CN101991410B - Pulse rate searching and calculating method - Google Patents
Pulse rate searching and calculating method Download PDFInfo
- Publication number
- CN101991410B CN101991410B CN 200910169674 CN200910169674A CN101991410B CN 101991410 B CN101991410 B CN 101991410B CN 200910169674 CN200910169674 CN 200910169674 CN 200910169674 A CN200910169674 A CN 200910169674A CN 101991410 B CN101991410 B CN 101991410B
- Authority
- CN
- China
- Prior art keywords
- pulse frequency
- pulse
- search
- signal
- trough
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Landscapes
- Measuring Pulse, Heart Rate, Blood Pressure Or Blood Flow (AREA)
Abstract
The invention discloses a pulse rate searching and calculating method which comprises the following steps of: acquiring a pulse wave signal; preprocessing the acquired pulse wave signal to obtain an alternating current component in the pulse signal; carrying out pulse frequency searching on one section of obtained alternating current signal to obtain a pulse frequency of the pulse wave signal; carrying out pulse frequency tracking on a calculating result of the obtained pulse frequency, updating and correcting the pulse frequency in real time; carrying out interference reorganization on a pulse signal by adopting an interference reorganization algorithm while updating and correcting the pulse frequency in real time, and correspondingly processing a signal with larger interference for keeping the pulse frequency stable. Through analyzing and processing the pulse signal, accurate pulse frequency values can be searched and calculated, especially, the accuracy and the stability of the pulse frequency calculation can be improved when the pulse signals are interfered. Compared with the traditional pulse frequency searching and calculating method, the invention has the advantages of accuracy, stability, real time updating, interference resistance, simple and easy operation, and the like, and has better practical application value.
Description
Technical field
The present invention relates to a kind of processing method to pulse frequency in the pulse signal, pulse frequency search in particular and Calculation Method are particularly suitable for pulse frequency search and calculating in the pulse frequency instrument.
Background technology
Pulse signal is an important physical signal in the biomedical measurement, wherein includes abundant physiologic information, and the size of pulse frequency can reflect the patient's body situation, and this parameter often is used to the diagnosis to the patient.Existing pulse blood oxygen instrument generally adopts multiband light electric induction technology, through being carried out date processing and analysis by the pulse information that optical signal contained of finger or the absorption of other position, calculates physiological parameters such as pulse frequency and blood oxygen saturation.
Generally in 0.5-5Hz, before carrying out the calculating of pulse frequency and blood oxygen saturation, general elder generation carries out pretreatment to data, some high-frequency noises of filtering and some low-frequency disturbance to the frequency range of human pulse; But the interference that causes for motion such as hands, generally all in the pulse effective band, therefore just can't be through the filtering removal, this calculating to pulse frequency has increased difficulty.
So proposed different pulse search and computational methods.Common method generally has three kinds: first method is to adopt the search extreme point to find out the maximum and the minimum of pulse signal, through the screening of extreme point being found out the crest and the trough of pulse wave, and then calculates the pulse frequency value.The threshold value screening method often is used to choosing of extreme point, relatively can filter out possible crest and trough through signal and preset threshold to sampling, searches out crest, the trough of pulse signal again through some corresponding processing methods, calculates pulse frequency thus.This method is simple and clear, is easy to realize, can obtain result more accurately for signal more stably, if but signal generation baseline drift tends to produce the bigger error of calculation.
Second method is a calculus of finite differences, and it confirms the position of Wave crest and wave trough through the variation of signal difference score value, but this method searches wrong pulse frequency value easily to the signal that is interfered.
The third method is a Spectrum Method, and it can to a certain degree improve the problems referred to above, but the energy leakage is the inherent shortcoming that frequency spectrum calculates; When interference such as low frequency wonder or sign mutation occurred, the frequency spectrum algorithm also was powerless, and the frequency spectrum algorithm needs a certain amount of data; Can cause delay; Amount of calculation is big, has increased hsrdware requirements, often is restricted in its use of some occasion.
To sum up, at present also there are deficiency in search and computational methods in the pulse frequency processing.
Summary of the invention
Carry out in some conventional methods of pulse frequency searching and computing to pulse signal at present; In the time of can not solving pulse signal well and be interfered particularly motion artifacts; Problems such as the accuracy of pulse rate calculations and maintenance numerical stability; The present invention can effectively address the above problem, and has improved the interference free performance of pulse rate calculations, has improved accuracy, real-time and the stability of pulse rate calculations.Interference recognizer of the present invention can be discerned the influence whether pulse signal receives larger interference exactly, thereby has improved the problem of pulse rate calculations under significantly disturbing.
In order to realize above-mentioned thought, concrete technical scheme of the present invention is following:
A kind of search of pulse frequency and computational methods, it may further comprise the steps:
A, collection pulse wave signal;
B, the pulse wave signal that collects is carried out pretreatment, obtain the alternating component in the pulse wave signal;
C, the one section alternating component that is obtained is carried out pulse frequency search, obtain the pulse frequency of pulse wave signal;
D, carry out pulse frequency according to the result of calculation of the pulse frequency that obtains and follow the tracks of, pulse frequency is carried out real-time update and correction;
E, pulse frequency being carried out real-time update and revising also to adopt simultaneously and disturb recognizer that pulse wave signal is disturbed identification, carry out handled to disturbing bigger signal, keep the stable of pulse frequency.
When the acquisition alternating component is carried out the pulse frequency search; Be that extreme point is repeatedly searched for; Extreme point is classified as the intrinsic extreme point class corresponding to different frequency, the extreme point class is analyzed to judge whether such extreme point class is the crest and the trough of pulse wave, and calculate pulse frequency according to this.
If find crest or trough then search for end; Maybe count and be less than setting value and then search for end when remaining extreme value.
It is uniformly that crest or trough are judged as the extreme point spacing, and such as the distance values that has some extreme point, their mean square deviation satisfies predetermined condition; And successive during the extreme point spacing, satisfy predetermined condition such as the mean square deviation of the distance values of adjacent extreme point.
If search stops and still do not search crest or trough, then gather one piece of data again, and repeat the pulse frequency search step.
Carry out pulse frequency according to the result of calculation of the pulse frequency that obtains and follow the tracks of, comprise,
1, the one section alternating component that is obtained is carried out the pulse frequency search, obtain the pulse frequency of pulse wave signal, the position of predicting next crest and trough;
2, between current crest and next crest, search out minima, and, between current trough and next trough, search out maximum with the trough value of this minima as reality, and with the crest value of this maximum as reality;
3, according to the actual crest that is searched out and trough value spacing and pulse frequency value are revised and upgraded, or negate.
Pulse frequency is revised and upgraded; Or negate; Mainly be to compare with the historical distance values that had before calculated according to the crest of current calculating or trough distance values; The current distance values that calculates is closed within the scope in certain of history value, then thinks stabilization signal, and the pulse frequency value is carried out suitable correction and renewal.
Compare with the historical distance values that had before calculated according to the crest of current calculating or trough distance values,, think that then variation has taken place signal really, the pulse frequency value is revised and upgraded if continual and steady variation takes place.
Crest or trough distance values according to current calculating compare with the historical distance values that had before calculated, and have no rule if change, and differ greatly, and think that then signal is unusual really, and the one section pulse wave signal of then sampling carries out the pulse frequency search again.
Adopt when pulse frequency is carried out real-time update and revised and disturb recognizer; Disturb recognizer can discern the bigger pulse wave signal of noise amplitudes, and, keep the stability of pulse frequency value with this search and calculating of revising pulse frequency.
The present invention can search for and calculate pulse frequency value accurately through to pulse wave signal analysis and processing, and this method especially can improve the accuracy and the stability of pulse rate calculations when pulse wave signal is interfered.Than search of traditional pulse frequency and computational methods, the pulse frequency search that the present invention proposes has accurately with computational methods, stablizes, real-time update, advantage such as anti-interference, simple, has practice value preferably.Method of the present invention, simple, the algorithm computation amount is little; Hsrdware requirements are low; Calculate accurately, reliable results especially still can guarantee the accuracy of numerical value and keep data stability under the situation that signal is interfered; This method is the variation of real-time tracking pulse wave signal also, and the real-time of maintenance data is also followed the variation of pulse frequency fast.
Description of drawings:
Fig. 1 is a system flow chart of the present invention;
Pulse frequency searching algorithm flow chart among Fig. 2 the present invention;
Fig. 3 pulse wave signal search procedure of the present invention instance 1;
Fig. 4 pulse wave signal search procedure of the present invention instance 2;
Fig. 5 pulse wave signal search procedure of the present invention instance 3;
Fig. 6 pulse frequency trace flow of the present invention figure;
Fig. 7 the present invention disturbs the identification sketch map in pulse blood oxygen instrument.
The specific embodiment
Pulse frequency processing method disclosed by the invention can be used for the pulse frequency search of pulse wave signal and calculate; Also can be used for search and calculating that any periodic physiological signal carries out the cycle; Like electrocardiosignal, breath signal etc., below in conjunction with the concrete implementation of accompanying drawing this method of explaination in pulse blood oxygen instrument.
In the pulse blood oxygen instrument commonly used, send HONGGUANG and infrared signal through the control luminous tube, during light transmission finger by various tissue absorption part light; Along with beating of pulse; Corresponding alternate also appears in absorbed optical signal, and the optical signal that receiving tube receives passes through Signal Pretreatment then through links such as change-over circuit, amplifying circuit and samplings; Can obtain pulse wave signal, call the pulse frequency algorithm computation then and go out the pulse frequency value.It is HONGGUANG and infrared transmission or reflection finger or other position of health that pulse blood oxygen instrument generally adopts two waveband; Obtain two passage pulse wave signals simultaneously; Different by the absorbance of HbO2 Oxyhemoglobin and reduced hemoglobin according to HONGGUANG with infrared light, utilize lambert-Bill (Lamber-Beer) law to calculate oximetry value.According to Lambert-Beer's law, oximetry value is by the ratio decision that exchanges amplitude and the ratio of direct current amplitude of HONGGUANG and infrared light.And the oximetry value of human body can keep relative stability in a period of time, so in a period of time, their ratio also can keep relative stability.
The present invention mainly comprises two links, and pulse frequency search and pulse frequency are followed the tracks of, and the pulse frequency search is general through the crest of finding out pulse wave signal, the position of trough point, and calculates pulse frequency thus, and the information of all the other points is not very paid close attention to.Rule of thumb mode is decomposed the thought of " signal is that the high frequency periodic signal superposition is on low-frequency periodic signal ", and the extreme point of signal also demonstrates the characteristic of the mutual superposition of low-and high-frequency.The high-frequency signal superposition on low frequency signal, the extreme point of high-frequency signal also superposition on low frequency signal, through extreme point being searched for once more the extreme point that extreme point then can find the outer residue signal of high-frequency signal.So repeated searching then can obtain " the intrinsic extreme point class " of a series of corresponding different frequency signals successively, then discern from these extreme point apoplexy due to endogenous wind whether extreme points are crest, the trough of pulse wave signal, thus calculate pulse frequency.Because this method no longer needs " the eigen mode state function " of signal calculated and need not carry out the envelope match to extreme point; Only be that extreme point is handled and analyzed; Simplified computational process greatly and reduced the amount of deal with data, used so be highly suitable in the small-sized embedded system.The pulse frequency track algorithm can guarantee the instantaneity of pulse rate calculations, follows the variation of pulse frequency in real time.After the pulse frequency search finishes, utilize result of calculation that crest subsequently, the position of trough point are predicted, check signal whether to change according to the actual pulse ripple signal of gathering subsequently, and according to this pulse frequency value is revised in real time.
The flow process of whole system is as shown in Figure 1.Earlier one section pulse wave signal is carried out the pulse frequency search, adopt the pulse frequency tracking that pulse frequency is carried out real-time update and correction after searching pulse frequency, follow the tracks of in pulse frequency and also added the interference recognizer simultaneously, judge whether to disturb, and make corresponding processing.
The detailed process of pulse frequency search is following: gather one piece of data earlier, require data length to comprise several (such as 4) complete pulse cycle at least, to guarantee to search crest and trough.The data that collect are repeatedly searched for the method for extreme point and seek crest and trough; From maximum point, search out crest, from minimum point, search out trough, this algorithm flow is as shown in Figure 2; With the search trough is example (step of search crest is roughly the same), comprises the following step:
1, all minimum points is found out, calculated their spacing respectively.
2, according to the mutual relation of spacing size, confirm the trough point.If do not find satisfactory trough point, then new search data more is about to all minimum points and is used as new data, more all minimum points is found out, and calculates their spacing respectively.
3, when remaining data are very few can't proceed to search for the time, interrupt search, gather one section new data again and search for again.
Fig. 3,4,5 is the trough search procedure of 3 groups of different pulse wave signals, and Fig. 3 is the less pulse wave signal of interference that receives, and searching algorithm just can find trough point accurately through search first; And the pulse wave signal of Fig. 4 receives bigger interference; Traditional searching method searches out wrong pulse frequency value easily, under pulse frequency searching algorithm of the present invention, through extreme point search for the second time; Can find the trough point exactly; The suffered interference of the pulse wave signal of Fig. 5 is even more serious than Fig. 4's, but after through three extreme points search, has also accurately found the trough point.This shows that this algorithm has better anti-disturbance performance.
After searching pulse frequency, get into pulse frequency and follow the tracks of link.The flow process that pulse frequency is followed the tracks of is as shown in Figure 6; According to before the spacing that obtains and the pulse frequency position that dopes next crest and trough, between current trough and prediction trough, search for maximum, and be assumed to the actual wave crest point of current pulse cycle; Between current crest and prediction crest, search for minima; And be assumed to the actual trough point of current pulse cycle, and actual Wave crest and wave trough point that searches out and predictive value are compared, whether the check changes in spacing is normal.If changes in spacing within more among a small circle, is then thought normal signal, spacing and pulse frequency value are upgraded; Otherwise count unusual; Think that if anomalous counts reaches some bigger variation takes place pulse wave signal, do not exist, then search for pulse frequency again if identify to disturb.
Pulse frequency is followed the tracks of link and is included the interfering signal recognizer, can discern disturbing excessive signal, improves the accuracy and the stability of pulse rate calculations in such cases.HONGGUANG and infrared light dual pathways pulse wave signal are arranged in the oxygen saturation measurement appearance simultaneously; The optical signal of different-waveband is different by the light intensity ratio of finger; But both keep ratio when blood oxygen saturation is constant relatively stable, so can whether ANOMALOUS VARIATIONS is discerned interferential existence according to their ratio.
The concretism of interfering signal recognizer is following:
For the pulse alternating component, establishing detected HONGGUANG alternating component is I
Rd, the infrared light alternating component is I
Ir, they are respectively by pulse wave signal part (S
Rd, S
Ir) and interfering signal part (N
Rd, N
Ir) form.Pulse wave signal part S
RdWith S
IrProportional, its ratio value of keeing relative stability is r under the constant situation of measurement environment and blood oxygen saturation.Pulse wave signal S in the red signal light
Rd, deduct the pulse wave signal S that infrared light produces
IrWith the product of ratio r, its result is residual noise signal N only, and this noise signal can be used as judges whether signal contains the foundation that disturbs.
I
rd＝S
rd+N
rd
I
ir＝S
ir+N
ir
Wherein:
S
rd＝r*S
ir
So:
N＝I
rd-r*I
ir
＝S
rd+N
rd-r*(S
ir+N
ir)
＝S
rd-r*S
ir+N
rd-r*N
ir
＝N
rd-r*N
ir
When the N value is big, the existence that disturbs is described, otherwise, disturb less.
Can judge through above-mentioned theory whether pulse wave signal receives larger interference, thereby improve the stability of the pulse rate calculations under larger interference, prevent to calculate wrong pulse frequency value.Fig. 7 is red signal light, infrared signal and the HONGGUANG of a pulse wave signal and the error amount that infrared signal subtracts each other in proportion.Glitch-free the time, error signal is very little, and under middle two sections noisy situation; The amplitude of error signal is relatively large; Variation through this error signal just can be discerned interfering signal and state no interference signal, and noisy the time, it is more careful that pulse frequency is upgraded; Such as keeping numerical value to be locked on the history value, can effectively prevent the measurement mistake that causes because of interference.
Claims (1)
1. the search of a pulse frequency and computational methods, it may further comprise the steps:
A, collection pulse wave signal;
B, the pulse wave signal that collects is carried out pretreatment, obtain the alternating component in the pulse wave signal;
C, the one section alternating component that is obtained is carried out pulse frequency search; Said the one section alternating component that is obtained is carried out pulse frequency when search; Be that extreme point is repeatedly searched for; Extreme point is classified as the intrinsic extreme point class corresponding to different frequency, the extreme point class is analyzed to judge whether this intrinsic extreme point class is the crest and the trough of pulse wave, thereby and calculate to obtain the pulse frequency of pulse wave signal according to this;
D, carry out pulse frequency according to the result of calculation of the pulse frequency that obtains and follow the tracks of, pulse frequency is carried out real-time update and correction;
Said result of calculation according to the pulse frequency that obtains is carried out the pulse frequency tracking, comprising:
Step 1, by the pulse frequency of pulse wave signal, predict the position of next crest and trough;
Step 2, between current crest and next crest, search out minima, and with the trough value of this minima as reality; Between current trough and next trough, search out maximum, and with the crest value of this maximum as reality;
Step 3, spacing and pulse frequency value are revised and upgraded, or negate according to the actual crest that is searched out and trough value;
E, when pulse frequency being carried out real-time update and revising, adopt and disturb recognizer that pulse wave signal is disturbed identification, carry out handled, keep the stable of pulse frequency disturbing bigger signal.
2, according to the search and the computational methods of the said a kind of pulse frequency of claim 1, it is characterized in that, in step c, if find crest or trough then search for end; Maybe count and then search for end when being less than setting value when remaining extreme value.
3, the search of a kind of pulse frequency according to claim 2 and computational methods is characterized in that, it is even and successive that crest or trough are judged as the extreme point spacing; Said extreme point spacing is uniformly, is meant the distance values that exists some extreme point, and their mean square deviation satisfies predetermined condition; Said extreme point spacing is successive, is meant that the mean square deviation of the distance values of adjacent extreme point satisfies predetermined condition.
4, the search of a kind of pulse frequency according to claim 2 and computational methods is characterized in that, in step c, if search stops and still do not search crest or trough, then gather one piece of data again, and repeat the pulse frequency search step.
5, the search of a kind of pulse frequency according to claim 1 and computational methods; It is characterized in that, pulse frequency is revised and upgraded, or negate; Mainly be to compare with the historical distance values that had before calculated according to the crest of current calculating or trough distance values; The current distance values that calculates is then thought stabilization signal within the scope of history value, the pulse frequency value is carried out suitable correction and renewal.
6, the search of a kind of pulse frequency according to claim 5 and computational methods; It is characterized in that; Crest or trough distance values according to current calculating compare with the historical distance values that had before calculated; If continual and steady variation takes place, think that then variation has taken place signal really, carries out suitable correction and renewal to the pulse frequency value.
7, according to the search and the computational methods of the said a kind of pulse frequency of claim 5; It is characterized in that; Crest or trough distance values according to current calculating compare with the historical distance values that had before calculated, and have no rule if change, and differ greatly; Think that then abnormal signal, the one section pulse wave signal of then sampling carry out the pulse frequency search again.
8, the search of a kind of pulse frequency according to claim 1 and computational methods is characterized in that, pulse frequency is carried out real-time update disturb recognizer with revising to adopt; Disturb recognizer can discern the bigger pulse wave signal of noise amplitudes, and, keep the stability of pulse frequency value with this search and calculating of revising pulse frequency.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200910169674 CN101991410B (en) | 2009-08-31 | 2009-08-31 | Pulse rate searching and calculating method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200910169674 CN101991410B (en) | 2009-08-31 | 2009-08-31 | Pulse rate searching and calculating method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101991410A CN101991410A (en) | 2011-03-30 |
| CN101991410B true CN101991410B (en) | 2012-09-19 |
Family
ID=43782526
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 200910169674 Active CN101991410B (en) | 2009-08-31 | 2009-08-31 | Pulse rate searching and calculating method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101991410B (en) |
Families Citing this family (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102885616A (en) * | 2012-07-17 | 2013-01-23 | 桂林电子科技大学 | Method for removing baseline drift from pulse wave signal |
| JP5987578B2 (en) * | 2012-09-13 | 2016-09-07 | オムロンヘルスケア株式会社 | Pulse measuring device, pulse measuring method and pulse measuring program |
| CN103230267B (en) * | 2013-05-14 | 2015-06-03 | 北京理工大学 | Anti-movement-interference extraction method for pulse rates |
| CN103549945B (en) * | 2013-10-31 | 2015-07-15 | 广州视源电子科技股份有限公司 | Method for identifying pulse rate and blood oxygen saturation degree through heart contraction process characteristics |
| CN106604679B (en) * | 2014-09-09 | 2020-11-06 | 日本电信电话株式会社 | Heartbeat detection method and heartbeat detection device |
| CN105534475A (en) * | 2015-06-30 | 2016-05-04 | 宇龙计算机通信科技(深圳)有限公司 | Physiological parameter data updating method and system |
| CN105832346A (en) * | 2015-12-02 | 2016-08-10 | 梁云 | Blood oxygen supply detecting instrument |
| CN107577986B (en) * | 2017-07-31 | 2021-07-06 | 来邦科技股份公司 | Respiration and heartbeat component extraction method, electronic equipment and storage medium |
| CN111528830B (en) * | 2020-05-20 | 2023-03-17 | 广东工业大学 | Electrocardiogram monitoring device |
| CN112587133A (en) * | 2020-12-09 | 2021-04-02 | 深圳市奋达智能技术有限公司 | Method for measuring blood oxygen saturation |
| CN113100727B (en) * | 2021-05-12 | 2023-09-19 | 深圳市通久电子有限公司 | Method for analyzing and identifying pulse wave crest in real time |
| CN113545753B (en) * | 2021-07-21 | 2024-04-05 | 芯原微电子（成都）有限公司 | Waveform detection method and device, wearable equipment and computer readable storage medium |
| CN116439675A (en) * | 2023-03-28 | 2023-07-18 | 深圳市莱康宁医用科技股份有限公司 | Measurement method, device and system applied to human circulatory system |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1830386A (en) * | 2005-03-10 | 2006-09-13 | 深圳迈瑞生物医疗电子股份有限公司 | Method of measuring blood oxygen under low filling |
| CN1985750A (en) * | 2005-12-21 | 2007-06-27 | 深圳迈瑞生物医疗电子股份有限公司 | Pulse wave detecting method and device by means of cardiac symbol signal |
| CN101327121A (en) * | 2007-06-22 | 2008-12-24 | 香港中文大学 | Physiological parameter measurement mechanism |
| CN101433463A (en) * | 2008-12-19 | 2009-05-20 | 北京航空航天大学 | Method for detecting blood oxygen saturation |
-
2009
- 2009-08-31 CN CN 200910169674 patent/CN101991410B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1830386A (en) * | 2005-03-10 | 2006-09-13 | 深圳迈瑞生物医疗电子股份有限公司 | Method of measuring blood oxygen under low filling |
| CN1985750A (en) * | 2005-12-21 | 2007-06-27 | 深圳迈瑞生物医疗电子股份有限公司 | Pulse wave detecting method and device by means of cardiac symbol signal |
| CN101327121A (en) * | 2007-06-22 | 2008-12-24 | 香港中文大学 | Physiological parameter measurement mechanism |
| CN101433463A (en) * | 2008-12-19 | 2009-05-20 | 北京航空航天大学 | Method for detecting blood oxygen saturation |
Non-Patent Citations (1)
| Title |
|---|
| JP特开2008-79813A 2008.04.10 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101991410A (en) | 2011-03-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101991410B (en) | Pulse rate searching and calculating method | |
| EP1611847A1 (en) | Validating pulse oximetry signals in the potential presence of artifact | |
| KR100499139B1 (en) | Method of removing abnormal data and blood constituent analysing system using spectroscopy employing the same | |
| US6122535A (en) | Method and apparatus for determining the concentration of a component | |
| CN103027667B (en) | Characteristic parameter extraction of pulse wave | |
| KR20160094265A (en) | Computing device and method for analyzing biophysiological rate features thereof | |
| CN103565427A (en) | Method for detecting quasi-periodicity physiological signal feature points | |
| JP2003505754A (en) | Method of identifying valid signal in measurement signal | |
| CN100589759C (en) | Saturation oxygen detection method | |
| CN108042107A (en) | A kind of PPG signals puppet difference correcting method | |
| CN106798552A (en) | Heart rate with status switch optimization estimates equipment | |
| WO2017214870A1 (en) | Method for calculating physiological parameters and corresponding medical equipment | |
| CN105943012B (en) | Heart rate measuring method for removing motion noise in photoplethysmography signals | |
| CN108937957A (en) | Detection method, device and detection device | |
| CN103034837A (en) | Relevancy of characteristic parameters and pulse condition elements | |
| CN101683260A (en) | Rhythmicity physiological signal optimization method and device | |
| US20220304610A1 (en) | Method and apparatus for processing an electrocardiogram signal and electronic device | |
| CN115702782A (en) | Heart rate detection method based on deep learning and wearable device | |
| CN103230267A (en) | Anti-movement-interference extraction method for pulse rates | |
| CN105816165B (en) | real-time dynamic heart rate monitoring device and monitoring method | |
| CN114532993A (en) | Automatic detection method for electroencephalogram high-frequency oscillation signals of epileptic | |
| Kumar et al. | STSR: spectro-temporal super-resolution analysis of a reference signal less photoplethysmogram for heart rate estimation during physical activity | |
| CN104095640A (en) | Oxyhemoglobin saturation detecting method and device | |
| CN101940476A (en) | Blood oxygen saturation detection method and system | |
| CN106214143B (en) | Transient motion interference identification method and heart rate confidence calculations method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |