rejection of invention patent application after publication |