CN106037720B - A medical application system of mixed continuous information analysis technology - Google Patents
A medical application system of mixed continuous information analysis technology Download PDFInfo
- Publication number
- CN106037720B CN106037720B CN201510880861.8A CN201510880861A CN106037720B CN 106037720 B CN106037720 B CN 106037720B CN 201510880861 A CN201510880861 A CN 201510880861A CN 106037720 B CN106037720 B CN 106037720B
- Authority
- CN
- China
- Prior art keywords
- heart rate
- event
- module
- heartbeat
- patient
- 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.)
- Active
Links
- 238000004458 analytical method Methods 0.000 title claims abstract description 89
- 238000005516 engineering process Methods 0.000 title claims abstract description 32
- 230000002159 abnormal effect Effects 0.000 claims abstract description 104
- 238000001514 detection method Methods 0.000 claims abstract description 63
- 238000012545 processing Methods 0.000 claims abstract description 56
- 230000000747 cardiac effect Effects 0.000 claims abstract description 41
- 238000003745 diagnosis Methods 0.000 claims abstract description 18
- 239000000284 extract Substances 0.000 claims abstract description 9
- 230000009429 distress Effects 0.000 claims abstract description 4
- 230000033764 rhythmic process Effects 0.000 claims description 60
- 238000000718 qrs complex Methods 0.000 claims description 37
- 238000000034 method Methods 0.000 claims description 20
- 238000012544 monitoring process Methods 0.000 claims description 14
- 230000007774 longterm Effects 0.000 claims description 10
- 208000001871 Tachycardia Diseases 0.000 claims description 7
- 238000007781 pre-processing Methods 0.000 claims description 7
- 208000024891 symptom Diseases 0.000 claims description 7
- 230000006794 tachycardia Effects 0.000 claims description 7
- 239000003814 drug Substances 0.000 claims description 5
- 230000008569 process Effects 0.000 claims description 5
- 108010076504 Protein Sorting Signals Proteins 0.000 claims description 4
- 101000712600 Homo sapiens Thyroid hormone receptor beta Proteins 0.000 claims description 3
- 102100033451 Thyroid hormone receptor beta Human genes 0.000 claims description 3
- 206010044565 Tremor Diseases 0.000 claims description 3
- 229940079593 drug Drugs 0.000 claims description 3
- 210000002837 heart atrium Anatomy 0.000 claims description 3
- 208000006218 bradycardia Diseases 0.000 claims description 2
- 230000036471 bradycardia Effects 0.000 claims description 2
- 208000019622 heart disease Diseases 0.000 claims description 2
- 239000002131 composite material Substances 0.000 claims 7
- 238000004364 calculation method Methods 0.000 claims 1
- 230000001684 chronic effect Effects 0.000 claims 1
- 239000003550 marker Substances 0.000 claims 1
- 230000001575 pathological effect Effects 0.000 claims 1
- 230000005540 biological transmission Effects 0.000 abstract description 10
- 230000005856 abnormality Effects 0.000 abstract description 6
- 238000013461 design Methods 0.000 description 13
- 238000010586 diagram Methods 0.000 description 12
- 206010047302 ventricular tachycardia Diseases 0.000 description 11
- 208000000418 Premature Cardiac Complexes Diseases 0.000 description 9
- 208000009729 Ventricular Premature Complexes Diseases 0.000 description 8
- 206010015856 Extrasystoles Diseases 0.000 description 7
- 230000006870 function Effects 0.000 description 7
- 239000000203 mixture Substances 0.000 description 7
- 230000008859 change Effects 0.000 description 6
- 230000002861 ventricular Effects 0.000 description 6
- 206010047289 Ventricular extrasystoles Diseases 0.000 description 5
- 230000000694 effects Effects 0.000 description 4
- 238000011160 research Methods 0.000 description 4
- 230000002547 anomalous effect Effects 0.000 description 3
- 230000006399 behavior Effects 0.000 description 3
- 150000001875 compounds Chemical class 0.000 description 3
- 201000010099 disease Diseases 0.000 description 3
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 3
- 238000007689 inspection Methods 0.000 description 3
- 230000007170 pathology Effects 0.000 description 3
- 230000001568 sexual effect Effects 0.000 description 3
- 230000011664 signaling Effects 0.000 description 3
- 238000012360 testing method Methods 0.000 description 3
- 208000024172 Cardiovascular disease Diseases 0.000 description 2
- 206010003119 arrhythmia Diseases 0.000 description 2
- 230000006793 arrhythmia Effects 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 208000029078 coronary artery disease Diseases 0.000 description 2
- 230000009191 jumping Effects 0.000 description 2
- 230000000737 periodic effect Effects 0.000 description 2
- 238000004321 preservation Methods 0.000 description 2
- 239000004575 stone Substances 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 206010003658 Atrial Fibrillation Diseases 0.000 description 1
- 206010007558 Cardiac failure chronic Diseases 0.000 description 1
- 208000002330 Congenital Heart Defects Diseases 0.000 description 1
- 206010020772 Hypertension Diseases 0.000 description 1
- 206010049418 Sudden Cardiac Death Diseases 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 description 1
- 230000001746 atrial effect Effects 0.000 description 1
- 210000000467 autonomic pathway Anatomy 0.000 description 1
- 230000003796 beauty Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000008280 blood Substances 0.000 description 1
- 210000004369 blood Anatomy 0.000 description 1
- 230000036772 blood pressure Effects 0.000 description 1
- 238000007675 cardiac surgery Methods 0.000 description 1
- 230000000052 comparative effect Effects 0.000 description 1
- 230000001447 compensatory effect Effects 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000013144 data compression Methods 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 238000002565 electrocardiography Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 230000005831 heart abnormality Effects 0.000 description 1
- 230000000302 ischemic effect Effects 0.000 description 1
- 230000003902 lesion Effects 0.000 description 1
- 210000005036 nerve Anatomy 0.000 description 1
- 229910052760 oxygen Inorganic materials 0.000 description 1
- 239000001301 oxygen Substances 0.000 description 1
- 238000001615 p wave Methods 0.000 description 1
- 238000011056 performance test Methods 0.000 description 1
- 238000005498 polishing Methods 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 238000004393 prognosis Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 230000009329 sexual behaviour Effects 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
- 230000002269 spontaneous effect Effects 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 230000002889 sympathetic effect Effects 0.000 description 1
- 238000010998 test method Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 230000001515 vagal effect Effects 0.000 description 1
Landscapes
- Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
Abstract
The invention discloses a kind of medical application systems for mixing continuous information analytical technology, module, which is obtained, by the ECG signal in electrocardio mixing continuous analysis system obtains ECG signal from electrocardio equipment, then ECG signal is pre-processed by electrocardiogram detecting module, treated, and signal extracts signal characteristic by electrocardiogram detection method and basic heart rate event one by one is made so that complex event processing heartbeat identification module is analyzed, complex event processing heartbeat identification module is compared basic heart rate event and the information in patient's history's database module by the continuous analysis module of mixing therein, identify abnormal cardiac rate event, when there is serious abnormal cardiac rate event to pre- diagnosis output module transmission abnormality heartbeat alarm and pre- diagnostic message, pre- diagnosis output module is sent to the doctor of patient and/or family members Alarm and distress signals.
Description
Technical field
The present invention relates to a kind of application method for mixing continuous information analytical technology, especially a kind of mixing continuous informations point
The medical application system of analysis technology.
Background technique
Electrocardiogram (electrocardiograms, ECG) is heart companion when each cardiac cycle cardiac is beated
With biological Electrical change, by electrocardiograph from body surface draw diversified forms potential change figure, it gives each
The function detail of human heart simultaneously can help to analyze the abnormal heartbeats in ECG signal.Periodic P can be observed in electrocardiogram,
QRS and T wave train, QRS wave group has peak swing in this sequence, helps to calculate its week to its detection
The P enclosed, T wave and other features of heartbeat.
Cardiovascular disease is always one of to endanger human health, cause the main reason for human death, the identification point of ECG
Clinically tool has very important significance for analysis.The ECG clinical application in China is broadly divided into three kinds of forms at present: one is patients to connect
Usually the scanning of the heartbeat of patient is no more than one minute when such as physical examination and generic cardiac detection by the short time test of electrocardiogram,
And the ECG signal of patient can be directly printed upon on paper, be analyzed it by doctor, then inform that patient's electrocardiogram has He Yi
Often.Be for second be in peril of one's life for inpatient or heart in terms of disease when, ECG monitor can be provided to detect patient
Heartbeat blood pressure, the data such as blood oxygen amount.The purpose of this mode is intended to detect patient when there is the case where threat to life, electrocardio prison
Shield instrument will do it alarm, if patient trembles, situations such as stop jumping, and most patient ECG information and without recording.This
Sample just to be unable to get good grasp into the ECG situation doctor of a period of time for patient.For in intensive care unit
(ICU) except patient.The third is that after cardiac surgery patient or serious cardiac monitor within electrocardio 24 hours,
It is commonly called as " knapsack ".This is to connect a portable electrocardiogram recording instrument with the lead for connecing in human body, makes patient voluntarily movable, with note
Record the electrocardiogram (ECG) data that patient goes over 24 hours.These data are imported in computer from taking-up in recorder then and are analyzed.
For three of the above form, jointly will in face of the problem of there are three: firstly, the medium-term and long-term electrocardiogram (ECG) data of patient
It does not all achieve, that is to say, that can not be by analyzing the rehabilitation situation of patient or sb.'s illness took a turn for the worse to the analysis of history ECG
Degree.Secondly, the analysis for electrocardiogram, patient, which takes, to be divided in addition to looking for the doctor of profession after the electrocardiogram being recorded on paper
Analysis, it is helpless for the redemption of sudden heart disease without real-time.Third, only urgent patient can obtain medicine side
The profession in face helps, for generally suffering from hypertension, a large amount of the elderlys of coronary heart disease they not have approach to understand oneself own daily
Physical condition.And due to the difference of human body individual, everyone normal heartbeat is also not quite similar.We pass through Internal retrieval
It can be found that the normal heartbeat of human body is 60-100 beats/min, then this value just has 1 problem, such as 1 normal heart of people
Jumping is 60 beats/min, if his beats are 100 beats/min suddenly, although the numerical value is still within the range of normal heartbeat,
But tachycardia is in for himself.There are also be exactly somebody itself normal heartbeat itself it is higher or
Person is relatively low, and 110 beats/min of somebody's normal heartbeat or more, somebody's normal heartbeat is at 60 beats/min hereinafter, if with a masses
Numerical value it is judged if, inevitably there is mistake.
Therefore, the heartbeat situation for designing a system combination patient itself, in due course detects the heartbeat of patient, works as disease
When abnormal heartbeats occurs in people, abnormal heartbeats can be shown, and by the case history and patient of the performance of abnormal heartbeats and patient
Past heartbeat historical data is compared, and alarms serious abnormal heartbeats, and the auxiliary provided for each patient is pre-
The case where diagnosis, help doctor preferably grasps patient is a urgent problem to be solved.
Summary of the invention
The object of the present invention is to provide a kind of application method of mixing Continuous Analytical Technique medically.This method can
To combine the heartbeat situation of patient itself, the heartbeat of patient is detected in due course, it, can be right when abnormal heartbeats occurs in patient
Abnormal heartbeats are shown, and the case history of the performance of abnormal heartbeats and patient and the past heartbeat historical data of patient are compared
It is right, it alarms serious abnormal heartbeats, the auxiliary provided for each patient diagnoses in advance, and doctor is helped preferably to grasp disease
The case where people.
Technical solution of the present invention: a kind of medical application system mixing continuous information analytical technology is mixed by electrocardio
ECG signal in continuous analysis system obtains module and obtains ECG signal from electrocardio equipment, is then visited by electrocardiogram
It surveys module to pre-process ECG signal, treated, and signal extracts signal characteristic by electrocardiogram detection method and makes
At basic heart rate event one by one so that complex event processing heartbeat identification module is analyzed, complex event processing heartbeat is known
Other module by the continuous analysis module of mixing therein by the information in basic heart rate event and patient's history's database module into
Row comparison, identifies abnormal cardiac rate event, when there is serious abnormal cardiac rate event to the pre- diagnosis output module transmission abnormality heart
Alarm and pre- diagnostic message are jumped, the pre- output module that diagnoses is to the doctor of patient and/or family members' transmission alarm and distress signals.
In the medical application system of mixing continuous information analytical technology above-mentioned, the ECG signal obtains module by the heart
Electric patient monitor or portable electrocardiograph obtain signal, and signal is by Bluetooth transmission to mobile phone or PC.
In the medical application system of mixing continuous information analytical technology above-mentioned, the electrocardiogram detecting module is from electrocardiogram
Each heartbeat is identified in the voltage signal sequence that signal acquisition module transmission comes, and is extracted between the heart rate of heartbeat, RR wave
Every, P wave, QRS complex and T wave property parameter, and these characterisitic parameters are fabricated to basic heart rate event and are transferred to compound
Event handling heartbeat identification module is used for the matching of abnormal cardiac rate event；Electrocardiogram (ECG) data is stored in history database module simultaneously
In.
In the medical application system of mixing continuous information analytical technology above-mentioned, the electrocardiogram detecting module includes letter
Number preprocessing module and electrocardio detecting module；Electrocardio detecting module includes QRS complex detection module and P, T wave detection module；Its
The voltage signal that ECG signal obtains module transfer is removed work using wavelet transformation first by middle signal pre-processing module
Frequency interferes and is directed to baseline drift problem, and adjusts baseline with fitting of a polynomial；QRS complex detection module uses dynamic thresholding
Detection method carries out point-by-point variance, point-by-point square and the point-by-point square operation of signal amplitude to the data that real-time Transmission comes,
There is the integral that moving window is done at the peak R；P, T wave detection module is for analyzing from QRS the window phase of one complete heartbeat
The corresponding P of heartbeat, the position of T wave and information are found in signal before and after wave group.
In the medical application system of mixing continuous information analytical technology above-mentioned, the dynamic thresholding detection method includes
Following steps:
A, the window that 1 size is 1000 sample points is established, as new signal inputs, old sample point removes window
Mouthful；
B, pretreated signal is subjected to sliding average processing, the sliding window of 11 sample points is averaged；
C, it is equal to the average value of sample point and the sum of sample average in sliding window using dynamic threshold THR1 to filter out
Lower than the signal section of threshold value；
D, using dynamic threshold THR2 be equal to the average value of sample point and sample average in sliding window difference filter
Fall to be higher than threshold signal part；
E, the interval in the non-zero section that comparison step C and step D is generated, when interval less than 50, non-zero section sample point
When, merge by adjacent latter one non-zero section zero setting, and by step C with the result of step D；
F, the part after step E processing is all to find point of the most value point as the peak R in wave crest in QRS complex；
G, it after step D processing, if non-zero section is not fallen after rise at sliding window edge, is regarded as QRS wave group and does not reach
To highest point, the newest R peak value that both step F had been found is there are error, the newest R peak value that step G is analyzed
Label；
If H, treated that non-zero section has been fallen after rise in window edge by step D, then executes a step F for new knot
The peak position R found in fruit comparison sliding window so that it is determined that a newest peak R position；
J, two sizes for calculating the section edge being not zero after step E processing are to find out second dervative in 20 sample intervals
Section of the sample point as R wave that changes of sign；
K, by former and later two RR wave sections, calculate the section of intermediate complete heartbeat, and to the window before and after QRS complex into
Row analysis, finds out the section for meeting subthreshold THR3 as P, the section of T wave is analyzed；
L, by the interval time of QRS complex, P wave, T wave in each heartbeat section, peak value, the sample point of state pause judgments
It sets information preservation and issues complex event processing heartbeat identification module.
In the medical application system of mixing continuous information analytical technology above-mentioned, the complex event processing heartbeat identifies mould
Block monitors each basic heart rate event that electrocardiogram detecting module exports by the continuous analysis module of mixing therein,
Identify whether each basic heart rate event meets normal heartbeat by the EPL sentence in complex event processing heartbeat identification module
Condition, for abnormal cardiac rate event by in patient's history's database module abnormal cardiac rate table carry out matching judgment its be
Any abnormal cardiac rate event, and the position in event tree or occurrence diagram is in judge this abnormal heart according to abnormal cardiac rate event
Can rate event form other increasingly complex abnormal cardiac rate events, and judge whether to remove corresponding pattern matching statement,
Finally complex event processing heartbeat identification module is sent by the new complicated event of gained；And by all basic of each heartbeat
Heart rate event is as in historical data deposit history database module.
In the medical application system of mixing continuous information analytical technology above-mentioned, the history database module includes ill
Basic Information Table, storage basal heart rate table, abnormal cardiac rate summary table, the abnormal cardiac rate of people divides table and heart rate variability analysis numerical value
Table；Each table is divided into two storage sections of day and night according to time difference.
It is described to mix continuous analysis module for patient in the medical application system of mixing continuous information analytical technology above-mentioned
The medium-term and long-term index of heart rate is recorded, and heart pathology event caused by due to individual difference is adjusted using these indexs
It screens, and being applied in complicated heart rate event handling when making complicated heart rate event handling used match pattern can be with
Dynamically it is changed and adjusts.
In the medical application system of mixing continuous information analytical technology above-mentioned, the realization of the continuous analysis module of mixing
Including there are three parts: mixing elementary event, mixing complicated event and comprehensive pre- diagnosis；Wherein mixing elementary event by pair
The analysis and statistics of the normal heartbeat of history obtain the distribution relative to QRS complex of P wave and T wave in everyone normal cardiac rate
Position, and past all single hearts rate are excavated or analyzed to obtain the history feature value of the rhythm of the heart, the two uses mixing
Continuous analysis mode is compared, to analyze the category attribute of basic heart rate；Mixing complicated event passes through in compound thing
Part handles heartbeat identification module in the snoop procedure of abnormal cardiac rate event, extracts and occurred in patient's history's database module
Different abnormal cardiac rate events characteristic value, and these characteristic values are used for the real-time abnormal cardiac rate event detection of patient；When real-time
When detecting abnormal heart rhythm and being converted into the complicated event of abnormal heart rhythm, rapidly judged whether there is by comprehensive pre- diagnosis
Life danger, and information relevant to symptom is sent to medical staff.
Beneficial effects of the present invention: compared with prior art, the heartbeat feelings of application method combination patient itself of the invention
Condition in due course detects the heartbeat of patient, when abnormal heartbeats occurs in patient, can show to abnormal heartbeats, and will
The performance of abnormal heartbeats is compared with the case history of patient with the past heartbeat historical data of patient, to serious abnormal heartbeats into
The case where row alarm, the auxiliary provided for each patient diagnoses in advance, doctor is helped preferably to grasp patient.Present invention incorporates
The actual demand of people devises electrocardio mixing continuous analysis system and relevant processing method, proposes electrocardio detection method
And Complex event processing module.It is proposed that a kind of electrocardiogram detection method detects the basal heart rate of electrocardiogram, and by its
Conversion is basic flow of event, flows Event processing engine using Esper to construct continuous analysis system, produces to electrocardiogram detection method
Raw basic flow of event carries out pattern match.Heart real time data and abnormal cardiac rate data that system generates are deposited into history number
According in library.The EPL sentence for the Common Abnormity rhythm of the heart is devised simultaneously, it will a large amount of complicated thing in the form of complicated event tree
Relationship between part, which is together in series, carries out Dynamic Pattern Matching, reduces the pressure of system.Finally, realizing electrocardio history number
According to and feature taken out from database and be supplied to flow of event analysis engine and carry out more accurate and personalized pattern match.?
It is fitted on timely early warning after preset abnormal heart rhythm event, and relevant historical record, the medicining condition etc. of patient are quickly provided
Relevent information, to improve the treatment rate of medical staff.The advantages of summarizing mainly including the following aspects:
1, a kind of real-time electrocardio detection method is proposed for the special requirement of ecg analysis, this method can be preferable
It is that cardioelectric monitor analysis provides real-time event support.
2, the electrocardio mixing continuous analysis system based on Complex event processing is devised, by Complex event processing analytical technology
It is introduced into electrocardio monitoring and analysis.
3, a kind of continuous analysis framework of mixing is devised, traditional Continuous Analytical Technique can only set threshold value in advance, if
It then alarms more than threshold value；Everyone constitution and heartbeat feature are had nothing in common with each other relatively, and mixing Continuous Analytical Technique can be certain
This problem is preferably solved in degree.
4, electrocardio mixing continuous analysis system uses complex event processing techniques, and data-driven is changed to event-driven mould
Formula, saving calculating power also save a large amount of initial data memory spaces and only store relatively small event data.
Detailed description of the invention
Attached drawing 1 is the structural schematic diagram of electrocardio mixing continuous analysis system of the invention；
Attached drawing 2 is the level schematic diagram of complicated event；
Attached drawing 3 is the flow chart of complex event processing heartbeat identification module；
Attached drawing 4 is HR Heart Rate hybrid monitoring flow chart；
Attached drawing 5 is the event tree schematic diagram of ventricular tachycardia event；
Attached drawing 6 is ventricular bigeminy schematic diagram；
Attached drawing 7 is room property trigeminy schematic diagram；
Attached drawing 8 is the error rate schematic diagram of MIT ecg database；
Attached drawing 9 is identification schematic diagram of the complicated event technology to abnormal heart rhythm；
Attached drawing 10 is mixing Continuous Analytical Technique abnormal cardiac rate discrimination schematic diagram；
Specific embodiment
The present invention is further illustrated with reference to the accompanying drawings and examples, but be not intended as to the present invention limit according to
According to.
The embodiment of the present invention: present invention research is exactly based on CEP (again with the electrocardio mixing continuous analysis system of design
Close event handling) technology analyzes and saves to a large amount of biology character data that patient's each moment generates.Electrocardio mixing
The research and design of continuous analysis system can solve the problem of a large amount of biology character data are directly dropped, and these are counted
According to analysis for save life and improve physical condition.
Electrocardiogram (electrocardiograms, ECG) is heart companion when each cardiac cycle cardiac is beated
With biological Electrical change, by electrocardiograph from body surface draw diversified forms potential change figure, it gives each
The function detail of human heart simultaneously can help to analyze the abnormal heartbeats in ECG signal.Periodic P can be observed in electrocardiogram,
QRS and T wave train, QRS wave group has peak swing in this sequence, helps to calculate its week to its detection
The P enclosed, T wave and other features of heartbeat.Research to the combination for analyzing and mixing Continuous Analytical Technique of Electrocardiogram
It is the main object of the present invention with trial.It is explained below and is set for the medicine ECG frame for being mixed the system continuously analyzed
Meter and module design.
Electrocardio mixing continuous analysis system Frame Design
Electrocardio mixing continuous analysis system needs obtain ECG signal from various electrocardio equipment, for these signals CEP
(complex event processing) technology can not directly carry out processing and obtain information, and there are also the influences such as noise to sentence for possible signal source
The factor of disconnected quality, so the first step is exactly to need to carry out ECG signal a series of pretreatment so that the quality of signal meets
The requirement of detection method.Next cardiac electrical signal characteristic can be extracted by ECG detection method for CEP system point
Analysis.CEP system alarms to serious abnormal heartbeats to abnormal heartbeats are found out after the electrocardiogram (ECG) data analysis sent.And it will be different
The performance of normal heartbeat is compared with the case history of patient with the past heartbeat historical data of patient, provides for the auxiliary of each patient
The case where helping pre- diagnosis, doctor helped preferably to grasp patient.
According to above analysis, the ECG real-time monitoring analyzing system based on CEP can effectively to ECG signal at
Reason, identifies the event of improper heartbeat, heartbeat event is compared with historical informations such as the case histories of patient and provides pre- examine
Disconnected information.The patient common for help understands the physical condition of oneself and makes the decision offer support of medical treatment in time.Electrocardio
It is as shown in Fig. 1 to mix continuous analysis system frame.The system is that ECG signal obtains mould respectively by being divided into five modules
Block, electrocardiogram detecting module, CEP heartbeat identification module, history database module diagnose output module in advance.It below will be right respectively
The main functional modules and the relevant technologies of system are introduced.
ECG signal obtains module design
ECG signal is obtained to be completed by ECG monitor, portable electrocardiograph etc., these signals pass through Bluetooth transmission to mobile phone
Or pc, the detection of ECG is carried out in mobile phone terminal or the end pc, it can't be by the electrocardiosignal real-time Transmission of 360HZ in order to save bandwidth
To server end, but gone out by the program instrumentation of mobile phone terminal or the end pc after the characteristic information of each heartbeat and after compressing, with every point
The ECG characteristic signal that clock or longer a cycle upload this primary period is handled to server end；If in terminal
The ECG compressed data for finding that improper heartbeat was then transmitted at once in this existing period in detection is divided to server end
Analysis.The information received by serial ports is used as to the data source of next step Signal Pretreatment by the oracle listener at the end PC.Monitor journey
Sequence can recorde the frequency of signal source and the strong and weak value of each signal for preprocessing module use.
What is generally stored in ecg database is the electrocardiogram (ECG) data of one or more lead, the signal source that includes need according to
Different databases are decoded, and general ecg database can include header file and diagnostic file, and header file generally provides electrocardio
Relevant information, such as the signal frequency of electrocardio use, signal base position, unit, the quantity of lead and position and patient
Some relevant informations: age, gender, medicining condition etc..And diagnostic file will indicate the type and time of origin of abnormal heartbeats, it will
The signaling point that specific abnormal heartbeats occur is marked with correlation.Reading markup information greatly will help developer to verify
The cardioelectric monitor algorithm or program of oneself.
The design of electrocardiogram detecting module
The function of electrocardiogram detecting module is the heartbeat identified from the voltage signal sequence of heart one by one, and is extracted
Some of which characterisitic parameter such as heart rate out, RR wave spacing, P wave, QRS complex, T wave.These characterisitic parameters are constituted into the basic heart
Rule event is simultaneously transferred to complex event processing heartbeat identification module for the matching use of abnormal heart rhythm event.Electrocardiogram detecting module
Mainly consist of two parts: signal pre-processing module and electrocardio detecting module.Electrocardio detecting module includes QRS complex detection identification
And P, T wave detection identification two parts.The function of this three parts will be illustrated with design respectively below.
Signal pre-processing module design: for the discrimination of ECG signal have an important influence be ECG record quality,
The electrocardiosignal usually actually obtained has various interference and drift, so carrying out pretreatment for signal is very
It is necessary to behavior.ECG signal there are two important interference source, one be 50Hz/60Hz and its harmonic wave Hz noise；It is another
A a little baseline drifts for being less than 1Hz.Hz noise is removed using wavelet transformation first and for baseline drift problem, selection
Fitting of a polynomial (polynomial fitting) adjusts baseline.
The design of electrocardio detecting module: it includes the detection of signal QRS complex, can not be to complete because analyzing real time data
The feature of whole heart rate record extracts, therefore uses a kind of dynamic threshold detection method, the data to come to real-time Transmission into
The point-by-point variance of row, point-by-point square, point-by-point square of signal amplitude is waited operation, so that the data after output are positive, and non-linear
The signal for being exaggerated differential output, the high frequency section of prominent signal more highlights the peak R, and it is positive to reduce vacation caused by T wave
Property；There is the integral for doing moving window at the peak R, extracting the other information of R wave, such as slope, width and raising QRS are comprehensive
The accuracy rate of multiplex detection.It further include having the detection of P, T wave, it, can by the calculating of detection and RR wave spacing to QRS complex
Easily substantially to calculate the window phase of a complete heartbeat.This window phase is analyzed before and after QRS complex
The possible position of corresponding P, T wave of this heartbeat and other information are found in signal；But and not all P, T signal are ok
Identified, such as tachycardia, trembled in such a way, and others P, T wave it is Chong Die with other waves the case where just
It is difficult to be identified, but usual P, T wave, which can not identify, can also become a kind of mark, can also be used to do abnormal electrocardiogram
Judgement.
It needs for the essential information details of heartbeat to be compressed into from the voltage value of several hundred a sample points to each for this system
The master data information of heartbeat, dynamic threshold detection method are mainly described as follows.
Step1. it establishes in the window that a size is 1000 sample points, as new signal inputs, old sample
Point grand window.
Step2. pretreated signal is subjected to sliding average processing, the sliding window of 11 sample points is averaged.
The acquirement of sliding average can effectively reduce dynamic threshold mistake and filter out signal section lower than its mean value.
N=1,2 ... .1000
Step3. dynamic threshold THR1 is equal to the average value of sample point and the sum of sample average in sliding window and filters out
Lower than the signal section of threshold value.
N=1,2 ... .1000
Step4. dynamic threshold THR2 be equal to the average value of sample point and sample average in sliding window difference filter
Fall to be higher than threshold signal part.The position of R wave can be positioned.
N=1,2 ... .1000
The interval in the non-zero section Step5. generated in comparison Step3 and Step4, when interval less than 50, non-zero section sample
At this, merge by adjacent latter one non-zero section zero setting, and by Step3 with the result in Step4.The master done so
Syllabus is to abstract practical QRS complex, can quickly navigate to the position of QRS complex.For next step to the position of P wave into
Row estimation is prepared.
Step6. the part after Step5 processing is all to find in wave crest most value point as the peak R in QRS complex
Point.
Step7. after Step4 processing, if non-zero section is not fallen after rise at sliding window edge, it is regarded as QRS complex
It may not peak, there may be errors for the newest R peak value that both Step6 has been found, and the 7th step is analyzed
A newest R peak markers.
If Step8. treated that non-zero section has been fallen after rise in window edge by Step4, then executing a Step6 will
The peak position R found in new Comparative result sliding window so that it is determined that a newest peak R position.
Step9. two sizes for calculating the section edge being not zero after Step5 processing are to find out second order in 20 sample intervals
Section of the sample point that the sign of derivative changes as R wave.
Step10. by the section of this complete heartbeat among the calculating in former and later two sections RR, and to QRS complex before
Window afterwards is analyzed, and finds out the section for meeting subthreshold THR3 as P, the section of T wave is analyzed.
Step11. by each heartbeat section QRS complex, P wave, T wave (if P wave, T wave are detectable) detailed number
According to such as interval time, peak value, the sample point position information preservation of state pause judgments simultaneously issues identification module.
Electrocardiography module is foundation stone of the system to heartbeat type analysis accuracy, is played a crucial role.
The design of complex event processing heartbeat identification module
Complex event processing (CEP) heartbeat identification module is that the core of electrocardio mixing continuous analysis system is also hybrid analysis
Basis.CEP technology is utilized to handle and analyze the anomalous event in real-time heart rate, single heart rate anomalous event in this module
It can cause the complicated event of a series of abnormal heart rhythm.By identifying obtained elementary event to the monitoring of electrocardiogram detecting module
It monitors, various abnormal heart rhythm events can be listened to, and the generation for the abnormal heart rhythm event that multiple specific rules occur can be recognized
To be the abnormal heart rhythm complicated event for meeting expression pattern in EPL sentence.System for the monitoring of these complicated events can and
When notify medical staff and provide patient symptom data at the first time to help to save the life of patient to a certain extent.It is attached
Fig. 2 show the level schematic diagram of the complicated event arrived used in system.First by the electrocardiogram detecting module in system Lai
Generate basic heartbeat event is indicated with box；All ellipses indicate complicated event, pass through EPL sentence in figure and carry out mode
Match, the generation of certain specific elementary events will trigger EPL sentence and generate complicated event, the complicated event generated by elementary event
We position level-one complicated event, that is, in hierarchy chart complicated event first layer.The complicated event of generation can be also flowed into
In system, meets the complicated event of mode when certain EPL sentence listens to and have occurred, then generate the complicated event of more high-level.
The level of complicated event is higher, shows that heart abnormality symptom is more clear and the generation of this event is for sufferer
It is more of vital importance.Such as a common patients with coronary heart disease, it is i.e. popular that thousands of Premature Ventricular Beats perhaps can occur daily
The premature beat of title, such situation are too frequently not enough to judge whether patient serious symptom occurs.If but the hair of proiosystole
Raw frequency and mode induce a secondary event such as VT tachycardia event, and it is unpredictable clinically to mean that patient has
Potential fatal implementations may occur；If the time of VT event duration reaches a certain level the complicated thing that the third level will occur
Part, such as cardiac ischemic event, that just means that patient needs to carry out first aid at once, if taking no action to that life will be jeopardized.
The process of complex event processing heartbeat identification module is as shown in Fig. 3, and CEP is dividing in real time to continuous signal source
Analysis, so there is no the terminals that method marks identification process for flow chart.After any cardiac monitoring data source access system, electrocardio
Figure detecting module will carry out pretreatment and waveforms detection to the signal of sampling.The heartbeat message that then each is detected seals
Underlying rhythm event one by one is dressed up, and is deposited into basic heartbeat.When underlying rhythm event generates, system is responsible for identifying base
The EPL sentence of this rhythm of the heart event will identify whether the condition for meeting normal heartbeat, monitor next time if not abnormal
Heartbeat event.It, can be according to existing some matchings if CEP monitors the data of this rhythm of the heart and do not meet the definition of normal cardiac rhythm
Rule is stored into abnormal heart rhythm table to match this heartbeat be any abnormal heart rhythm.Complicated event system can be according to different
Normal heart rate event is in the position in event tree or occurrence diagram to judge that it is other increasingly complex that can this abnormal heart rhythm form
Abnormal heart rhythm event.If upper one layer different there is no the top layer of arrival event tree, is added in this event in the mode that CEP is monitored
Normal cardiac rhythm pattern.And it is top if having arrived at event tree, illustrate that event has been monitored, then needs to remove corresponding
Pattern matching statement.Finally complex event processing heartbeat identification module is sent by the new complicated event of gained.Compound thing later
Part processing heartbeat identification module will ceaselessly recycle the step in execution flow chart.
The essential information of heartbeat is made basic heart rate event and is sent to complex event processing heartbeat knowledge by electrocardiogram detecting module
Other module, and by ECG data compression deposit history database module in case search later；Complex event processing heartbeat identifies mould
Block will carry out pattern match after receiving the heartbeat event transmitted, using all underlying rhythm events of each heartbeat as
Historical data is stored in database, the feature of abnormal heartbeats will be stored in off-note data if meeting the feature of abnormal heartbeats
Library is in case diagnosis output module is called in advance later.
Heartbeat event is constituted: the data read out from ECG monitor etc. detect after being pre-processed through electrocardiogram
Method can detect the detailed features value an of heartbeat, as QRS complex, P wave, T wave details as underlying rhythm thing
Part, underlying rhythm event is only to be able to reflect some parameters of the current heartbeat real-time detected, although it can also be provided perhaps
Mostly relevant detailed information, but the target of these information and heartbeat exception to be automatically detected is there are also at a distance from very big, it can not
Express a complete meaning.Therefore it needs underlying rhythm event aggregation into the complicated thing for having practical medical meaning one by one
Part.Here is the basic description to event.
Underlying rhythm event: each underlying rhythm event has an only available name, here by patient Uid+
BeatPart+BeatID composition.Source list shows ECG signal source from mobile device or ECG monitor, Yi Jishe
Standby number；There are also the positions that database of case history name shows medical records storage.Attribute list includes: ecg_id is which lead be shown to be
Information；Beat_ID be used for mark be which heartbeat attribute event；Strat_time and end_time indicates event
At the beginning of and End Event, if event do not have duration end_time can for sky.
Complicated event: the more subevent subevent and constraint list constraint conditions compared with elementary event, it is multiple
Miscellaneous affair part is made of multiple subevents and its constraint relationship.Complicated event is mainly used to indicate some distinctive improper hearts
Rate event, such as VT Ventricular Tachycardia, ventricular bigeminy, room property trigeminy.
The function of the continuous analysis module of mixing in complex event processing heartbeat identification module is by the historical data of patient
Information extract the continuous analysis for real time data.Mix medium-term and long-term digit synbol of the continuous analysis module by patient heart rate
Record is got off, and the examination of heart pathology event caused by due to individual difference is adjusted using these indexs, and applied to
Used match pattern when Complex event processing is allowed dynamically to be changed and adjust in Complex event processing.
History database module design
Analysis system needs to be added electrocardio history database module and goes through to store to compare with history electrocardiogram (ECG) data
History data.History database module is broadly divided into five class tables of data.
The first kind is the basic breath table of patient, contains the patient number of patient, name, year in patient's Basic Information Table
The basic documents such as age, gender and medication history.Daily behavior information note, such as whether there is the behavior of big physical sport or labour.
What is stored in basic heart rate table is by being sent to each of complex event processing heartbeat identification module after the identification of electrocardiogram detection method
The essential information parameter of secondary heartbeat.
Second class is storage basal heart rate table, both the related data of heartbeat each time, such as date, the interval heart rate ID, RR, P
Wave, QRS complex, detail parameters of T wave etc.；And this kind of basal heart rate storage table has that can be divided into two class one kind be that the same day is of that month
Table data store, another kind of is the table data store of history electrocardiogram (ECG) data.Storing in storage basal heart rate table is every month
Basic heart rate, literary name segment structure indicate consistent with basic heart rate, and monthly system will copy to the basic heart rate table of this month and deposit
It stores up in basal heart rate table, and basic heart rate table is emptied to store all heart rate informations of this month.
Third class is abnormal cardiac rate summary table, stores all abnormal hearts rate, the neither ID of normal heartbeat, abnormal cardiac rate class
Type, duration and abnormal cardiac rate ID etc.；Two classes, the of that month abnormal heart rate of one kind storage, another kind of storage can also be equally divided into
The abnormal heart rate of history.
4th class is that abnormal cardiac rate divides table, has both been divided and has been stored by ventricle, atrium, the region of junctional area three and is different different
Normal heart rate, such as Ventricular Tachycardia, atrial premature beats, the rich abnormal heartbeats type of junctional area ease.Table database is divided to be divided into three, point
It Wei not room sexual abnormality rhythm of the heart database, room sexual abnormality rhythm of the heart data and junctional area heart rate exception database.
5th class is that HRV analyzes numerical tabular, is a kind of heart rate analysis numerical value using electrocardiogram as data source, these pass through meter
The numerical value obtained can be very good to represent the dirty some features of institute's thought-read, these features can be used to analysis measured and suffer from
There are a possibility that chronic heart failure exhausts and degree.HRV (heart rate variability) reaction is autonomic nerves system activity and qualitative assessment
Cardiac sympathetic nerve and vagal tone and balance, to judge its state of an illness and prognosis to cardiovascular disease.It is prediction
One extremely valuable index of sudden cardiac death and arrhythmia cordis sexual behavior part.
Pre- diagnosis output module design
After complex event processing heartbeat identification module identifies abnormal heart rhythm, mixing inquiry will start, while to going through
History ECG data, abnormal cardiac rate pattern base, case history are inquired and are compared with real-time ECG situation.When comparison find patient case history in
Relevant part, and to pre- diagnosis output module transmission abnormality heartbeat alarm and pre- diagnostic message when having a high-risk grade, in advance
Alarm or distress signals can be sent to the doctor and families of patients that patient history is set by diagnosing output module, receive information to other people
It is given treatment to.
To understand the heart rate situation of a people, medium-term and long-term electrocardiogram (ECG) data will bring greatly help and benefit.One
The situation of change of personal medium-term and long-term heart rate can directly embody the changing rule and lesion situation of heart of patient.And it can
To apply in the heart rate monitoring of individual as the heart rate reference index of this person.
It as shown in Fig. 4, is HR Heart Rate hybrid monitoring flow chart, when underlying rhythm event detects mould by electrocardiogram in figure
Block generate after can be monitored by complex event processing heartbeat identification module, the heart rate of day and night be slightly different thus through
Corresponding medium-term and long-term history average can be taken out after judgement from the historical data on night or daytime respectively.Real-time underlying rhythm thing
Judged after part and historical data base data mixing, is considered as bradycardia when being less than averaged historical heart rate 65%；Take care
Rate is considered to be tachycardia when being the 200~250% of history heart rate；And it is then considered as when being greater than the 250% of history heart rate
Heart is trembled.The heart rate range of the normal person introduced in today that technical level continues to develop, in Traditional Textbooks with
And occur tachycardia, tremble and the confining spectrum of heart escape beat be no longer desirable for individual monitoring demand.For
The grasp of medium-term and long-term heart rate index will can help to carry out personalized heart rate monitoring.Historical data can be directed to everyone heart rate
Benchmark come carry out individuation threshold value setting and event analysis.
Complicated event --- CEP heartbeat identification
The electric signal of the data source of complex event processing heartbeat identification module, each lead of ECG monitor is passing through ECGD
Underlying rhythm event can be packaged into after detecting module.Complicated event is by one or more underlying rhythm events or complicated event
It constitutes.The generation of complicated event is the mode in CEP system analysis EPL sentence and is matched in elementary event stream corresponding
Mode and generate, complicated event generate after equally can be also sent to by the form of a flow of event in CEP system, and may
By other mode detections and matching；
Attached drawing 5 is the event tree schematic diagram of ventricular tachycardia event, and diagrammatically shown is the room complicated event VT property mistake aroused in interest
The case where event of fast basic composition situation and next stage.Ventricular Tachycardia (ventricular tachycardia,
VT): abbreviation ventricular tachycardia, refer to originating from ventricle, it is spontaneous, continuous 3 or 3 or more, frequency be greater than 100 beats/min of proiosystole
The rhythm of the heart of composition.
Complex event processing heartbeat identification module detects the various features of underlying rhythm event to be matched to room property first
Proiosystole, and secondary abnormal heart rhythm event is added in flow of event.The abnormal cardiac rate event of Premature Ventricular Beats is added into
Associated EPL sentence will be dynamically activated while flow of event, that is, increase subsequent matching mould relevant to Premature Ventricular Beats
Formula is monitored.VT Ventricular Tachycardia as shown in Fig. 5, attached ventricular bigeminy shown in fig. 6, attached room property three shown in Fig. 7
The mode of rule is monitored.
Hereafter mode contained by all EPL sentences on the listening-in line that complex event processing heartbeat identification module will continue to.
If hereafter continue to have continuous two elementary events to be identified as Premature Ventricular Beats, and this continuous three heartbeat event
The room the VT property heart when average value that the average frequency of RR wave spacing all exceeds 100 i.e. RR wave spacings per minute is just matched less than 216
The dynamic feature overrun.This thing system can generate VT complicated event and be added in current time stream.If have it is more complicated based on
The mode of VT Ventricular Tachycardia, CEP will activate the monitoring of associative mode.
Due to the cardiac characteristic of each patient be it is different, usually all rhythm of the heart events will not all be monitored, only
It can be monitored by the event that single underlying rhythm event monitoring is got in event tree.When specific heart rate anomalous event
After being activated, it will be activated in the father node EPL sentence of each event tree containing this abnormal heart rhythm event, and in CEP
In monitored.It can effectively economize on resources in this way and mitigate system burden.
The complicated event of VT only needs following information, and patient ID, patient's heartbeat number, date and abnormal heart rhythm event are opened
The time of beginning and the time of end.It can be used as the time that VT event starts, then when heart rate event no longer matches this complicated thing
This logout end time is given when part.This abnormal heart rhythm complicated event is stored in abnormal heart rhythm correspondence database later.
Premature Ventricular Beats are characterized in: QRS complex occurs ahead of time, paramophia, and the time limit is most > and 0.12 second, T wave
Contrary with the main wave of QRS wave, ST is shifted with T wave, preceding without P wave.The ventricular premature beat of bundle branch proximal end, QRS complex occurs
It can not be broadening.There is complete compensatory pause after Premature Ventricular Beats mostly.When underlying rhythm is slower, Premature Ventricular Beats are inserted into
Twice between sinus property heartbeat, insert type Premature Ventricular Beats are formed.The accidental inverse retrograde P wave for reaching atrium, often comes across ST sections
On.
Usually go deep into research, the number of plies of event tree is to gradually increase event tree in other words can grow up, same to current events
The number of part tree also will increase.It can make to continue to support more complicated things while not being modified system in this way
Part, this is also the flexible place of CEP technology.
The continuous analysis of mixing
The upper section of complicated event analysis in to(for) electrocardio monitoring, which is realized, has been made description.This for universal heart rate not
Neat patient is very helpful, but everyone heart rate reference index and same heart rate are abnormal in practical application
Electrocardiogram pattern caused by phenomenon is different, this virtually brings challenge to accurate match of the system to individual.Mixing
The original intention for the application continuously analyzed is precisely in order to enable to this set that can have based on the ecg analysis system of Complex event processing
There are stronger individual compatibility and accuracy.And can be realized the physical condition for different patients oneself carry out analysis and
Prediction.
Based on complicated event, three parts are classified into the realization for mixing continuous analysis module: mixing elementary event,
Mix complicated event and comprehensive pre- diagnosis.
The detection of basic heart rate is completed by electrocardiogram detection method, and this method is to have for the detection of QRS complex
Higher accuracy；But be so far still a difficult medical problem for the accurate detection of P wave and T wave, because of most of feelings
Under condition, the difference of the heart pathology feature of patient leads to the P wave of different patients and the relative position relative to QRS complex of T wave
There is very big difference, is to be difficult to be accurately positioned and be supplied to CEP to be analyzed with fixed test method, so mixing is continuous
Analysis can solve the problems, such as this to a certain extent.
The electrocardiogram of the normal complete heartbeat of everyone single is easiest to orient P wave and T wave, to the normal of history
The distribution position relative to QRS complex of specific P wave and T wave in the analysis of heartbeat and everyone available normal cardiac rate of statistics
It sets.Using this more everyone exclusive accurate distributing position section, to largely solve determining for P wave and T wave
Position problem.
Further, the electrocardiogram of everyone single complete heartbeat can extract, the characteristic value of corresponding heartbeat type.
The history feature value of the available particular kind of rhythm of the heart, the two if the single heart rate all to the past is excavated or analyzed
Being compared can use the continuous analysis mode of mixing to analyze the attributes such as the type of basic heart rate.
There is independent feature for everyone basic heart rate situation and the electrocardiogram pattern of presentation, for patient
The extraction of each complete heartbeat characteristic value is that public relatively-stationary abnormal heart rhythm can will be directed in medical domain
Standard more individuation.
The abnormal heart of the difference occurred in patient's history's record is extracted under the basis of original Complex event processing system
The characteristic value of rule, and these characteristic values are used for the real-time abnormal heart rhythm of patient and are detected.The historical record of patient is in the database
Daily saved.Daytime, the HR Heart Rate of people can be higher than night, more excited.So historical record was needed daytime and night
Evening has separated analysis characteristic value.It is medically 9AM to 8PM and 9 points of that morning to 8 points at night by the section definition on daytime；
The time interval at night is defined as 9PM to next day 8AM., be to going through so as it can be seen that although data are daily stored
When history data are analyzed, the section on date and daytime and night is but referred to.So needing to repair the EPL sentence of CEP system
Change and the access modules to historical data base are added in systems.
To handling it is required that EPL sentence can support the acquisition to historical data for mixing complicated event；In Nesper
Support introduces custom function in EPL sentence, is by ODBC Open Database Connection or JDBC using custom function
Java database connects to link corresponding database, takes out the mean eigenvalue of the corresponding abnormal heart rhythm of type in database simultaneously
Pattern match is carried out used as the threshold value in EPL and play.
Pre- diagnosis being completed for task of output module in fact aiming at real-time detect abnormal heart rhythm be converted into it is different
When the complicated event of Chang Xinlv, quickly judge whether to be in peril of one's life, and information relevant to symptom is sent to medical care people
Member.If in patient's essential information, installing pacemaker additional, and historical data before then loses reference value and patient one
Directly in the safety problem taken which drug medical worker can be helped quickly to judge diagnosis and treatment method and prescription.
System effect and performance test
Database is chosen: the standard cardioelectric database of mainstream mainly has 3 in the world: European AT-T ecg database, beauty
The MIT-BIH database that the AHA database and the Massachusetts Institute of Technology that heart association, state provides provide, wherein the BIH number of MIT
According to having in library including multiple database such as arrhythmia cordis, ST sections of changes, atrial fibrillations.
The assessment of electrocardio detection method
Table 1MIT database QRS complex testing result
Electrocardio detection method is as shown in table 1 to the testing result in MIT-BIH heart rate data library, and wherein error rate is by following public affairs
Formula obtains.It is not difficult to find out that can be detected for most Heart Rate from result, however for heart rate polishing
It is more serious and with p wave, t wave disappear heart rate for error rate then will appear a degree of promotion.Simultaneously for room property
The electrocardio detection accuracy of the more frequent patient of proiosystole needs further promoted.
It is attached Fig. 8 shows be choose MIT-BIH ecg database in ten sections of electrocardiographic recordings detection error statistics situation.
Left figure is that missing inspection number and erroneous detection number count, and is respectively the missing inspection number and erroneous detection number of every record in figure, missing inspection number is to lose heartbeat
The quantity not identified, erroneous detection number are the quantity that the signaling point except positive and negative 5 signaling points of R wave crest is identified as to R wave crest；Right figure
It is total error rate statistic, error rate has the sum of rate of failing to report and rate of false alarm to get divided by heartbeat sum.
It can be seen that, detection method is that there is relatively more in the identification to record 200 and record 223 from attached drawing 8
Big error, respectively reach 4% or more.So particular for analyze record 200 and record 223 the characteristics of, record
There is the ventricular bigeminy symptom up to 12 minutes in 200 electrocardiographic recording and along with serious myoelectricity interference.For 30
The electrocardiographic recording of minute length, 12 minutes bigeminies have already taken up a big chunk time.It is up in record 223
1 point of 50 seconds VT tachycardia, the Detection accuracy of heart rate during which will receive bigger influence.It is recorded from this two
From the point of view of analysis, electrocardio detection method also deposits the identification of more severe signal environment and the faster continuous abnormal rhythm of the heart at present
In some problems.The success rate and accuracy of the recognition methods of heart rate are the foundation stones of heart rate analysis software, also determine system
Accuracy and availability.The performance and effect of this method can achieve the standard for meeting system design.
The assessment of electrocardio complicated event analysis system
CEP complicated event detection result: in order to verify effect of the CEP technology in heartbeat inspecting analysis, for MIT-BIH
Heartbeat record tested after obtain following experimental result, 223 exception has been counted respectively to record 200, record 208, record
The frequency of room property heartbeat, as shown in table 5.7.As can be seen from the table, there are ventricular premature beat in No. 200 heart rate records
826 times, system identification is 719 times, wherein number 71 times of bigeminy generation, practical identification 63 times, Ventricular Tachycardia 7 times,
Practical identification 6 times.
2 Room sexual abnormality heart rate of table identification statistics
As shown in Fig. 9, left figure is the sum to the ventricular premature beat in three heartbeat records and CEP system identification respectively
Number.What right figure indicated is recorded in 200 and record 223 to the identification mistake of ventricular premature beat, ventricular bigeminy, Ventricular Tachycardia
Accidentally rate.The identification error rate of ventricular premature beat in record 200 and record 223 is respectively reached from can be seen that in attached drawing 9
13% and 7.6%, and be 4% and 4.18% in the error rate of electrocardio detection method.This illustrate the accuracy in detection of CEP system by
To the influence of electrocardio detection method, for the identification of more severe signal environment and the faster continuous abnormal rhythm of the heart, there is also one
A little problems.
All it is respectively higher than 85% for the average recognition rate of three groups of records of MIT abnormal cardiac rate, experimental result is said well
The applicability of CEP technology is illustrated.
The continuous analysis outcome evaluation of mixing
The statistics to historical data, the available exception based on historical data are added in electrocardio mixing continuous analysis system
The fundamental characteristics statistical value of the rhythm of the heart can modify the EPL language for being responsible for that pattern match is carried out to abnormal heart rhythm by these statistical values
Match attribute value in sentence, to improve the accuracy of identification.
As shown in Fig. 10, left figure is the sum to the ventricular premature beat in three heartbeat records and the continuous analysis of mixing respectively
The identification number of system.What right figure indicated is record 200, record 208 and the CEP for recording 223 and the continuous analysis of mixing to room property morning
The error rate comparison of heart rate of fighting identification.
It can be seen from figure 10 that compared with for CEP system mix Continuous Analytical Technique for abnormal cardiac rate identification just
True rate is higher, is reduced to 6.77% from 12.95% to the identification error rate of record 200；Record 208 mistakes corresponding with 223
Rate is also reduced to 2.72% and 5.07% from 3.12% and 6.71% respectively.The reduction of error rate is it will be evident that main cause
It is as, there is a large amount of bigeminy phenomenon, the alternating generation of bigeminy phenomenon will interfere CEP to find premature beat in record 200
The phenomenon that rhythm of the heart event.Suitable attributes match model can be found to the statistics of patient's history's heart rate by mixing connecting technique
It encloses, so that the identification to the rhythm of the heart is more accurate.
It is demonstrated experimentally that being helpful by discriminance analysis of the statistical analysis to historical data for real-time heart rate, no
But it can be identified for personal physical fitness situation, and the precision of identification can be improved, be connected to demonstrate mixing
Continue the advisability of analytical technology.
Claims (8)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510880861.8A CN106037720B (en) | 2015-12-04 | 2015-12-04 | A medical application system of mixed continuous information analysis technology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510880861.8A CN106037720B (en) | 2015-12-04 | 2015-12-04 | A medical application system of mixed continuous information analysis technology |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106037720A CN106037720A (en) | 2016-10-26 |
| CN106037720B true CN106037720B (en) | 2019-04-19 |
Family
ID=57179374
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510880861.8A Active CN106037720B (en) | 2015-12-04 | 2015-12-04 | A medical application system of mixed continuous information analysis technology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106037720B (en) |
Families Citing this family (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107137077A (en) * | 2017-06-14 | 2017-09-08 | 西南大学 | Autonomic nerve recovers detection method, device and electronic equipment |
| CN107569227A (en) * | 2017-10-26 | 2018-01-12 | 太仓市同维电子有限公司 | The processing method and monitoring device of heart rate under a kind of motion state |
| CN108158578A (en) * | 2017-12-14 | 2018-06-15 | 北京顺源开华科技有限公司 | Noise segments recognition methods, ECG signal processing method and processing device |
| CN108836305B (en) * | 2018-05-08 | 2019-04-12 | 北京理工大学 | A kind of ECG feature extracting method of fusion Butterworth filtering and wavelet transformation |
| CN109199375B (en) * | 2018-11-30 | 2021-11-02 | 东南大学 | A non-invasive fetal ECG detection device and ECG signal data processing method |
| CN111317463B (en) * | 2018-12-13 | 2022-12-06 | 深圳市理邦精密仪器股份有限公司 | Heart beat insertion method, heart beat automatic insertion device and storage medium |
| CN113905667A (en) * | 2019-06-28 | 2022-01-07 | 深圳迈瑞生物医疗电子股份有限公司 | Monitoring processing method and device and storage medium |
| CN112006679B (en) * | 2020-08-28 | 2021-08-03 | 山东省人工智能研究院 | Wearable electrocardiosignal R wave detection method based on window variance transformation |
| CN115844420B (en) * | 2022-07-21 | 2024-12-24 | 成都怡康科技有限公司 | Group electrocardiogram arrhythmia diagnosis method |
| CN115879179B (en) * | 2023-02-24 | 2023-05-09 | 忻州师范学院 | Abnormal medical record detection device |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101911083A (en) * | 2008-01-07 | 2010-12-08 | 通用电气公司 | System, method and device for predicting sudden cardiac death risk |
| CN103584872A (en) * | 2013-10-29 | 2014-02-19 | 燕山大学 | Psychological stress assessment method based on multi-physiological-parameter integration |
| CN104116501A (en) * | 2014-03-27 | 2014-10-29 | 中国东方航空股份有限公司北京分公司 | Long-distance double-direction health monitoring integration system |
| CN104382582A (en) * | 2014-11-10 | 2015-03-04 | 哈尔滨医科大学 | Device for classifying dynamic electrocardio data |
| CN104523265A (en) * | 2014-12-31 | 2015-04-22 | 广州视源电子科技股份有限公司 | Abnormal electrocardiogram data judgment method and device |
| CN104720790A (en) * | 2015-03-30 | 2015-06-24 | 中国计量科学研究院 | Multichannel synchronous electrocardiograph calibrating apparatus |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP3435262A1 (en) * | 2010-03-15 | 2019-01-30 | Singapore Health Services Pte. Ltd. | A system for the detection of impending acute cardiopulmonary medical events |
-
2015
- 2015-12-04 CN CN201510880861.8A patent/CN106037720B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101911083A (en) * | 2008-01-07 | 2010-12-08 | 通用电气公司 | System, method and device for predicting sudden cardiac death risk |
| CN103584872A (en) * | 2013-10-29 | 2014-02-19 | 燕山大学 | Psychological stress assessment method based on multi-physiological-parameter integration |
| CN104116501A (en) * | 2014-03-27 | 2014-10-29 | 中国东方航空股份有限公司北京分公司 | Long-distance double-direction health monitoring integration system |
| CN104382582A (en) * | 2014-11-10 | 2015-03-04 | 哈尔滨医科大学 | Device for classifying dynamic electrocardio data |
| CN104523265A (en) * | 2014-12-31 | 2015-04-22 | 广州视源电子科技股份有限公司 | Abnormal electrocardiogram data judgment method and device |
| CN104720790A (en) * | 2015-03-30 | 2015-06-24 | 中国计量科学研究院 | Multichannel synchronous electrocardiograph calibrating apparatus |
Non-Patent Citations (1)
| Title |
|---|
| 基于CEP的ECG实时监测分析系统的研究与设计;戴震宇等;《计算机工程与设计》;20140216;第35卷(第2期);第731-735页 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106037720A (en) | 2016-10-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106037720B (en) | A medical application system of mixed continuous information analysis technology | |
| Pathinarupothi et al. | Instantaneous heart rate as a robust feature for sleep apnea severity detection using deep learning | |
| JP6199330B2 (en) | Identification of Chain Stokes breathing patterns using oximetry signals | |
| US10163174B2 (en) | Methods, systems, and computer program products for evaluating a patient in a pediatric intensive care unit | |
| CN112365978A (en) | Method and device for establishing early risk assessment model of tachycardia event | |
| CN113613559A (en) | Electrocardiogram processing system for delineation and classification | |
| Zhang et al. | Sleep stage classification using bidirectional lstm in wearable multi-sensor systems | |
| US20200205709A1 (en) | Mental state indicator | |
| CA2927807C (en) | System and method for providing multi-organ variability decision support for extubation management | |
| CN107252313A (en) | The monitoring method and system of a kind of safe driving, automobile, readable storage medium storing program for executing | |
| CN109620209B (en) | Dynamic electrocardiograph, respiration and motion monitoring system and method | |
| Choudhary et al. | Analyzing seismocardiographic approach for heart rate variability measurement | |
| US20210345896A1 (en) | Heart Rate Variability Monitoring and Analysis | |
| WO2014036651A1 (en) | Method for multi-scale quality assessment for variability analysis | |
| EP4216232A1 (en) | Methods and system for cardiac arrhythmia prediction using transformer-based neural networks | |
| Mohammadpoor Faskhodi et al. | Arousal detection by using ultra-short-term heart rate variability (HRV) analysis | |
| TW202143915A (en) | Electrocardiogram analysis | |
| Cuadros et al. | ECG multilead QT interval estimation using support vector machines | |
| Du-Yan et al. | Convolutional neural network is a good technique for sleep staging based on HRV: a comparative analysis | |
| Yin et al. | Intelligent monitoring of noxious stimulation during anaesthesia based on heart rate variability analysis | |
| Maurer et al. | Explainable Artificial Intelligence on Biosignals for Clinical Decision Support | |
| Konstantin et al. | Noise-resilient Automatic Interpretation of Holter ECG Recordings | |
| Roederer et al. | Towards non-invasive monitoring of hypovolemia in intensive care patients | |
| Ghista et al. | ECG waveform and heart rate variability signal analysis to detect cardiac arrhythmias | |
| Nazarchuk | Heart Rate Variability Analysis within a Data-Enabled Design research |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |