CN105193423A - Non-invasive blood glucose detection method, device and system - Google Patents
Non-invasive blood glucose detection method, device and system Download PDFInfo
- Publication number
- CN105193423A CN105193423A CN201510623388.5A CN201510623388A CN105193423A CN 105193423 A CN105193423 A CN 105193423A CN 201510623388 A CN201510623388 A CN 201510623388A CN 105193423 A CN105193423 A CN 105193423A
- Authority
- CN
- China
- Prior art keywords
- pulse wave
- wave signal
- detection
- amplitude
- signal
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Landscapes
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
A non-invasive blood glucose detection method comprises the steps that a pulse wave signal is obtained, and blood glucose concentration information of a sample to be detected is borne by the pulse wave signal; the obtained pulse wave signal is detected through first detection to obtain an initial detection value of blood glucose concentration; the obtained pulse wave signal is detected through second detection to obtain a detection interval which the blood glucose concentration belongs to; whether the detection belongs to the detection interval or not is judged, and if yes, the initial detection value serves as the blood glucose concentration value borne by the pulse wave signal. Due to the fact that judgment is conducted on the interval which the initial detection value belongs to, interference to the blood glucose concentration caused by other components can be effectively reduced, and the precision and the accuracy of blood glucose concentration detection are improved. The invention further discloses a non-invasive blood glucose detection device and system.
Description
Technical field
The present invention relates to medical detection field, be specifically related to a kind of Noninvasive Blood Glucose Detection Methods, Apparatus and system.
Background technology
Along with socioeconomic development, the living standard of people progressively promotes, but the thing followed also has some diseases, and presently, diabetes have become one of principal disease of modern society's harm humans health.Blood glucose too high or too low, not only affects the metabolism of patient, and also have some complication, as cardiovascular disease and neuropathy, these have very large threat for the healthy of patient.According to the report of World Health Organization (WHO), will have 300,000,000 diabeticss to the whole world in 2025, wherein the diabetics of China also will have greatly.In recent years, the patient of diabetes does not exist only in the middle of some old peoples, for some youngsters, starts to occur diabetic disorders yet.Diabetes are a kind of chronic diseases, are difficult to reach good effect by disposable treatment, so diabetics needs the blood sugar level understanding oneself in real time accurately.
But at present for the method for blood sugar test, hospital or patient oneself at home, be all adopt the blood sugar detecting method having wound, namely directly extract blood samples of patients, detect the blood sugar level of patient according to electrochemical method.This detection method causes certain physiology painful to patient, and repeatedly draws blood and easily cause infection.Further, electrochemical reaction reagent paper is expensive, for diabetics, is also a kind of larger financial burden.Noninvasive dynamics monitoring can eliminate the misery that patient detects, and frequently can detect, improve patients ' life quality.
At present, there is many Noninvasive Blood Glucose Detection Methods, the detection method of concentration of glucose in blood of human body is widely used in the research of noninvasive dynamics monitoring wherein based on light.But except glucose also exists other compositions many in blood, limit the precision of blood sugar test.The precision how improving blood sugar test becomes problem demanding prompt solution.
Summary of the invention
The application provides a kind of Noninvasive Blood Glucose Detection Methods, Apparatus and system, with improve to blood sugar concentration detect precision and accuracy.
According to first aspect, provide a kind of Noninvasive Blood Glucose Detection Methods in a kind of embodiment, comprising:
Obtain pulse wave signal, pulse wave signal carries sample to be tested blood sugar concentration information; The first pulse wave signal detecting acquisition is adopted to obtain the initial detecting value of blood sugar concentration; Adopting second to detect the pulse wave signal obtained obtains between the detection zone belonging to blood sugar concentration; Judge whether detected value belongs between detection zone, if initial detecting value is in detection zone, then the blood glucose concentration value of this initial detecting value then for carrying in pulse wave signal.
According to second aspect, a kind of noninvasive dynamics monitoring device is provided in a kind of embodiment, comprises:
Signal acquisition module, for obtaining pulse wave signal, pulse wave signal carries sample to be tested blood sugar concentration information; Preliminary detection module, detects for adopting the first detection the initial detecting value that the pulse wave signal obtained obtains blood sugar concentration; Interval detection module, detects for adopting the second detection the pulse wave signal obtained and obtains between the detection zone belonging to blood sugar concentration; Judge module, for judging whether detected value belongs between detection zone, if initial detecting value is in detection zone, then the blood glucose concentration value of this initial detecting value then for carrying in pulse wave signal.
According to the third aspect, provide a kind of noninvasive system for detecting blood sugar in a kind of embodiment, comprising:
Signal picker, for gathering and exporting pulse wave signal; The above-mentioned noninvasive dynamics monitoring device be connected with signal picker.
According to the Noninvasive Blood Glucose Detection Methods of above-described embodiment, detecting owing to adopting the second detection the pulse wave signal obtained obtains between the detection zone belonging to blood sugar concentration, and judge that the first detection detects the initial detecting value that the pulse wave signal that obtains obtains blood sugar concentration and whether belongs between detection zone, when initial detected value is in described detection zone, then this initial detecting value is blood glucose concentration value, judged by interval belonging to initial detecting value, effectively can reduce the interference that other composition causes blood sugar concentration, improve precision and the accuracy of blood sugar concentration detection.
Accompanying drawing explanation
Fig. 1 is a kind of noninvasive system for detecting blood sugar structural representation disclosed in the present embodiment;
Fig. 2 is noninvasive dynamics monitoring apparatus structure schematic diagram disclosed in the present embodiment;
Fig. 3 is a kind of Noninvasive Blood Glucose Detection Methods flow chart disclosed in the present embodiment;
A kind of waveform schematic diagram of Fig. 4 pulse wave signal disclosed in the present embodiment;
Fig. 5 is a kind of method flow diagram of the present embodiment pulse wave signal characteristics extraction;
Fig. 6 is a kind of schematic diagram of the present embodiment pulse wave signal wavelet transformation;
Fig. 7 is a kind of example flow diagram that the present embodiment blood sugar concentration detects;
Fig. 8 is the present embodiment blood sugar concentration Detection results schematic diagram.
Detailed description of the invention
By reference to the accompanying drawings the present invention is described in further detail below by detailed description of the invention.
Please refer to Fig. 1, a kind of noninvasive system for detecting blood sugar disclosed in the present embodiment, comprising: signal picker 1 and noninvasive dynamics monitoring device 2, wherein: signal picker 1 is for gathering person to be measured and exporting pulse wave signal.Noninvasive dynamics monitoring device 2 is connected with signal picker 1, detects for the pulse wave signal exported signal picker 1, to obtain the blood sugar concentration of person to be measured.
In a preferred embodiment, can also be connected with front end signal treatment circuit 2 between signal picker 1 and noninvasive dynamics monitoring device 2, front end signal treatment circuit 2 carries out pre-process for the pulse wave signal exported signal picker 1 collection.The signal such as exported signal picker 1 is changed, enlarge leadingly and filtering etc., and certainly, front end signal treatment circuit 2 can also provide power supply for signal picker 1.Particularly, when carrying out filtering to pulse wave signal, the noise in pulse wave signal and DC component can be removed, leaving required AC compounent; Pulse wave signal is amplified and analog digital conversion, as an example, 12 ADC can be adopted to sample, sample frequency can be such as 1KHz, to obtain the digital signal of pulse wave signal, and noninvasive dynamics monitoring device 2 is sent to by this digital signal to carry out follow-up process.
In a particular embodiment, pulse wave signal can be photoplethysmographic signal, also can be bioimpedance signal or pressure sensor signal.In the present embodiment, pulse wave signal is preferably photoplethysmographic signal (photoplethysmograph, PPG).Signal picker 1 comprises: detecting position and the light source 11 and the light-sensitive element 12 that are positioned at detecting position two ends, wherein, the place of detecting position for providing blood sugar concentration to detect, such as, may be used for placing the intensive tissue of the human capillary vessels such as human finger finger tip, ear-lobe or wrist; Light source 11 is for sending the optical signal at least comprising near infrared light to detecting position, as preferred embodiment, light source 11 can comprise multiple near-infrared luminous pipe in 800nm-1100nm wave band, and near-infrared luminous pipe peak wavelength deviation is ± 10nm, and radiant power is greater than 3mW; Light-sensitive element 12 is for receiving the optical signal after detecting position, and be converted into signal of telecommunication output, in a particular embodiment, the wave band that light-sensitive element 12 receives can be arranged, the optical signal wave band received to make light-sensitive element 12 is near infrared light wave band, and particularly, photosensitive receiving tube peak wavelength deviation is ± 10nm, photocurrent is greater than 10uA, and the peak wavelength deviation that photosensitive receiving tube 12 receives is less than ± 10nm.It should be noted that, in a preferred embodiment, after the wave band received light-sensitive element 12 is arranged, light source 11 can comprise the optical signal of other wave band, but the optical signal that demand fulfillment light source 11 sends at least comprises near infrared light.
The improvements of the present embodiment are also noninvasive dynamics monitoring device 2, please refer to Fig. 2, noninvasive dynamics monitoring apparatus structure schematic diagram disclosed in the present embodiment, this noninvasive dynamics monitoring device comprises: signal acquisition module 21, preliminary detection module 22, interval detection module 23 and judge module 24, wherein:
Signal acquisition module 21 is for obtaining pulse wave signal, and described pulse wave signal carries sample to be tested blood sugar concentration information; The initial detecting value of preliminary detection module 22 for adopting the pulse wave signal of the first detection detection acquisition to obtain blood sugar concentration; Interval detection module 23 obtains between the detection zone belonging to blood sugar concentration for adopting the second detection to detect the pulse wave signal obtained; Judge module 24 is for judging whether detected value belongs between detection zone, if described initial detecting value is in described detection zone, then and the blood glucose concentration value of this initial detecting value then for carrying in described pulse wave signal.
In a preferred embodiment, this noninvasive dynamics monitoring device also comprises: characteristics extraction module 25, and it is for extracting the eigenvalue of obtained pulse wave signal.Preliminary detection module is used for the initial detecting value that the eigenvalue of Detection and Extraction obtains blood sugar concentration; The eigenvalue that interval detection module is used for Detection and Extraction obtains between the detection zone of blood sugar concentration.
Please refer to Fig. 3, based on above-mentioned noninvasive dynamics monitoring device, the present embodiment also discloses a kind of Noninvasive Blood Glucose Detection Methods, for this Noninvasive Blood Glucose Detection Methods comprises the steps:
Step 100, pulse wave signal obtains.Alleged pulse wave signal carries sample to be tested blood sugar concentration information.In the present embodiment, pulse wave signal is preferably photoplethysmographic signal (photoplethysmograph, PPG).Glucose molecule formula contains multiple O-H, C-H chemical bond, there is absorption peak at 800nm-1100nm wave band and absorb peak valley, absorption peak wavelength is as critical wavelength, this wavelength is the peak wavelength of blood glucose to near-infrared absorption, the absorbing state of blood glucose near infrared light can be reflected, absorb peak valley wavelength as reference wavelength.The photoplethysmographic that critical wavelength produces not only contains the absorption information of blood glucose near infrared light, and comprises the absorption information of other materials in blood near infrared light.Combine with reference to wavelength and critical wavelength and carry out modeling, effectively can reduce the impact of other materials on near-infrared absorption.In the present embodiment, another major reason selecting wavelength to be less than the near infrared light of 1500nm is because these wavelength easily obtain, be all some common near-infrared wavelengths, such as typical gallium arsenide diode just can reach our demand, reduces the cost of noninvasive dynamics monitoring.
Step 200, initial detecting.The first pulse wave signal detecting acquisition is adopted to obtain the initial detecting value of blood sugar concentration.In a particular embodiment, the first detection can be carried out by neutral net (such as prediction neural network) to the pulse wave signal obtained, thus obtain the initial detecting value of blood sugar concentration.Certainly, it should be noted that, in the specific implementation, should first train neutral net, the training of neutral net can be online, can be also off-line, in the present embodiment, is preferably off-line training.
Step 300, interval detection.Adopt second to detect the pulse wave signal obtained to obtain between the detection zone belonging to blood sugar concentration.Similarly, in a particular embodiment, also the second detection can be carried out by neutral net to the pulse wave signal obtained, to obtain the interval at the pulse wave signal place obtained.Particularly, for human body, blood sugar concentration can be 1 carry out demarcation interval [3,4], [4,5], [5,6] according to step-length ... [24,25], thus human blood glucose concentration scope 3 ~ 25 will be contained be divided into multiple interval.Similarly, in the specific implementation, should first train neutral net, the training of neutral net can be online, can be also off-line, in the present embodiment, is preferably off-line training.
It should be noted that, in the present embodiment, the not execution sequencing of conditioning step " initial detecting " and step " interval detection ".
Step 400, judges whether detected value belongs between detection zone.Detect the initial detecting value and second obtained to detect to compare between the detection zone that obtains judge first, if the result judged be initial detecting value in detection zone, then the blood glucose concentration value of this initial detecting value then for carrying in pulse wave signal.As an example, such as to adopt first to detect the initial detecting value that obtains be 4.6, and be [4,5] between the detection zone adopting second to detect to obtain, then explanation initial detecting value belongs between this detection zone; Otherwise if adopting second to detect between the detection zone that obtains is [5,6], [3,4] or [9,10] etc., then explanation initial detecting value does not belong between this detection zone.Certainly, if the result judged be initial detecting value not in detection zone, then this initial detecting value differs comparatively greatly with actual blood glucose concentration value, can give up and change initial detecting value.
In a preferred embodiment, after execution pulse wave signal obtains, can also comprise:
Step 500, characteristics extraction.Extract the eigenvalue of the pulse wave signal obtained.Now, when performing step " initial detecting " and " interval detect ", should preferably using the eigenvalue of extraction as detected object.Particularly, detect in employing first pulse wave signal obtained and obtain in the initial detecting value of blood sugar concentration, the eigenvalue of Detection and Extraction obtains the initial detecting value of blood sugar concentration; In between employing second detects detection zone that the pulse wave signal that obtains obtains belonging to blood sugar concentration, the eigenvalue of Detection and Extraction obtains between the detection zone of blood sugar concentration.
In a particular embodiment, the eigenvalue extracted can be the amplitude integration in pulse wave signal unit period, also can be main wave-wave peak and main ripple rise time ratio and the main ripple relative altitude of subwave, in a preferred embodiment, please refer to Fig. 4, in one-period, the eigenvalue of extraction is the main peak amplitude h of pulse wave signal main crest P, main trough A, secondary wave crest T and secondary trough V
p, main paddy amplitude h
a, secondary peak amplitude h
t, secondary paddy amplitude h
v, main peak-secondary paddy interval t1, main peak-secondary peak interval t2, main peak-main paddy interval t3 and adjacent main peak time interval t4.
In a particular embodiment, the mode of wavelet transformation can be adopted when the eigenvalue extracting pulse wave signal: main peak amplitude h
p, main paddy amplitude h
a, secondary peak amplitude h
t, secondary paddy amplitude h
v, main peak-secondary paddy interval t1, main peak-secondary peak interval t2, main peak-main paddy interval t3 and adjacent main peak time interval t4, particularly, please refer to Fig. 5, comprise the steps:
Step 510, wavelet transformation.Wavelet transformation is carried out to the pulse wave signal obtained and obtains wavelet transformation sequence.Before wavelet transformation, first denoising can be carried out to the pulse wave signal (such as PPG) obtained, again Stationary Wavelet Transform is carried out to the purified signal after de-noising, obtain wavelet transformation sequence according to income value after Stationary Wavelet Transform, as shown in Figure 6.
Step 520, searches maximum.Search the modulus maximum in wavelet transformation sequence.After obtaining wavelet transformation sequence, suitable predetermined threshold value can be determined, to search the modulus maximum meeting predetermined threshold value, in this enforcement, the modulus maximum in wavelet transformation sequence comprises positive modulus maximum, negative modulus maximum and relevant secondary modulus maximum.
Step 530, extracts eigenvalue.Eigenvalue main peak amplitude h is extracted according to modulus maximum
p, main paddy amplitude h
a, secondary peak amplitude h
t, secondary paddy amplitude h
v, and obtain eigenvalue main peak-secondary paddy interval t1, main peak-secondary peak interval t2, main peak-main paddy interval t3 and adjacent main peak time interval t4 according to the position of main crest P, main trough A, secondary wave crest T and secondary trough V.
In a particular embodiment:
For main wave-wave peak P, in wavelet transformation sequence, after filtering out the negative modulus maximum exceeding predetermined threshold value, after finding, be close to the positive modulus maximum of this negative modulus maximum, connect negative modulus maximum and positive modulus maximum adjacent thereafter, obtain the first intersection point with axis of abscissas; Search the maximum point of signal amplitude at the first near intersections, this point is main crest P, and the amplitude of this point is main peak amplitude h
p, particularly, can centered by this first intersection point preset among a small circle in search, preset can rule of thumb determine among a small circle.
For main wave-wave paddy A, in wavelet transformation sequence, after filtering out the negative modulus maximum exceeding predetermined threshold value, before finding, be close to the positive modulus maximum of this negative modulus maximum, connect negative modulus maximum and positive modulus maximum adjacent before it, obtain the second intersection point with axis of abscissas; Search the minimum point of signal amplitude at the second near intersections, this point is main paddy A, and the amplitude of this some correspondence is main paddy amplitude h
a; Particularly, can centered by this second intersection point preset among a small circle in search, preset can rule of thumb determine among a small circle.
For secondary wave crest T, in wavelet transformation sequence, connect time negative norm maximum and positive modulus maximum adjacent thereafter, obtain the 3rd intersection point with axis of abscissas, search the maximum point of signal amplitude at the 3rd near intersections, the amplitude of this point is time peak amplitude h
t.Particularly, filter out the negative modulus maximum exceeding predetermined threshold value, back to back negative norm maximum (this point is defined as time negative norm maximum point) is found after the negative modulus maximum that this exceedes predetermined threshold value, secondary negative norm maximum point should be no more than predetermined threshold value, find positive modulus maximum being close to after this negative norm maximum point, connect time negative norm maximum and positive modulus maximum adjacent thereafter, obtain the 3rd intersection point with axis of abscissas; Search the maximum point of signal amplitude at the 3rd near intersections, this point is secondary wave crest T, and the amplitude of this point is time peak amplitude h
t, particularly, can centered by this first intersection point preset among a small circle in search, preset can rule of thumb determine among a small circle.
For secondary trough A, in wavelet transformation sequence, connect time negative norm maximum and mould secondary maximum value positive before it, obtain the 4th intersection point with axis of abscissas, search the maximum point of signal amplitude at the 4th near intersections, the amplitude of this point is time paddy amplitude h
v.Filter out the negative modulus maximum exceeding predetermined threshold value, back to back negative norm maximum (this point is defined as time negative norm maximum point) is found after the negative modulus maximum that this exceedes predetermined threshold value, secondary negative norm maximum point should be no more than predetermined threshold value, find positive modulus maximum being close to before this negative norm maximum point, connect time negative norm maximum and positive modulus maximum adjacent before it, obtain the 4th intersection point with axis of abscissas; Search the maximum point of signal amplitude at the 4th near intersections, this point is time trough V, and the amplitude of this point is time peak amplitude h
v, particularly, can centered by the 4th intersection point preset among a small circle in search, preset can rule of thumb determine among a small circle.
It should be noted that, behind the position finding main peak P, main paddy A, secondary peak T and secondary paddy V, those skilled in the art can try to achieve main peak-secondary paddy interval t1, main peak-secondary peak interval t2, main peak-main paddy interval t3 and adjacent main peak time interval t4, do not repeat them here.
Please refer to Fig. 7, understand the technical scheme of the present embodiment for ease of those skilled in the art, hereafter for gather pulse wave signal for photoplethysmographic (PPG), be described in conjunction with concrete example:
Utilize the PPG training signal of standard to train prediction neural network and Classification Neural respectively, obtain the prediction neural network net1 that trains and Classification Neural net2 respectively.In a particular embodiment, the PPG eigenvalue training signal of standard can be adopted to train, for prediction neural network, input is the eigenvalue of photoplethysmographic, the wound that has of correspondence is detected blood glucose concentration value as output, then uses such as MATLAB Neural Network Toolbox to train, wherein, it is three layers that neutral net as training hides the number of plies, and the number of hidden layer node is 15-25.For Classification Neural, input is the eigenvalue of photoplethysmographic, the blood glucose concentration value of correspondence is classified, as blood glucose concentration value belongs to interval [3,4] first kind is designated as, belong to interval [4,5] Equations of The Second Kind is designated as, belong to interval [5,6] and be designated as the 3rd class, by that analogy as exporting, until the interval [3 of all blood glucose values will occurred in training process, 25] be included, then use such as MATLAB Neural Network Toolbox to train, network the highest for wherein discrimination is used for modeling.
Gather PPG signal, in this example, select ear-lobe or finger tip as the position of extracting PPG signal, ear-lobe or finger tip are placed the detecting position of signal picker 1.The blood of finger tip and ear-lobe is abundanter, along with the periodic cycle of heart, the photosignal that photoelectric sensor can detect periodically changes, in order to obtain stable photoplethysmographic, need external influence factor is dropped to minimum or becomes controlled, such as ambient temperature and humidity, to sum up, ear-lobe or finger tip are the positions of extraction PPG signal the most suitable.Photoplethysmographic is obtained through near infrared spectral transmission human skin tissue or through human skin tissue reflection.In a preferred embodiment, the mode that signals collecting obtains multi-group data by one-shot measurement improves data volume, each measurement needs 2min, institute's image data is divided into many groups, often organize PPG data to deliver to prediction-classification forecast model respectively and detect, can be used for forecast error analysis and the precision raising of forecast model.
Dual-network regression analysis: the eigenvalue extracting PPG acquired signal, and feeding prediction neural network net1 and Classification Neural net2 carries out the first detection and the second detection respectively.Utilize prediction neural network net1 to carry out the first detection and obtain initial detecting value (blood glucose value R1), utilize Classification Neural net2 to carry out the second detection and obtain (the interval R2 of blood glucose) between detection zone.Utilize R1 to judge belonging to blood glucose value interval, if belong to R2 between R1 location, so then think that R1 is correct, retention forecasting result R1; Otherwise, then think and abandon the mistake that predicts the outcome the R1 that predicts the outcome, predict next time.
In order to improve the accuracy predicted the outcome further, in noninvasive dynamics monitoring, enough photoplethysmographic can be collected each time, will the result of multiple data prediction be obtained, remove maximum and minima to a blood glucose prediction result, then average, predict the outcome as final.Fig. 8 illustrates the interpretation of result of the present embodiment blood glucose prediction, and Fig. 8 shows, adopts detection method disclosed in the present embodiment to detect the blood glucose concentration value error obtained less.
Noninvasive Blood Glucose Detection Methods disclosed in the present embodiment and system, detecting owing to adopting the second detection the pulse wave signal obtained obtains between the detection zone belonging to blood sugar concentration, and judge that the first detection detects the initial detecting value that the pulse wave signal that obtains obtains blood sugar concentration and whether belongs between detection zone, when initial detected value is in described detection zone, then this initial detecting value is blood glucose concentration value, judged by interval belonging to initial detecting value, effectively can reduce the interference that other composition causes blood sugar concentration, improve precision and the accuracy of blood sugar concentration detection.
It will be appreciated by those skilled in the art that, in above-mentioned embodiment, all or part of step of various method can be carried out instruction related hardware by program and completes, this program can be stored in a computer-readable recording medium, and storage medium can comprise: read only memory, random access memory, disk or CD etc.
More than applying specific case to set forth the present invention, just understanding the present invention for helping, not in order to limit the present invention.For those skilled in the art, according to thought of the present invention, some simple deductions, distortion or replacement can also be made.
Claims (10)
1. a Noninvasive Blood Glucose Detection Methods, is characterized in that, comprising:
Obtain pulse wave signal, described pulse wave signal carries sample to be tested blood sugar concentration information;
The first pulse wave signal detecting acquisition is adopted to obtain the initial detecting value of blood sugar concentration;
Adopting second to detect the pulse wave signal obtained obtains between the detection zone belonging to blood sugar concentration;
Judge whether detected value belongs between detection zone, if described initial detecting value is in described detection zone, then the blood glucose concentration value of this initial detecting value then for carrying in described pulse wave signal.
2. Noninvasive Blood Glucose Detection Methods as claimed in claim 1, is characterized in that, after acquisition pulse wave signal, also comprises: the eigenvalue extracting the pulse wave signal obtained;
Detecting in employing first pulse wave signal obtained obtains in the initial detecting value of blood sugar concentration, and the eigenvalue of Detection and Extraction obtains the initial detecting value of blood sugar concentration;
In between employing second detects detection zone that the pulse wave signal that obtains obtains belonging to blood sugar concentration, the eigenvalue of Detection and Extraction obtains between the detection zone of blood sugar concentration.
3. Noninvasive Blood Glucose Detection Methods as claimed in claim 2, it is characterized in that, the eigenvalue of described extraction comprises:
Main peak amplitude (the h of pulse wave signal
p), main paddy amplitude (h
a), secondary peak amplitude (h
t), secondary paddy amplitude (h
v), main peak time paddy interval (t1), main peak secondary peak interval (t2), the main paddy interval (t3) of main peak and adjacent main peak time interval (t4);
Or, the amplitude integration in pulse wave signal unit period;
Or, main wave-wave peak and main ripple rise time ratio and the main ripple relative altitude of subwave.
4. Noninvasive Blood Glucose Detection Methods as claimed in claim 3, is characterized in that, when the eigenvalue extracted comprises: the main peak amplitude (h of pulse wave signal
p), main paddy amplitude (h
a), secondary peak amplitude (h
t), secondary paddy amplitude (h
v), main peak time paddy interval (t1), main peak secondary peak interval (t2), the main paddy interval (t3) of main peak and adjacent main peak time interval (t4) time, the eigenvalue of the pulse wave signal that extraction obtains comprises:
Wavelet transformation is carried out to the pulse wave signal obtained and obtains wavelet transformation sequence;
Search the modulus maximum in wavelet transformation sequence;
Eigenvalue is extracted according to modulus maximum.
5. Noninvasive Blood Glucose Detection Methods as claimed in claim 4, is characterized in that, determine eigenvalue position according to the modulus maximum sequence that wavelet transformation obtains, and then extracts eigenvalue, comprising:
In wavelet transformation sequence, connect negative modulus maximum and positive modulus maximum adjacent thereafter, obtain the first intersection point with axis of abscissas; Search the maximum point of signal amplitude at the first near intersections, the amplitude of this point is main peak amplitude (h
p);
In wavelet transformation sequence, connect negative modulus maximum and positive modulus maximum adjacent before it, obtain the second intersection point with axis of abscissas; Search the minimum point of signal amplitude at the second near intersections, the amplitude of this point is main paddy amplitude (h
a);
In wavelet transformation sequence, connect time negative norm maximum and positive modulus maximum adjacent thereafter, obtain the 3rd intersection point with axis of abscissas; Search the maximum point of signal amplitude at the 3rd near intersections, the amplitude of this point is time peak amplitude (h
t);
In wavelet transformation sequence, connect time negative norm maximum and mould secondary maximum value positive before it, obtain the 4th intersection point with axis of abscissas; Search the maximum point of signal amplitude at the 4th near intersections, the amplitude of this point is time paddy amplitude (h
v).
6. the Noninvasive Blood Glucose Detection Methods as described in claim 1-5 any one, is characterized in that, described pulse wave signal is photoplethysmographic signal, bioimpedance signal or pressure sensor signal.
7. a noninvasive dynamics monitoring device, is characterized in that, comprising:
Signal acquisition module, for obtaining pulse wave signal, described pulse wave signal carries sample to be tested blood sugar concentration information;
Preliminary detection module, detects for adopting the first detection the initial detecting value that the pulse wave signal obtained obtains blood sugar concentration;
Interval detection module, detects for adopting the second detection the pulse wave signal obtained and obtains between the detection zone belonging to blood sugar concentration;
Judge module, for judging whether detected value belongs between detection zone, if described initial detecting value is in described detection zone, then the blood glucose concentration value of this initial detecting value then for carrying in described pulse wave signal.
8. noninvasive dynamics monitoring device as claimed in claim 7, is characterized in that, also comprise:
Characteristics extraction module, for extracting the eigenvalue of obtained pulse wave signal;
Described preliminary detection module is used for the initial detecting value that the eigenvalue of Detection and Extraction obtains blood sugar concentration; The eigenvalue that described interval detection module is used for Detection and Extraction obtains between the detection zone of blood sugar concentration.
9. a noninvasive system for detecting blood sugar, is characterized in that, comprising:
Signal picker, for gathering and exporting pulse wave signal;
The noninvasive dynamics monitoring device as claimed in claim 7 or 8 be connected with described signal picker.
10. noninvasive system for detecting blood sugar as claimed in claim 9, it is characterized in that, described signal picker comprises: detecting position and the light source and the light-sensitive element that are positioned at detecting position two ends, wherein,
The place of detecting position for providing blood sugar concentration to detect;
Light source is used for sending to detecting position the optical signal at least comprising near infrared light;
Light-sensitive element for receiving the optical signal after detecting position, and is converted into signal of telecommunication output; The optical signal wave band that described light-sensitive element receives is near infrared light wave band.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510623388.5A CN105193423A (en) | 2015-09-25 | 2015-09-25 | Non-invasive blood glucose detection method, device and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510623388.5A CN105193423A (en) | 2015-09-25 | 2015-09-25 | Non-invasive blood glucose detection method, device and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105193423A true CN105193423A (en) | 2015-12-30 |
Family
ID=54941662
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510623388.5A Pending CN105193423A (en) | 2015-09-25 | 2015-09-25 | Non-invasive blood glucose detection method, device and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105193423A (en) |
Cited By (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106901752A (en) * | 2017-01-23 | 2017-06-30 | 武汉市瑞达源科技有限公司 | Method for determining concentration of glucose in blood of human body matrix |
| CN108513665A (en) * | 2017-02-07 | 2018-09-07 | 华为技术有限公司 | The methods, devices and systems of user identity identification |
| TWI640297B (en) * | 2017-03-23 | 2018-11-11 | 國立交通大學 | Non-invasive blood glucose measuring device, method, and system with identification function |
| WO2018214299A1 (en) * | 2017-05-20 | 2018-11-29 | 深圳市前海安测信息技术有限公司 | Non-invasive blood glucose data collection apparatus |
| WO2018214298A1 (en) * | 2017-05-20 | 2018-11-29 | 深圳市前海安测信息技术有限公司 | Non-invasive blood glucose detection system and method |
| WO2018214297A1 (en) * | 2017-05-20 | 2018-11-29 | 深圳市前海安测信息技术有限公司 | Portable device and method for non-invasive blood glucose detection |
| CN109276258A (en) * | 2018-08-10 | 2019-01-29 | 北京大学深圳研究生院 | Blood glucose trend forecasting method, system and Medical Devices based on DTW |
| CN110141249A (en) * | 2019-06-18 | 2019-08-20 | 张远 | Woundless blood sugar monitoring method, system, equipment and medium based on PPG signal |
| CN110251115A (en) * | 2019-06-20 | 2019-09-20 | 张远 | PPG method for extracting signal, system, equipment and medium based on body surface video |
| CN110327058A (en) * | 2019-07-31 | 2019-10-15 | 清华大学 | A kind of non-invasive blood sugar instrument and blood sugar detecting method |
| CN111297374A (en) * | 2020-02-24 | 2020-06-19 | 京东方科技集团股份有限公司 | Physical sign parameter detection equipment and physical sign parameter detection method |
| CN111317484A (en) * | 2018-12-13 | 2020-06-23 | 三星电子株式会社 | Apparatus and method for estimating blood glucose |
| CN111599470A (en) * | 2020-04-23 | 2020-08-28 | 中国科学院上海技术物理研究所 | Method for improving near-infrared noninvasive blood glucose detection precision |
| CN111631733A (en) * | 2020-06-19 | 2020-09-08 | 浙江澍源智能技术有限公司 | Arterial blood spectrum detection method and device |
| WO2020228084A1 (en) * | 2019-05-13 | 2020-11-19 | 深圳六合六医疗器械有限公司 | Personalized blood sugar interval statistical method and device |
| CN112120711A (en) * | 2020-09-22 | 2020-12-25 | 博邦芳舟医疗科技（北京）有限公司 | Noninvasive diabetes prediction system and method based on photoplethysmography |
| JP2021045639A (en) * | 2020-12-22 | 2021-03-25 | 京セラ株式会社 | Electronic apparatus, estimation system, estimation method, and estimation program |
| CN113343200A (en) * | 2021-07-12 | 2021-09-03 | 武汉华星光电技术有限公司 | Electronic equipment unlocking system and electronic equipment |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2009128787A1 (en) * | 2008-04-16 | 2009-10-22 | Glucostats System Pte Ltd | Method and system for measuring a composition in a blood fluid |
| CN102217940A (en) * | 2011-06-14 | 2011-10-19 | 中国科学院长春光学精密机械与物理研究所 | Method for realizing no-wound human blood glucose measuring by utilizing volume pulse wave characteristic value |
| CN103690176A (en) * | 2013-12-26 | 2014-04-02 | 苏州大学 | Method for testing blood glucose in non-invasive manner |
| CN104224193A (en) * | 2014-09-12 | 2014-12-24 | 北京玉峰生物技术有限公司 | Non-invasive blood sugar detection instrument |
| CN104665840A (en) * | 2015-03-02 | 2015-06-03 | 桂林麦迪胜电子科技有限公司 | Noninvasive blood glucose measuring method and finger tip measuring probe |
-
2015
- 2015-09-25 CN CN201510623388.5A patent/CN105193423A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2009128787A1 (en) * | 2008-04-16 | 2009-10-22 | Glucostats System Pte Ltd | Method and system for measuring a composition in a blood fluid |
| CN102217940A (en) * | 2011-06-14 | 2011-10-19 | 中国科学院长春光学精密机械与物理研究所 | Method for realizing no-wound human blood glucose measuring by utilizing volume pulse wave characteristic value |
| CN103690176A (en) * | 2013-12-26 | 2014-04-02 | 苏州大学 | Method for testing blood glucose in non-invasive manner |
| CN104224193A (en) * | 2014-09-12 | 2014-12-24 | 北京玉峰生物技术有限公司 | Non-invasive blood sugar detection instrument |
| CN104665840A (en) * | 2015-03-02 | 2015-06-03 | 桂林麦迪胜电子科技有限公司 | Noninvasive blood glucose measuring method and finger tip measuring probe |
Cited By (27)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106901752A (en) * | 2017-01-23 | 2017-06-30 | 武汉市瑞达源科技有限公司 | Method for determining concentration of glucose in blood of human body matrix |
| CN108513665A (en) * | 2017-02-07 | 2018-09-07 | 华为技术有限公司 | The methods, devices and systems of user identity identification |
| TWI640297B (en) * | 2017-03-23 | 2018-11-11 | 國立交通大學 | Non-invasive blood glucose measuring device, method, and system with identification function |
| WO2018214298A1 (en) * | 2017-05-20 | 2018-11-29 | 深圳市前海安测信息技术有限公司 | Non-invasive blood glucose detection system and method |
| WO2018214297A1 (en) * | 2017-05-20 | 2018-11-29 | 深圳市前海安测信息技术有限公司 | Portable device and method for non-invasive blood glucose detection |
| WO2018214299A1 (en) * | 2017-05-20 | 2018-11-29 | 深圳市前海安测信息技术有限公司 | Non-invasive blood glucose data collection apparatus |
| CN109276258A (en) * | 2018-08-10 | 2019-01-29 | 北京大学深圳研究生院 | Blood glucose trend forecasting method, system and Medical Devices based on DTW |
| CN109276258B (en) * | 2018-08-10 | 2021-08-03 | 北京大学深圳研究生院 | DTW-based blood glucose trend prediction method and system and medical equipment |
| CN111317484A (en) * | 2018-12-13 | 2020-06-23 | 三星电子株式会社 | Apparatus and method for estimating blood glucose |
| WO2020228084A1 (en) * | 2019-05-13 | 2020-11-19 | 深圳六合六医疗器械有限公司 | Personalized blood sugar interval statistical method and device |
| CN110141249A (en) * | 2019-06-18 | 2019-08-20 | 张远 | Woundless blood sugar monitoring method, system, equipment and medium based on PPG signal |
| CN110141249B (en) * | 2019-06-18 | 2022-02-18 | 张远 | Non-invasive blood glucose monitoring method, system, equipment and medium based on PPG signal |
| CN110251115A (en) * | 2019-06-20 | 2019-09-20 | 张远 | PPG method for extracting signal, system, equipment and medium based on body surface video |
| CN110251115B (en) * | 2019-06-20 | 2022-02-18 | 张远 | PPG signal extraction method, system, equipment and medium based on body surface video |
| US11759129B2 (en) | 2019-07-31 | 2023-09-19 | Tsinghua University | Noninvasive glucometer and blood glucose detection method |
| CN110327058A (en) * | 2019-07-31 | 2019-10-15 | 清华大学 | A kind of non-invasive blood sugar instrument and blood sugar detecting method |
| CN111297374B (en) * | 2020-02-24 | 2022-05-27 | 京东方科技集团股份有限公司 | Physical sign parameter detection equipment and physical sign parameter detection method |
| CN111297374A (en) * | 2020-02-24 | 2020-06-19 | 京东方科技集团股份有限公司 | Physical sign parameter detection equipment and physical sign parameter detection method |
| CN111599470A (en) * | 2020-04-23 | 2020-08-28 | 中国科学院上海技术物理研究所 | Method for improving near-infrared noninvasive blood glucose detection precision |
| CN111631733A (en) * | 2020-06-19 | 2020-09-08 | 浙江澍源智能技术有限公司 | Arterial blood spectrum detection method and device |
| CN111631733B (en) * | 2020-06-19 | 2024-01-26 | 浙江澍源智能技术有限公司 | Arterial blood spectrum detection method and device |
| WO2022063047A1 (en) * | 2020-09-22 | 2022-03-31 | 博邦芳舟医疗科技（北京）有限公司 | Photoplethysmography-based non-invasive diabetes prediction system and method |
| CN112120711A (en) * | 2020-09-22 | 2020-12-25 | 博邦芳舟医疗科技（北京）有限公司 | Noninvasive diabetes prediction system and method based on photoplethysmography |
| CN112120711B (en) * | 2020-09-22 | 2023-10-13 | 博邦芳舟医疗科技（北京）有限公司 | Noninvasive diabetes prediction system and method based on photoplethysmography pulse waves |
| JP2021112603A (en) * | 2020-12-22 | 2021-08-05 | 京セラ株式会社 | Electronic apparatus, estimation system, estimation method, and estimation program |
| JP2021045639A (en) * | 2020-12-22 | 2021-03-25 | 京セラ株式会社 | Electronic apparatus, estimation system, estimation method, and estimation program |
| CN113343200A (en) * | 2021-07-12 | 2021-09-03 | 武汉华星光电技术有限公司 | Electronic equipment unlocking system and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105193423A (en) | Non-invasive blood glucose detection method, device and system | |
| US20190295729A1 (en) | Universal non-invasive blood glucose estimation method based on time series analysis | |
| CN100515335C (en) | Blood oxygen measuring method and device capable of eliminating moving inteference | |
| WO2018214298A1 (en) | Non-invasive blood glucose detection system and method | |
| CN107296616A (en) | Portable non-invasive blood sugar test device and method | |
| CN101683260B (en) | Rhythmicity physiological signal optimization method and device | |
| CN105266772A (en) | Measuring method for physiological parameter | |
| CN106901705A (en) | A kind of unaware human Body Physiology Multi-parameter harvester and acquisition method and application | |
| Forouzanfar et al. | Automatic analysis of pre‐ejection period during sleep using impedance cardiogram | |
| CN101919704B (en) | Heart sound signal positioning and segmenting method | |
| CN109662720A (en) | Multiband optoacoustic Woundless blood sugar concentration prediction system based on deep learning | |
| CN110141249B (en) | Non-invasive blood glucose monitoring method, system, equipment and medium based on PPG signal | |
| CN108324286B (en) | Infrared noninvasive blood glucose detection device based on PCA-NARX correction algorithm | |
| CN204926092U (en) | Identification terminal and system | |
| CN106659404A (en) | Continuous blood pressure measurement method, apparatus and device | |
| Hossain et al. | Estimation of blood glucose from PPG signal using convolutional neural network | |
| Zanelli et al. | Transfer learning of CNN-based signal quality assessment from clinical to non-clinical PPG signals | |
| CN115137319A (en) | Skin microvascular endothelial function detection equipment and detection method thereof | |
| CN115624328A (en) | Infrared emitter of noninvasive glucose meter and glucose meter | |
| CN112022167A (en) | Noninvasive blood glucose detection method based on spectral sensor | |
| CN113069091A (en) | Pulse condition classification device and method for PPG (photoplethysmography) signals | |
| CN110575182A (en) | Method and device for detecting blood sugar | |
| LU502768B1 (en) | Intelligent blood pressure prediction method based on multi-scale residual network and ppg signal | |
| CN115770028A (en) | Blood pressure detection method, system, device and storage medium | |
| CN114903445A (en) | Intelligent monitoring and early warning system for cardiovascular and cerebrovascular diseases |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20151230 |