CN204520670U - A kind of electrocardiogram monitor system based on Internet of Things - Google Patents

Abstract

This utility model relates to a kind of electrocardiogram monitor system based on Internet of Things, comprise electrocardiogram acquisition module, main intelligent mobile terminal, cardioelectric monitor module, electrocardiogram acquisition module is connected with main intelligent mobile terminal, main intelligent mobile terminal and the two-way communication of cardioelectric monitor module, electrocardiogram acquisition module comprises the signals collecting end connected successively, data processing module, signal emission module, data processing module is sent to main intelligent mobile terminal by after the ECG's data compression collected through signal emission module, electrocardiosignal is sent to cardioelectric monitor module by main intelligent mobile terminal, cardioelectric monitor module comprises remote communication module, diagnostic analysis module, ecg signal data storehouse, remote communication module is used for communicating with main intelligent mobile terminal, and the electrocardiosignal received is sent to ecg database storage, the signal that diagnostic analysis module extracts ecg database reception carries out diagnostic analysis, after diagnostic analysis, result is sent to main intelligent mobile terminal by remote communication module.

Description

A kind of electrocardiogram monitor system based on Internet of Things

Technical field

This utility model relates to human biology detection technique field, particularly a kind of electrocardiogram monitor system based on Internet of Things.

Background technology

Heart disease has PD slow and hidden, and once the feature that sb.'s sickness becomes critical of falling ill, therefore, the life of the mankind in serious threat.Generally, the main cause of heart disease patient sudden death is malignant arrhythmia and heart failure etc.; Special heart attack has no time rule, and this kind of patient can show symptom in working at ordinary times or living, but when doing electrocardiographic examination to hospital, symptom may disappear, therefore, cannot abnormal electrocardiographic pattern be detected, cause doctor can not make the state of an illness and judge accurately, delay best occasion for the treatment.If mortality arrhythmia can be identified in time clinically, take effective Prevention and Curation measure, the success rate saving patient vitals can be improved to a great extent.

For many years, due to the extensive use of electrocardiosignal automatic analysis technology, huge facility is brought to numerous cardiac, but, existing electrocardiogram monitor system electrocardiogram acquisition processing unit and discharger is blended in a special mobile phone, by hand-held, make mobile phone adopting bottom electrode contact thoracic wall form loop to gather electrocardiosignal, then by mobile communication network, ecg information is sent to data processing centre.Solve ecg signal acquiring and transmission problem under mobile status, greatly reduce the restriction of time, region.But, there is following shortcoming in this electrocardiogram monitor system: 1, does not possess intellectual analysis warning function, automatically cannot identify and send abnormal electrocardiogram signal, patient must manually gather and send, when happen suddenly serious heart disease and patient have a rest as sleep, often cannot perform manual operation and lose best opportunity for the treatment of.2, cannot realize electrocardio in real time, continuously, multi-lead monitoring, in the monitoring of clinical practice especially ischemic heart desease, complicated arrhythmia, meaning is very limited.

Utility model content

The technical problems to be solved in the utility model is a kind of ECG detecting and early warning system automatically, can electrocardiogram acquisition and transmission automatically, and realizes the cardioelectric monitor system of real-time, continuous, the multi-lead detection of electrocardiosignal.

For solving the problems of the technologies described above, technical solution adopted in the utility model is:

A kind of electrocardiogram monitor system based on Internet of Things, comprise electrocardiogram acquisition module, main intelligent mobile terminal, cardioelectric monitor module, described electrocardiogram acquisition module is connected with described main intelligent mobile terminal, described main intelligent mobile terminal and the two-way communication of described cardioelectric monitor module, described electrocardiogram acquisition module comprises the signals collecting end connected successively, data processing module, signal transmission module, described main intelligent mobile terminal is sent to through described signal transmission module after the ECG's data compression that described signals collecting end collects by described data processing module, the electrocardiosignal of acquisition is sent to described cardioelectric monitor module by described main intelligent mobile terminal, described cardioelectric monitor module comprises remote communication module, diagnostic analysis module, ecg signal data storehouse, described remote communication module is used for communicating with main intelligent mobile terminal, and the electrocardiosignal received is sent to the storage of described ecg database, the signal that described diagnostic analysis module extracts the reception of described ecg database carries out diagnostic analysis, after diagnostic analysis, result is sent to main intelligent mobile terminal by described remote communication module.

Preferably, in technique scheme, described signals collecting end comprises core signal sensor, pulse transducer, oximetry sensor, temperature sensor, is respectively used to detect electrocardiosignal, pulse frequency, blood oxygen saturation and body temperature.

Preferably, in technique scheme, described data processing module comprises the front end amplifying unit, signal filtering unit, rear end amplifying unit, wave trap, the A/D converter that connect in turn, and described front end amplifying unit adopts precision amplifier AD620 circuit; Described signal filtering unit adopts Butterworth filter; Described rear end amplifying unit adopts a scaling circuit to connect an add circuit by a voltage follower; Described wave trap adopts 50Hz wave trap chip LMF100, and uses chip LMF100 wave trap mode of operation.

Preferably, in technique scheme, described cardioelectric monitor module also comprises case search module and system help module, and described case search module helps doctor to analyze for inquiring about patient history's case, and described system help module is used for User Defined and arranges.

Preferably, in technique scheme, described signal emission module is Bluetooth communication modules or Zigbee transmission module.

Preferably, in technique scheme, described electrocardiogram monitor system also comprises secondary intelligent mobile terminal, and described secondary intelligent mobile terminal is by described remote communication module and described cardioelectric monitor module communication.

Preferably, in technique scheme, described remote communication module is GPRS wireless transport module or 3G wireless transport module or 4G wireless transport module or Socket data transmission module.

Preferably, in technique scheme, described main intelligent mobile terminal and described secondary intelligent mobile terminal are smart mobile phone or panel computer.

Compared with prior art, the beneficial effects of the utility model are:

This utility model electrocardiogram acquisition module and main intelligent mobile terminal adopt separate type, and this structure avoids the passivity needing user manually to gather electrocardiosignal, achieve real-time, continuous, that multi-lead detects electrocardiosignal function; Electrocardiogram acquisition module acquires to electrocardiosignal pass the signal to cardioelectric monitor module by main intelligent mobile terminal, in signal detection module, diagnostic analysis is carried out in the ecg signal data storehouse storing electrocardiosignal by diagnostic analysis module, and diagnostic result is sent to detection and warning function that main intelligent mobile terminal achieves electrocardiosignal.Electrocardiosignal diagnostic analysis module can make diagnosis fast, not only shortens the waiting time of patient, also mitigates the misery of patient, and all right fast treating, this seems more crucial concerning acute cardiac patient.

In preferred version of the present utility model, signals collecting end can not only gather electrocardiosignal, the parameters such as pulse frequency, blood oxygen saturation, body temperature can also be gathered, the type of exercise that user does can be judged by these parameters, and these parameters are conducive to the state of an illness, health that user and doctor analyze user, also more fully understand health.

In preferred version of the present utility model, signal emission module preferentially selects wireless transmission method, as Bluetooth transmission or Zigbee transmission, select the mode of wireless transmission to make use more convenient, structure is simpler, use assembling also more convenient, provide larger facility to user.

All select best operation element in preferred version of the present utility model in data processing module, make the electrocardiosignal that detects more accurate.

In preferred version of the present utility model, secondary intelligent mobile terminal can be the family members of patient, also secondary intelligent mobile terminal can be issued while the result of cardioelectric monitor module analysis diagnosis issues main intelligent mobile terminal, family numbers of patients is allowed to know the situation of patient, and the relief made promptly and accurately determines, improve cure rate.

In sum, the movable information of patient when this utility model can record anomalous ecg exactly, and realize the classification of motion and intensity judges, relevant data can pass through mobile phone real time inspection.When there is anomalous ecg kinestate, alarming short message can be sent out from trend family members and hospital, can patient, set up real-time contact between family and doctor, realize electrocardiosignal stablize, accurately, real-time transmission, the range of activity of patient will no longer limit by hospital, the doctor of hospital recognizes the cardiac function situation of monitored patient outside institute by server, be conducive to doctor understand in time and find the state of an illness, the efficiency of diagnosis will be improved, the loss because delay treatment causes patient opportunity can be reduced, and make corresponding Diagnosis and Treat.The shared resources of medical resource can be increased substantially simultaneously, reduce the cost of public's medical treatment, alleviate doctor-patient relationship, for medical treatment brings new service and facility, and the service network system of round-the-clock cardiac monitoring outside institute can be provided to lay a good foundation for building one in the whole nation.

A kind of electrocardiogram monitor system based on Internet of Things that this utility model provides has following meaning: one is have positive role to the daily monitoring of the heart of sub-health population and health care, is that routine work is busy, stress is comparatively large, the ideal tools of the self-cardiac monitoring of people from all walks of life that are that lack exercise; Two is outside institute, observe curative effect of medication to patients with arrhythmia and state of illness monitoring has clinical meaning; Three be contribute to cardiovascular patient leave hospital after long-term state of illness monitoring; Four is one of prevention and the important means reducing the generation of some heart disease malignant event.

Accompanying drawing explanation

Below in conjunction with accompanying drawing, the utility model is described in further detail:

Fig. 1 is this utility model the general frame;

Fig. 2 is amplifying unit circuit theory diagrams in front end in this utility model;

Fig. 3 is signal filtering element circuit schematic diagram in this utility model;

Fig. 4 is amplifying unit circuit theory diagrams in rear end in this utility model;

Fig. 5 is the circuit theory diagrams of wave trap in this utility model;

Fig. 6 is the Wavelet-denoising Method schematic diagram related in this utility model;

Fig. 7 is the Wavelet-denoising Method flow chart related in this utility model.

Detailed description of the invention

As shown in Figure 1, a kind of electrocardiogram monitor system based on Internet of Things that this utility model provides, comprise electrocardiogram acquisition module, main intelligent mobile terminal, cardioelectric monitor module, electrocardiogram acquisition module is connected with main intelligent mobile terminal, main intelligent mobile terminal and the two-way communication of cardioelectric monitor module, electrocardiogram acquisition module comprises the signals collecting end connected successively, data processing module, signal emission module, data processing module comprises the front end amplifying unit connected in turn, signal filtering unit, rear end amplifying unit, wave trap, A/D converter, signal transmission module is Bluetooth communication modules or Zigbee transmission module, the preferred Bluetooth communication modules of the present embodiment, main intelligent mobile terminal is sent to through signal transmission module after the ECG's data compression that signals collecting end collects by data processing module, the electrocardiosignal of acquisition is sent to cardioelectric monitor module by main intelligent mobile terminal, cardioelectric monitor module comprises remote communication module, diagnostic analysis module, ecg signal data storehouse, remote communication module is used for communicating with main intelligent mobile terminal, and the electrocardiosignal received is sent to ecg database storage, the signal that diagnostic analysis module extracts ecg database reception carries out diagnostic analysis, after diagnostic analysis, result is sent to main intelligent mobile terminal by remote communication module, electrocardiogram monitor system described in the present embodiment also comprises secondary intelligent mobile terminal, cardioelectric monitor module is communicated with secondary intelligent mobile terminal by remote communication module.

The software analysis module of a kind of electrocardiogram monitor system based on Internet of Things that this utility model provides is development platform with VS2012, uses OO C language and complete in conjunction with SQL Server data base.

The transmission means of the present embodiment medium-long range communication module can be GPRS wireless transmission, or 3G wireless transmission, or 4G wireless transmission, or Socket data transmission module, the preferred Socket data transmission module of the present embodiment.

In the present embodiment, signals collecting end comprises core signal sensor, pulse transducer, oximetry sensor, temperature sensor, is respectively used to detect electrocardiosignal, pulse frequency, blood oxygen saturation and body temperature.

In the present embodiment, cardioelectric monitor module also comprises case search module and system help module, and case search module helps doctor to analyze for inquiring about patient history's case, and system help module is used for User Defined and arranges, and asks for help Xiang doctor.

In the present embodiment, main intelligent mobile terminal and secondary intelligent mobile terminal are smart mobile phone, or panel computer.

This utility model can be analyzed according to the various signals detected, and exactly record anomalous ecg time patient movable information, and realize classification and the intensity judgement of motion, relevant data patient can pass through mobile phone real time inspection, when there is anomalous ecg kinestate, cardioelectric monitor module to be transmitted messages alarming information from trend family members and hospital, can patient, set up real-time contact between family and doctor, realize electrocardiosignal stablize, accurately, real-time transmission, the range of activity of patient will no longer limit by hospital.

In the present embodiment, the specific design of data processing module is as follows:

(1) design of front end amplifying unit:

Electrocardiosignal derives from the potential difference of body surface, and when heart working, in each cardiac cycle, pacemaker and atrium, ventricle alternating excitation, can cause bioelectric change, be out electrocardiosignal by this bioelectric change detection.The design of front end amplifying unit in the present embodiment, selects high precision amplifier AD620.The features such as the bandwidth that this amplifier has that common mode rejection ratio is high, precision is high, amplify, low in energy consumption, the little and noise coefficient of drifting about is low.Because AD620 input resistance is very large, therefore use it for the collection terminal of small-signal, the gain G of AD620 and external control gain resistor R _grelation meet formula (1):

$G=\frac{49.4\mathrm{k\Ω}}{R_G}+1---\left(1\right)$

Due in the signal that collects containing part interference, first order amplification should not arrange too high, arranges too high meeting and is unfavorable for the process of subsequent conditioning circuit to noise, therefore, set and signal is amplified 5 times, be i.e. G=5, as calculated external control gain resistor R _g=12.3k Ω.

The interference had the greatest impact in ECG detecting process is 50Hz Hz noise, in order to reduce the impact of this interference on ECG detecting, needs to add driven-right-leg circuit to reduce common mode disturbances in ECG detecting circuit.In the present embodiment, front end amplifying unit circuit theory as shown in Figure 2.

(2) design of signal filtering unit

The electrocardiosignal that signals collecting end gathers is amplified after also preliminary removal interference through front end amplifying unit, and still have DC component and interference, therefore, signal filtering unit selects filter circuit, reaches the object of filtering DC component and elimination baseline drift.Wherein, interference mainly comprises High-frequency Interference and 50Hz interference, because rear end amplifying unit will introduce 50Hz interference, therefore wouldn't process 50Hz interference in this Unit Design.

Owing to containing multiple wave bands such as P ripple, QRS ripple, T ripple in electrocardiosignal, residing frequency domain and amplitude different, need more each waveform duration and amplitude when observing waveform, so require level and smooth as much as possible in the passband of wave filter.Preferentially Butterworth filter is selected in the present embodiment.

First allowing the signal of front end amplifying unit output by butterworth high pass filter, filtering DC component and elimination baseline drift, is 0.7Hz according to electrocardiosignal feature and designing requirement setting cut-off frequency.Butterworth high pass filter is by two resistance R ₁, R ₂, two electric capacity C ₁, C ₂form with an operational amplifier.The transfer function of this wave filter is:

$H_1\left(S\right)=\frac{S^2}{S^2+S\left(\frac{C_1+C_2}{R_1{C}_1{C}_2}\right)+\frac{1}{R_1{R}_2{C}_1{C}_2}}+1---\left(2\right)$

Choose the coefficient a of second order Butterworth network function ₁₁=1.414, cut-off angular frequency ω _c=0.7 × 2 π, and set C ₁=C ₂=0.1 μ F, calculates R according to parameter and formula (2) ₁and R ₂:

$R_1=\frac{C_1+C_2}{C_1{C}_2{\mathrm{\ω}}_C{a}_{11}}=1.61\mathrm{M\Ω}---\left(3\right)$

$R_2=\frac{C_1+C_2}{C_1{C}_2{\mathrm{\ω}}_C{R}_1}=3.21\mathrm{M\Ω}---\left(4\right)$

By butterworth high pass filter filtering DC component with after eliminating baseline drift, adopting Butterworth LPF filter away high frequency noise, is 100Hz according to electrocardiosignal feature and designing requirement setting cut-off frequency.Butterworth LPF circuit is by two electric capacity C ₃, C ₄, two resistance R ₃, R ₄and an operational amplifier is formed.The transfer function of this wave filter is:

$H_1\left(S\right)=\frac{\frac{1}{R_3{R}_4{C}_3{C}_4}}{S^2+S\left(\frac{R_3+R_4}{R_3{R}_4{C}_4}\right)+\frac{1}{R_3{R}_4{C}_3{C}_4}}---\left(5\right)$

Choose the coefficient a of second order Butterworth network function ₁₁=1.414, cut-off angular frequency ω _c=100 × 2 π, and set C ₃=C ₄=0.1 μ F.R is calculated according to parameter and formula (5) ₃and R ₄:

$R_3=\frac{a_{11}}{{\mathrm{\ω}}_C{C}_3}=22.5\mathrm{k\Ω}---\left(6\right)$

$R_4=\frac{1}{R_3{C}_3{C}_4{{\mathrm{\ω}}_C}^2}=11.3\mathrm{k\Ω}---\left(7\right)$

The design of this unit is the drift of adjustment signal base line and filtering interfering, and the feature such as operational amplifier demand fulfillment bipolarity, low noise, low input offset voltage therefore selected in two wave filter, in the present embodiment, two operational amplifiers preferentially select OP07.OP07 is the bipolar operational amplifier of a low noise, non-chopper-zero-stabilized, meets the condition of this Unit Design.In the present embodiment, the circuit theory of signal filtering unit as shown in Figure 3.

(3) design of rear end amplifying unit

After signal filtering unit filtering, eliminate DC component and High-frequency Interference, need to carry out scale amplifying to this signal, the design of this unit is that signal is amplified 100 times, reaches the degree can observing waveform.Simultaneously, because the signal finally exported will be changed through A/D, and the integrated A/D function of the single-chip microcomputer selected is unipolar, so signal will be added DC component, make it reach within the scope of 0 ~ 5v, this Unit Design mainly comprises two parts circuit: scaling circuit and add circuit.

The operational amplifier that the amplifying circuit of this unit in the present embodiment is selected requires the OP07 operational amplifier of higher open-loop gain, low noise and low input offset voltage; Add circuit process be amplify after signal, be not very high for precision, noise requirements, the present embodiment preferentially selects double operational chip NE5532.

In scaling circuit, the amplification of circuit is by RF ₁with RF ₂ratio determine, selected RF ₁=1k Ω, RF ₂=100k Ω.In order to avoid scaling circuit backend load affects amplification effect, between scaling circuit and add circuit, add a voltage follower.Add circuit part needs signal base line to promote 2.5V, namely realizes v+2.5, and input value is signal v and+5V power supply.Signal v and+5V are respectively through RF ₃, RF ₄to be connected and through RF with holding with operational amplifier ₅ground connection.For realizing v+2.5, then RF ₃/ RF ₄=0.5, if RF ₃=RF ₅=10k Ω, RF ₄=20k Ω, then:

$\frac{{\mathrm{RF}}_7}{{\mathrm{RF}}_6}=\frac{{\mathrm{RF}}_3}{{\mathrm{RF}}_4}+\frac{{\mathrm{RF}}_4}{{\mathrm{RF}}_5}=\frac{3}{2}---\left(8\right)$

RF is set by formula (8) ₆=10k Ω, then RF ₇=15k Ω.According to the circuit theory of the rear end amplifying unit of parameter designing as shown in Figure 4.

(4) design of wave trap

In ecg signal data processing procedure, the interference had the greatest impact is exactly 50Hz Hz noise, therefore needs to carry out design circuit for filtering 50Hz Hz noise.For the interference of single-frequency, best bet uses wave trap.Preferential use LMF100 filtering chip in this unit, LMF100 filtering chip is primarily of two independently general high performance wave filter compositions, and each wave filter an external external clock and 2-4 resistance can realize the function of various single order and second order filter.Each wave filter comprises three output ports, and one of them port can be used for configuring the output of high pass, all-pass or trap, and two other is used as the logical and low pass output port of band respectively.The characteristic of wave filter and mid frequency can be arranged by the frequency of the proportioning of resistance and external clock.

The mode of operation of LMF100 filtering chip is determined by the outer meeting resistance method of different modes.Use the wave trap mode of operation of LMF100 filtering chip in the present embodiment, namely there are two wave filter MODE2, LMF100 inside, and its front ten pins and rear ten pins represent a wave filter respectively.Wave trap only needs a wave filter just can realize, and signal output part is positioned at 3 (18) feet.The connection of resistance determines the mode of operation of wave filter and the quality factor of wave trap, and under current connected mode, quality factor q is:

$Q=\frac{\sqrt{R_2/R_4+1}}{R_2/R_3}---\left(9\right)$

The frequency parameter of wave trap is then determined by outside input clock, and by the setting to 12 pins, the mid frequency of wave trap is f _cLK/ 100.Therefore when using this filtering chip, needing, a slice single-chip microcomputer is set separately and provide external clock for it.In this unit, what need filtering is 50Hz Hz noise, so provide f by single-chip microcomputer _cLK=5kHz.And according to the frequency response of wave trap under the different quality factor of LMF100 filtering chip, meet design requirement about known quality factor q=1.

For filtering chip provides the single-chip microcomputer of external clock only need generate the square-wave signal of a road 5kHz, in the present embodiment, select single-chip microcomputer 89C2051.Just the electrocardiosignal that can observe can be obtained after trap process is completed to signal.In the present embodiment, the circuit theory of wave trap as shown in Figure 5.

In the present embodiment, in cardioelectric monitor module, the specific design of diagnostic analysis module is as follows:

Cardioelectric monitor module is primarily of signal detection and diagnostic analysis two parts composition, and ECG signal sampling algorithm is the core of cardioelectric monitor system, directly affects the accuracy of diagnostic result.Because the electrocardiogram amplitude of normal person and interval have certain scope, if find that electrocardiogram has unusual waveforms, the analyzing and diagnosing of relevant disease can be made, therefore accurately detect that the characteristic parameter of electrocardiosignal is most important.ECG signal sampling algorithm of the present utility model mainly comprises the detection algorithm of the Wavelet Algorithm of electrocardiosignal, P ripple and QRS ripple, is mainly used in follow-up software diagnostics analysis module.

As shown in Figure 6, the principle of wavelet threshold denoising method be by signal after the pretreatment of data processing module through wavelet transformation multi-resolution decomposition again through each scale coefficient denoising, signal after denoising is carried out wavelet inverse transformation reconstruct, the final signal after denoising.The flow chart of wavelet threshold denoising method as shown in Figure 7.

Wavelet threshold is divided into two kinds:

(1) hard-threshold: its computing formula is as follows:

${\hat{d}}_{j_{0}k}=\left\{\begin{array}{cc}{d}_{j_{0}k}& \left|d_{j_{0}k}\right|\&GreaterEqual;{\mathrm{\&lambda;}}_j\\ 0& \left|d_{j_{0}k}\right|<{\mathrm{\&lambda;}}_j\end{array}\right.---\left(10\right)$

(2) soft-threshold: its computing formula is as follows:

${\hat{d}}_{j_{0}k}=\left\{\begin{array}{cc}\mathrm{sgn}\left(d_{j_{0}k}\right)\left(\right|d_{j_{0}k}|-{\mathrm{\&lambda;}}_j)& \left|d_{j_{0}k}\right|\&GreaterEqual;{\mathrm{\&lambda;}}_j\\ 0& \left|d_{j_{0}k}\right|<{\mathrm{\&lambda;}}_j\end{array}\right.---\left(11\right)$

Hard-threshold and soft-threshold function have four kinds of threshold value principles: be respectively and estimate principle, fixed threshold principle, heuristic threshold value principle and extreme value threshold value principle without partial likelihood.

(SURE) principle is estimated: a kind of software threshold estimator without partial likelihood.Suppose that signal x (k) is a discrete-time series k=1,2,3 ..., n, make signal y (k) be | x (k) | ascending sequence, then make y ₁(k)=y (k) ², then threshold value thr ₁computing formula as follows:

$\left\{\begin{array}{c}{y}_2\left(k\right)={\mathrm{\&Sigma;}}_{i=1}^k{y}_1\left(i\right)\\ r\left(k\right)=\frac{n-2k+y_2\left(k\right)+(n-k)y_1\left(k\right)}{n}\\ {\mathrm{thr}}_1=\sqrt{\min \left(r\right)}\end{array}\right.---\left(12\right)$

Fixed threshold principle: fixed threshold thr ₂computing formula as follows:

${\mathrm{thr}}_2=\sqrt{2\log \left(n\right)}---\left(13\right)$

In formula, n is the length of signal x (k).

Heuristic threshold value principle: if the length of signal x (k) is n, makes the size of variable eta and crit be respectively:

$\mathrm{eta}=\frac{{\left|\right|x\left|\right|}^2-n}{n},\mathrm{crit}=\frac{{[\log \left(n\right)/\log 2]}^{1.5}}{\sqrt{n}}---\left(14\right)$

Then threshold value thr ₃be calculated as follows:

Extreme value threshold value principle: if the length of signal x (k) is n, then threshold value thr ₄be calculated as follows:

${\mathrm{thr}}_3=\left\{\begin{array}{cc}0& n\&le;32\\ 0.3936+\frac{0.1829\log \left(n\right)}{\log 2},& n>32\end{array}\right.---\left(16\right)$

Estimate and extract the baseline of electrocardiosignal drift, then it is removed from primary signal, obtain the signal after filtering.The expression formula of gauge signal denoising effect:

$\left\{\begin{array}{c}\mathrm{SNR}=20\&times;{\log }_{10}\left(\frac{{\mathrm{\&Sigma;}}_{i=1}^N{y}_i^2}{{\mathrm{\&Sigma;}}_{i=1}^N{(x_i-y_i)}^2}\right)\\ \mathrm{MSE}=\sqrt{\frac{1}{N}{\mathrm{\&Sigma;}}_{i=1}^N{(y_i-x_i)}^2}\end{array}\right.---\left(17\right)$

X in formula _iexpression standard primary signal, y _irepresent estimated signal after treatment, N represents the hits of electrocardiosignal.Wherein, signal to noise ratio snr is larger and mean square error MSE is less, illustrates that denoising effect is better.As long as Rational choice wavelet basis and threshold value, denoising effect is generally better than the denoising effect of wavelet function feedback algorithm.

In order to solve P ripple in electrocardiosignal, T ripple signal is complicated, faint, identification difficulty is large and recognizer execution efficiency is low and problem that is that easily lost efficacy, the fast characteristic of execution speed is analyzed because small echo carries out time and frequency zone to signal, propose and lifting wavelet transform is combined with calculus of differences, structure utilizes Lifting Wavelet to electrocardiosignal denoising, the composite algorism utilizing calculus of finite differences to identify P ripple, T ripple in the low frequency signal of the corresponding level of reconstruct.

The effect of diagnostic analysis module to combine after denoising by the electrocardiosignal received with Traditional Wavelet level discharge rating Denoising Algorithm by wavelet threshold denoising method, carry out identification and the location of R ripple, when detecting R crest value point, adopt forward R ripple and the mode of being inverted R ripple separate detection, improve the accuracy rate that R ripple detects.Utilize difference, amplitude, the method that position is combined together is to determine R crest value point position.Have higher recall rate like this, and algorithm is simple, for real-time detection, is easy to realize.Detect R ripple algorithm realization as follows:

(1) detection algorithm of R ripple: comprise the following steps:

1. initial detecting threshold value is determined

The initialization of detection threshold approximately needs the electrocardiogram (ECG) data (electrocardiosignal after data processing module process) of 30s.Concrete methods of realizing is as follows:

S1: take away the electrocardiogram (ECG) data in beginning 30s, is divided into 10 sections by the electrocardiogram (ECG) data of 30s, and has a QRS wave group in ensureing every section at least, i.e. every section of 3s;

S2: calculate forward difference value by following formula (18-1) and (18-2), obtains the difference maximum of every section;

Δy(n)＝y(n+1)-y(n) (18-1)

Δy(n)＞δ (18-2)

S3: by the difference maximum sequence of trying to achieve in step S2, remove maximum and minima, ask the arithmetic mean of instantaneous value of residue difference maximum, be designated as Δ _m0, and determine that initial threshold is by data simulation test:

${\mathrm{\&delta;}}_{10}=\frac{2}{5}{\mathrm{\&Delta;}}_{m0},{\mathrm{\&delta;}}_{20}=\frac{2}{5}{\mathrm{\&Delta;}}_{m0}+2,{\mathrm{\&delta;}}_{30}=\frac{2}{9}{\mathrm{\&Delta;}}_{m0}---\left(19\right)$

S4: after 8 R ripples being detected, tamper detection threshold value;

Threshold value modification method is as follows: establish Δ _mibe the meansigma methods of the maximum difference of front 8 the QRS wave groups comprising current QRS wave group, new detection threshold computing formula is:

${\mathrm{\&delta;}}_{1i}=\frac{1}{4}{\mathrm{\&Delta;}}_{\mathrm{mi}}+C_1,{\mathrm{\&delta;}}_{2i}=\frac{1}{4}{\mathrm{\&Delta;}}_{\mathrm{mi}}+C_2,{\mathrm{\&delta;}}_{3i}=\frac{1}{8}{\mathrm{\&Delta;}}_{\mathrm{mi}}+C_3---(20-1)$

Wherein C ₁, C ₂, C ₃be defined as:

$C_1=\frac{1}{10}{\mathrm{\&Delta;}}_{m0},C_2=\frac{1}{10}{\mathrm{\&Delta;}}_{m0}+2,C_3=\frac{1}{10}{\mathrm{\&Delta;}}_{m0}---(20-2)$

In formula: Δ _mirepresent the maximum difference meansigma methods of 8 QRS wave groups, i represents the number of the R ripple that current detection arrives, i=8,9,10, C ₁, C ₂, C ₃all constants.

2. the detection of forward R ripple: comprise the following steps:

S1: according to step S1 in 1., S2, S3 and formula (18-1), (18-2), (19), calculate and determine δ ₁₀, δ ₂₀, δ ₃₀value;

S2: determine R crest value point: forward difference is done to suspect signal, uses δ ₁₀, δ ₂₀, δ ₃₀detect R ripple, if continuous two difference values are all greater than δ ₁₀and δ ₂₀, and negative sense difference is there is in 100ms afterwards, and the absolute value of negative sense difference is greater than δ ₃₀, then this suspect signal is R crest value point;

S3: repeat step S2, determine 8 R crest value points, revise threshold value according to formula (20-1) and formula (20-2), continue to detect with new detection threshold, if detect, new R ripple just revises threshold value, then continues to detect the next one, until ED.

3. the detection of R ripple is inverted:

The design detects from position and amplitude two aspect and is inverted R ripple.First following explanation is done to sign flag: R _irepresent current R ripple position; PR _irepresent current R wave amplitude; RR _irepresent the RR interval of current R ripple and last adjacent R ripple; represent the meansigma methods of 3 normal RR-intervals above; represent 3 normal R wave amplitude averages above.

If current RR interval RR _imeet: then may there is following situation: the 1. undetected R ripple of forward; 2. undetected inverted R ripple; 3. there is bradycardia, or stopped fighting.For head it off, specific practice is as follows:

1. at pR _i-1after (representing the previous R wave amplitude of current R ripple) maximum point is found, if this amplitude satisfies condition in scope: this point is undetected forward R ripple.

If 2. exist the maximum point found in scope does not meet then illustrate and occurred the inverted situation of R ripple, in this region, find amplitude smallest point be denoted as H _d, get minimum amplitude H _dwith the forward R wave amplitude H be front adjacent _zmeansigma methods, the height H of baseline _j(namely ), then think and an inverted R ripple detected.

If 3. exist do not find maximum value or minimum value point in scope, then think and may occur bradycardia or stop fighting, then this region does not have R ripple.

The form of ecg wave form often can change, by the impact of many factors.In order to solve the change of threshold value energy adaptation signal, the R ripple detection algorithm adopted in the present embodiment can not only detect is inverted R ripple, and can detect undetected R ripple.

(2) Q ripple, S ripple detection algorithm:

The concrete detecting step of Q, S ripple is as follows:

(1) adopt forward difference to ask first derivative to pretreated electrocardiosignal y (n), and adopt backward difference to ask second dervative on this basis, that is:

B(n)＝y(n+1)-y(n) (21)

C(n)＝B(n)-B(n-1)＝y(n+1)-2y(n)+y(n-1) (22)

(2) respectively square operation is carried out to first-order difference and second differnce, that is:

B ₁(n)＝[B(n)] ²(23)

C ₁(n)＝[C(n)] ²(24)

This is equivalent to and has carried out a rectification and Image magnify to signal, and the radio-frequency component of electrocardiosignal is highly improved, and process reaches the object of smooth commutation like this.

(3) formula (23) and direct addition of formula (24) are obtained new data sequence:

D(n)＝B ₁(n)-B ₁(n) (25)

(4) now, centered by the position of R crest value point, find Local modulus maxima, be designated as b and c in the scope of its forward and backward 50ms, their amplitude is designated as H respectively _band H _c, using 1/30 of these 2 amplitudes as threshold value, that is:

${\mathrm{Th}}_1=\frac{1}{30}{H}_b,{\mathrm{Th}}_2=\frac{1}{30}{H}_c---\left(26\right)$

In formula: Th ₁represent the threshold value determining Q ripple starting point, Th ₂represent the threshold value determining S ripple terminal.Data between 30ms to b before b point is put successively with threshold value Th ₁compare, if continuous 3 points meet:

D(n)≥Th ₁(27)

Then think that first is not less than Th ₁point be the starting point of Q ripple.In like manner with 30ms after c point for terminal, the data between 30ms to c after c point is put successively with threshold value Th ₂compare, if continuous 3 points meet:

D(n)≥Th ₂(28)

Then think that first is not less than Th ₂point be the terminal of S ripple.Data between Q ripple starting point and the position of S ripple terminal are put 1, and remainder data sets to 0 and just can obtain square chart.

Above-mentioned embodiment is intended to illustrate that this utility model can be professional and technical personnel in the field and realizes or use; modifying to above-mentioned embodiment will be apparent for those skilled in the art; therefore this utility model includes but not limited to above-mentioned embodiment; any these claims or description of meeting describes; meet and principle disclosed herein and novelty, the method for inventive features, technique, product, all fall within protection domain of the present utility model.

Claims (8)

1. the electrocardiogram monitor system based on Internet of Things, comprise electrocardiogram acquisition module, main intelligent mobile terminal, cardioelectric monitor module, it is characterized in that, described electrocardiogram acquisition module is connected with described main intelligent mobile terminal, described main intelligent mobile terminal and the two-way communication of described cardioelectric monitor module, described electrocardiogram acquisition module comprises the signals collecting end connected successively, data processing module, signal transmission module, described main intelligent mobile terminal is sent to through described signal transmission module after the ECG's data compression that described signals collecting end collects by described data processing module, the electrocardiosignal of acquisition is sent to described cardioelectric monitor module by described main intelligent mobile terminal, described cardioelectric monitor module comprises remote communication module, diagnostic analysis module, ecg signal data storehouse, described remote communication module is used for communicating with main intelligent mobile terminal, and the electrocardiosignal received is sent to the storage of described ecg database, the signal that described diagnostic analysis module extracts the reception of described ecg database carries out diagnostic analysis, after diagnostic analysis, result is sent to main intelligent mobile terminal by described remote communication module.

2. a kind of electrocardiogram monitor system based on Internet of Things according to claim 1, it is characterized in that, described signals collecting end comprises core signal sensor, pulse transducer, oximetry sensor, temperature sensor, is respectively used to detect electrocardiosignal, pulse frequency, blood oxygen saturation and body temperature.

3. a kind of electrocardiogram monitor system based on Internet of Things according to claim 1, it is characterized in that, described data processing module comprises the front end amplifying unit, signal filtering unit, rear end amplifying unit, wave trap, the A/D converter that connect in turn, and described front end amplifying unit adopts precision amplifier AD620 circuit; Described signal filtering unit adopts Butterworth filter; Described rear end amplifying unit adopts a scaling circuit to connect an add circuit by a voltage follower; Described wave trap adopts 50Hz wave trap chip LMF100, and uses chip LMF100 wave trap mode of operation.

4. a kind of electrocardiogram monitor system based on Internet of Things according to claim 1, it is characterized in that, described cardioelectric monitor module also comprises case search module and system help module, described case search module helps doctor to analyze for inquiring about patient history's case, and described system help module is used for User Defined and arranges.

5. a kind of electrocardiogram monitor system based on Internet of Things according to claim 1, is characterized in that, described signal transmission module is Bluetooth communication modules or Zigbee transmission module.

6. a kind of electrocardiogram monitor system based on Internet of Things according to claim 1, it is characterized in that, described electrocardiogram monitor system also comprises secondary intelligent mobile terminal, and described secondary intelligent mobile terminal is by described remote communication module and described cardioelectric monitor module communication.

7. a kind of electrocardiogram monitor system based on Internet of Things according to claim 1, is characterized in that, described remote communication module is GPRS wireless transport module or 3G wireless transport module or 4G wireless transport module or Socket data transmission module.

8. a kind of electrocardiogram monitor system based on Internet of Things according to claim 1, is characterized in that, described main intelligent mobile terminal and described secondary intelligent mobile terminal are smart mobile phone or panel computer.