WO2017214870A1 - 一种生理参数的计算方法及相应的医疗设备 - Google Patents

一种生理参数的计算方法及相应的医疗设备 Download PDF

Info

Publication number
WO2017214870A1
WO2017214870A1 PCT/CN2016/085777 CN2016085777W WO2017214870A1 WO 2017214870 A1 WO2017214870 A1 WO 2017214870A1 CN 2016085777 W CN2016085777 W CN 2016085777W WO 2017214870 A1 WO2017214870 A1 WO 2017214870A1
Authority
WO
WIPO (PCT)
Prior art keywords
spectral
domain signal
coefficient
spectral peak
frequency domain
Prior art date
Application number
PCT/CN2016/085777
Other languages
English (en)
French (fr)
Inventor
韩飞
叶飞
姚祖明
张飞
何先梁
刘立汉
李明
金星亮
Original Assignee
深圳迈瑞生物医疗电子股份有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 深圳迈瑞生物医疗电子股份有限公司 filed Critical 深圳迈瑞生物医疗电子股份有限公司
Priority to CN201680073670.3A priority Critical patent/CN108471961B/zh
Priority to PCT/CN2016/085777 priority patent/WO2017214870A1/zh
Priority to CN202110094270.3A priority patent/CN112932473B/zh
Priority to EP16904987.1A priority patent/EP3473171B1/en
Publication of WO2017214870A1 publication Critical patent/WO2017214870A1/zh
Priority to US16/203,068 priority patent/US11154250B2/en
Priority to US17/505,155 priority patent/US11872060B2/en

Links

Images

Classifications

    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/145—Measuring characteristics of blood in vivo, e.g. gas concentration, pH value; Measuring characteristics of body fluids or tissues, e.g. interstitial fluid, cerebral tissue
    • A61B5/1455—Measuring characteristics of blood in vivo, e.g. gas concentration, pH value; Measuring characteristics of body fluids or tissues, e.g. interstitial fluid, cerebral tissue using optical sensors, e.g. spectral photometrical oximeters
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
    • A61B5/7235—Details of waveform analysis
    • A61B5/7253—Details of waveform analysis characterised by using transforms
    • A61B5/7257—Details of waveform analysis characterised by using transforms using Fourier transforms
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
    • A61B5/024—Detecting, measuring or recording pulse rate or heart rate
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/145—Measuring characteristics of blood in vivo, e.g. gas concentration, pH value; Measuring characteristics of body fluids or tissues, e.g. interstitial fluid, cerebral tissue
    • A61B5/1455—Measuring characteristics of blood in vivo, e.g. gas concentration, pH value; Measuring characteristics of body fluids or tissues, e.g. interstitial fluid, cerebral tissue using optical sensors, e.g. spectral photometrical oximeters
    • A61B5/14551—Measuring characteristics of blood in vivo, e.g. gas concentration, pH value; Measuring characteristics of body fluids or tissues, e.g. interstitial fluid, cerebral tissue using optical sensors, e.g. spectral photometrical oximeters for measuring blood gases
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
    • A61B5/7225—Details of analog processing, e.g. isolation amplifier, gain or sensitivity adjustment, filtering, baseline or drift compensation
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
    • A61B5/7235—Details of waveform analysis

Definitions

  • the invention relates to the field of physiological parameter monitoring, in particular to a method for calculating physiological parameters and corresponding medical equipment.
  • the metabolic process of the human body is a biological oxidation process, and the oxygen required for the metabolic process enters the human blood through the respiratory system, and combines with the reduced hemoglobin (Hb) in the red blood cells of the blood to form oxyhemoglobin (HbO2), which is then transported to the human body. Part of the tissue cells.
  • Hb reduced hemoglobin
  • HbO2 oxyhemoglobin
  • Blood oxygen saturation is the percentage of hemolytic hemoglobin (HbO2) capacity in the blood to the total hemoglobin (Hb + HbO2) capacity, ie the concentration of blood oxygen in the blood.
  • HbO2 hemolytic hemoglobin
  • Hb + HbO2 total hemoglobin
  • SpO2 blood oxygen saturation
  • pulse oximetry is so extensive that the requirements for pulse oximetry accuracy are also rising. It is known to those skilled in the art that there are two key factors in evaluating the pros and cons of pulse oximetry: weak perfusion performance and exercise performance. If the patient under test has poor perfusion performance or poor exercise performance, the accuracy and stability requirements for pulse oximetry performance are even more demanding.
  • oxyhemoglobin HbO2
  • deoxyhemoglobin HB
  • Fig. 1 oxyhemoglobin in the prior art.
  • the hemoglobin absorption spectrum curve is reduced; using this absorption characteristic, the blood oxygen saturation parameter can be further evaluated.
  • FIG. 2 it is a working diagram of blood oxygen measurement in the prior art; two light paths are emitted by using an arc tube (example)
  • red light and near-infrared light transmit the body tissue and receive the light signal transmitted through the body tissue by the receiving tube, thereby obtaining the blood pulsation component AC alternating current (for example, arterial blood) and the non-pulsating component DC direct current (for example: vein Blood, muscle, bones, skin, etc.).
  • the ratio of the mapping curve to the arterial oxygen saturation (SaO2) that is, the R coefficient table (as shown in Equation 1) can be obtained.
  • the infinite approximation pulse oximetry (SpO2) of arterial oxygen saturation (SaO2) can be obtained.
  • AC Red the amount of red light detected, that is, the amount of AC
  • DC Red the maximum amount of red light detected, that is, the amount of direct current
  • AC Ired the amount of detected infrared light
  • DC Ired detected The amount of infrared light in the infrared.
  • time domain technology has the characteristics of fast response and clear phase information. Since the time domain signal is a mixture of the useful signal and the noise signal, the noise outside the physiological bandwidth can be easily filtered by the high pass/low pass filter, and when the noise within the physiological bandwidth occurs, due to the variability of the noise, prior knowledge Scarce, the time domain method has almost no way to filter out this part of the noise. Therefore, time domain technology has natural defects in anti-sports performance.
  • the frequency domain technology can theoretically separate the frequency bands of noise and useful signals, thereby achieving the purpose of distinguishing and identifying real signals.
  • any waveform is composed of multiple sinusoidal waves, and the pulse wave of physiological parameters is no exception. Therefore, when the physiological pulse wave signal is converted into a frequency domain signal, the physiological parameter exhibits a fundamental frequency doubling characteristic.
  • the fundamental spectrum of the interference spectrum and the physiological parameters are aliased together, and it is difficult to identify which one of the real spectrum is. Therefore, although the frequency domain technology has congenital advantages, it is very difficult to accurately calculate blood oxygen related parameters in the interference state.
  • Equation 1 also applies to the frequency domain signal, where the AC amount corresponds to the energy change at each frequency point.
  • the red and infrared spectrum of the pulse wave is shown in Figure 3.
  • the ratio of the main frequency band and each octave band can be regarded as the corresponding energy ratio in this frequency band.
  • the meaning of the energy ratio is completely consistent with the meaning of the corresponding R value in Equation 1, that is, it corresponds uniquely to blood oxygen saturation.
  • the ratio of each frequency band is the same, which corresponds to the current blood oxygen saturation.
  • the blood oxygen saturation parameter can be obtained in any frequency band.
  • the frequency at which the fundamental frequency peak is located in FIG. 3 is the pulse rate value.
  • the method of routinely detecting the fundamental frequency of the fundamental frequency can be obtained by screening the base frequency peak, and the theory is based on the fundamental frequency.
  • the peak and harmonic peaks have a proportional relationship between energy and frequency.
  • the physiological pulse rate value can be recognized by this proportional relationship.
  • the pulse rate parameter and the blood oxygen parameter can be obtained by retrieving the position information of the fundamental frequency peak in the frequency domain signal and the energy ratio of the red light to the infrared light.
  • the spectral peak information of red and infrared light may be confused and annihilated by noise, and the fundamental frequency peak frequency information may be abnormally calculated due to unrecognizable or misidentified; and the energy ratio of red light and infrared light is also
  • the blood oxygen calculation deviation is large due to the mixing of noise.
  • an indication of the spectrum distribution under interference is given.
  • the spectral peak energy ratios of infrared light and red light are suddenly large and small, and the fundamental frequency peak is almost annihilated. In this case, it is almost impossible to correctly identify the spectral peaks and calculate accurate blood oxygen and pulse rate parameters using existing time-frequency domain techniques.
  • the technical problem to be solved by the embodiments of the present invention is to provide a method for calculating physiological parameters and a corresponding medical device, which can improve the accuracy of calculating physiological parameters (such as pulse rate values and blood oxygen parameters) under weak irrigation and exercise states. And the computational complexity is low, and the demand for computing resources is low.
  • an embodiment of the present invention provides a method for calculating a physiological parameter, and the method includes the following steps:
  • the compensation coefficient is used to compensate at least one of the time domain signal and/or the frequency domain signal, and the physiological parameter is calculated based on the compensated time domain signal and/or the frequency domain signal.
  • a medical device for calculating a physiological parameter comprising:
  • a sensor comprising at least one light emitting tube and at least one receiving tube, the light emitting tube emitting at least two optical signals of different wavelengths transmitted through the physiological tissue, the receiving tube receiving at least two optical signals transmitted through the physiological tissue, and converting to electric signal;
  • An analog to digital converter coupled to the sensor to convert the electrical signal into a digital signal comprising at least a portion of a characteristic of a physiological tissue
  • a digital processor coupled to the analog to digital converter, the digital processor performing the following processing:
  • the compensation signal is used to compensate the digital signal of the segment and/or the corresponding frequency domain signal, and the physiological parameter is calculated based on the compensated digital signal and/or the frequency domain signal.
  • a method for calculating a physiological parameter comprising the steps of:
  • the compensation coefficient is used to compensate at least one of the time domain signal and/or the frequency domain signal, and the physiological parameter is calculated based on the compensated time domain signal and/or the frequency domain signal.
  • a medical device for calculating a physiological parameter comprising:
  • a sensor comprising at least one light emitting tube and at least one receiving tube, the light emitting tube emitting at least two optical signals of different wavelengths transmitted through the physiological tissue, the receiving tube receiving at least two optical signals transmitted through the physiological tissue, and converting to electric signal;
  • An analog to digital converter coupled to the sensor to convert the electrical signal into a digital signal comprising at least a portion of a characteristic of a physiological tissue
  • a digital processor coupled to the analog to digital converter, the digital processor performing the following processing:
  • the domain signal is calculated based on the compensated digital signal and/or the frequency domain signal to obtain physiological parameters.
  • a method for calculating a physiological parameter comprising the steps of:
  • the compensation coefficient is used to compensate at least one of the time domain signal and/or the frequency domain signal, and the physiological parameter is calculated based on the compensated time domain signal and/or the frequency domain signal.
  • a medical device for calculating a physiological parameter comprising:
  • a sensor comprising at least one light emitting tube and at least one receiving tube, the light emitting tube emitting at least two optical signals of different wavelengths transmitted through the physiological tissue, the receiving tube receiving at least two optical signals transmitted through the physiological tissue, and converting to electric signal;
  • An analog to digital converter coupled to the sensor to convert the electrical signal into a digital signal comprising at least a portion of a characteristic of a physiological tissue
  • a digital processor coupled to the analog to digital converter, the digital processor performing the following processing:
  • the compensation signal is used to compensate the digital signal of the segment and/or the corresponding frequency domain signal, and the physiological parameter is calculated based on the compensated digital signal and/or the frequency domain signal.
  • the embodiment of the present invention is based on the frequency domain technology combined with the time domain technology, can greatly improve the calculation accuracy of the physiological parameters under the condition of weak perfusion and motion, and brings an excellent clinical performance experience to the customer, which can greatly Improve the application and promotion of physiological parameters (such as blood oxygen parameters);
  • the present invention is implemented by combining the characteristics of the time-frequency domain signal with the characteristics of the human physiological parameters, and in the case of interference, using the venous oxygen compensation method and the spectrum array pattern method, in the weak irrigation and transportation Improve the accuracy of calculating physiological parameters (such as pulse rate values and blood oxygen parameters) under dynamic conditions.
  • the venous oxygen compensation method can eliminate the deviation of blood oxygen measurement caused by interference, infinitely approach the true physiological blood oxygen value, and greatly provide the accuracy of the calculation of blood oxygen parameters under the interference state
  • the spectrum array map method can eliminate the interference caused by the interference Pulse rate measurement deviation, even in the long-term interference state, can accurately identify the physiological spectrum information, greatly providing the accuracy of the calculation of the pulse rate parameter in the interference state;
  • the method provided by the embodiment of the present invention has low computational complexity and low demand for computing resources.
  • 1 is a graph showing absorption spectra of oxyhemoglobin and reduced hemoglobin in the prior art
  • FIG. 2 is a schematic diagram of working blood oxygen measurement in the prior art
  • Figure 3 is a spectrum distribution diagram in the absence of interference
  • Figure 4 is a spectrum distribution diagram in the presence of an interference situation
  • FIG. 5 is a schematic diagram of a main flow of an embodiment of a method for calculating a physiological parameter provided by the present invention
  • FIG. 6 is a schematic diagram of a main flow of another embodiment of a method for calculating physiological parameters provided by the present invention.
  • FIG. 7 is a schematic diagram of a main flow of still another embodiment of a method for calculating a physiological parameter provided by the present invention.
  • FIG. 8 is a schematic structural diagram of an embodiment of a digital processor used in a medical device for calculating physiological parameters according to the present invention
  • FIG. 9 is a schematic structural diagram of a compensation coefficient construction unit of FIG. 8.
  • FIG. 10 is a schematic structural diagram of another embodiment of a digital processor used in a medical device for calculating physiological parameters according to the present invention.
  • FIG. 11 is a schematic structural diagram of a compensation coefficient construction unit in FIG. 10;
  • FIG. 12 is a schematic structural diagram of still another embodiment of a digital processor used in a medical device for calculating physiological parameters according to the present invention.
  • Figure 13 is a schematic structural view of the compensation processing unit of Figure 12;
  • FIG. 14 is a schematic diagram of a venous blood pulsation disturbing arterial blood pulsation according to an embodiment of the present invention
  • 15 is a schematic flow chart showing a calculation process of a compensation coefficient in an embodiment of the present invention.
  • 16 is a schematic diagram of blood oxygen distribution in a spectrum segment in an interference state according to an embodiment of the present invention.
  • Figure 17 is a schematic view showing the flow of the venous oxygen compensation method in an embodiment of the present invention.
  • FIG. 18 is a schematic diagram of a process of performing a base frequency screening process in an embodiment of the present invention.
  • Figure 19 is a flow chart showing the spectrum array pattern method in one embodiment of the present invention.
  • 20 is a flow chart showing the simultaneous use of an intravenous oxygen compensation method and a spectrum array pattern method in one embodiment of the present invention
  • FIG. 21 is a schematic structural diagram of an embodiment of a medical device for calculating physiological parameters according to the present invention.
  • VOC Venous Oxygen Compensation
  • Step S10 selecting a time domain signal of a section corresponding to at least one signal of red light and/or infrared light obtained by sampling, performing time-frequency domain conversion on the time domain signal of the interval, and obtaining a corresponding frequency domain signal;
  • Step S12 selecting, in the frequency domain signal, all the reasonable spectral peak information, calculating energy information of the selected reasonable spectral peak, and forming a spectral peak energy ratio sequence of infrared light and red light; wherein Reasonable spectral peak information is to satisfy the spectral energy relationship (eg, whether there is a fundamental frequency relationship), the spectral amplitude relationship (whether the amplitude is not lower than a fixed value), the spectral positional relationship, and the spectral peak shape relationship (eg, the shape of the spectral peak) At least one of factors such as whether the frequency point is bilaterally symmetric or the like;
  • Step S14 constructing a stability coefficient according to the spectrum peak energy ratio sequence, and if the stability coefficient is low, constructing a compensation coefficient by using a spectral peak energy ratio sequence; wherein the stability coefficient is based on the red light and infrared
  • the deviation statistics of the light energy ratio or the deviation statistics based on the blood oxygen parameter values are constructed or an empirical coefficient.
  • the empirical coefficient is abstracted according to the physiological state and the signal-to-noise ratio characteristics. For example, when it is recognized as motion interference, the physiological blood oxygen cannot be too low; under long-term exercise, the physiological blood oxygen will have a downward trend; when the motion interference is severe, when the noise component When it is much larger than the signal component, blood oxygen approaches 80%, and so on.
  • a digital empirical coefficient value can be formed. Specifically, the stability coefficient is compared with a set threshold to determine whether the stability coefficient is low.
  • the mean, standard deviation, maximum value, and minimum value of the statistical peak energy ratio deviation sequence or the blood oxygen deviation sequence may be utilized to construct the stability coefficient according to a specific algorithm;
  • the following methods can be used to construct the compensation coefficients:
  • tabR is the R coefficient value obtained by querying the mapping table of a predetermined blood oxygen and R curve coefficient with the molecular input source as the blood oxygen value
  • tabR (Denomnator) is the molecular input source.
  • the blood oxygen value the R coefficient value obtained by querying the mapping table of the predetermined blood oxygen and R curve coefficients
  • Factor compensation is the compensation coefficient.
  • Step S16 using the compensation coefficient to compensate at least one of the time domain signal and/or the frequency domain signal, and calculating the physiological parameter according to the compensated time domain signal and/or the frequency domain signal. It is at least one of a blood oxygen parameter, a pulse rate parameter, a waveform area parameter, and a perfusion index parameter.
  • the further detailed flow chart in FIG. 5 is exemplified by taking the physiological parameter as the blood oxygen saturation parameter as an example.
  • step S10 selecting a time domain signal of a section corresponding to at least one signal of red light and/or infrared light obtained by sampling, performing time-frequency domain conversion on the time domain signal of the interval, and obtaining a corresponding frequency Domain signal
  • step S12 among the frequency domain signals, all reasonable spectral peak information is selected, the energy information of the selected reasonable spectral peak is calculated, and a spectral peak energy ratio sequence of infrared light and red light is formed;
  • step S14 a stability coefficient is constructed according to the spectral peak energy ratio sequence, and if the stability coefficient is low, the compensation coefficient is constructed using the spectral peak energy ratio sequence. Specifically, among the plurality of feature information obtained in the foregoing step, a difference between the maximum value and the minimum value is used as a stability coefficient, and if the stable system is less than or equal to a set first threshold, determining that the stability coefficient is low Need to build a compensation coefficient;
  • the process of constructing the compensation coefficients can be obtained by:
  • the mean value is selected as the denominator input source of the compensation coefficient calculation formula; if the blood oxygen sequence exceeding the mean value of the blood oxygen sequence accounts for at least half of the total number of sequences, then Use the maximum + standard deviation as the molecular input source for the compensation coefficient calculation formula, and vice versa Large value – standard deviation as the molecular input source for the compensation coefficient calculation formula;
  • the repetition factor is obtained in a predetermined manner, as a molecular input source of the compensation coefficient calculation formula, and the average value is used as a denominator input source of the compensation coefficient calculation formula;
  • tabR is the R coefficient value obtained by querying the mapping table of a predetermined blood oxygen and R curve coefficient with the molecular input source as the blood oxygen value
  • tabR (Denomnator) is the molecular input source.
  • the blood oxygen value the R coefficient value obtained by querying the mapping table of the predetermined blood oxygen and R curve coefficients
  • Factor compensation is the compensation coefficient.
  • the step of obtaining the repetition factor in a predetermined manner is specifically:
  • the compensation time coefficient is used to compensate the time domain signal corresponding to at least one of the red and/or infrared light, and the physiological parameter is further calculated using at least one compensated time domain signal.
  • the physiological parameter is at least one of a blood oxygen parameter, a pulse rate parameter, a waveform area parameter, and a perfusion index parameter. Specifically, the following two methods can be adopted:
  • a Power Spectrum Array (Power Spectrum Array, PSA) compensation process specifically, the method includes the following steps:
  • Step S20 selecting a time domain signal of a section corresponding to at least one signal of red light and/or infrared light obtained by sampling, performing time-frequency domain conversion on the time domain signal of the interval, and obtaining at least one frequency domain signal;
  • Step S22 in the frequency domain signal, select all reasonable spectral peak information therein, and calculate the selected Selecting a reasonable spectral peak position information and forming a sequence of spectral peak positions; wherein the reasonable spectral peak information is at least one of a spectral energy relationship, a spectral amplitude relationship, a spectral position relationship, and a spectral peak shape relationship.
  • Step S24 constructing a stability coefficient according to the sequence of spectral peak positions, and if the stability coefficient is low, constructing a time-varying array map by using the spectral peak position sequence, and calculating a compensation coefficient by using the array image
  • the stability coefficient is constructed based on at least one of a spectral energy ratio deviation, a spectral blood oxygen deviation, a base frequency group state, and a reasonable number of spectral peaks, and may be determined according to the number of stability factors and/or the weight value. Whether the stability factor is low;
  • the stability factor is determined according to the state of the base octave group in the sequence of spectral peak positions; for example, when the pair of base octaves cannot be identified in the sequence of spectral peak positions, the stability coefficient is considered to be low;
  • the step of constructing the time-varying array map by using the sequence of spectral peak positions and calculating the compensation coefficient by using the array map includes:
  • the related information of the spectral peak includes: each spectral peak position information PeakArray; each spectral peak position weight information WeewedArray; the spectral peak storage quantity information ArrayIndex;
  • Step S161 filtering a predetermined number of spectral peaks in the frequency domain signal, sequentially filling the buffer, filling the position information of each spectral peak into the PeakArray, adding one to the corresponding position weight information of the WewokedArray, and adding the total length ArrayIndex.
  • the spectral peak position weight information WewokedArray in the buffer is directly added by one.
  • Step S162 in the buffer, assuming each spectral peak as a suspected fundamental frequency peak, traversing all other spectral peaks, determining whether other spectral peaks and the suspected fundamental frequency peak satisfy a multiple of the frequency, and if so, eliminating the Deriving the correlation information of the octave peak in the buffer, and converting the position weight coefficient of the frequency doubling peak according to the frequency doubling times and accumulating the position weight coefficient of the corresponding suspected fundamental frequency peak; otherwise, retaining the correlation of other spectrum peaks information;
  • Step S163 in the last remaining suspected spectral peaks, select three spectral peaks whose position weight coefficients are ranked in the front row, and determine one of them as the final reasonable spectral peak (which may be regarded as a fundamental frequency peak in some examples). And obtaining the pulse rate value according to the determined final reasonable spectral peak. Similarly, it can be understood that the corresponding blood oxygen saturation can be obtained by energy calculation according to the final reasonable spectral peak.
  • the step further includes:
  • the spectral peak is determined as the final reasonable spectral peak
  • the number of spectral peaks whose weight coefficient is greater than the third threshold is at least two, further according to Physiological characteristics, determining one of the spectral peaks as the final reasonable spectral peak, and the information of the final reasonable spectral peak described later as the compensation coefficient;
  • Step S26 using the compensation coefficient to compensate at least one of the time domain signal and/or the frequency domain signal, and calculating a physiological parameter based on the compensated time domain signal and/or the frequency domain signal;
  • the physiological parameter is at least one of a blood oxygen parameter, a pulse rate parameter, a waveform area parameter, and a perfusion index parameter; wherein, in one example, the compensation time coefficient is used to compensate the time domain signal and/or the frequency domain signal
  • the at least one path may be specifically configured to: construct a band pass filter according to the compensation coefficient, and perform filtering processing on at least one of the time domain signal and/or the frequency domain signal to implement a compensation function.
  • VOC Venous Oxygen Compensation
  • PSA Power Spectrum Array
  • Step S30 selecting a time domain signal of a section corresponding to at least one signal of red light and/or infrared light obtained by sampling, performing time-frequency domain conversion on the time domain signal of the interval, and obtaining at least one frequency domain signal;
  • Step S32 in the frequency domain signal, select all reasonable spectral peak information, calculate energy and or position information of the selected reasonable spectral peak, and form a spectral peak energy ratio sequence and a spectral peak position sequence;
  • the reasonable spectral peak information is to satisfy at least one of a spectral energy relationship, a spectral amplitude relationship, a spectral position relationship, and a spectral peak shape relationship; the process of forming a spectral peak energy ratio sequence and a spectral peak position sequence may be respectively referred to the foregoing pair.
  • Step S34 constructing a stability coefficient according to the signal characteristics of at least one of the spectral peak energy ratio sequence and the spectral peak position sequence, and determining whether the stability coefficient of at least one of the sequences is low, and if the stability coefficient is low, using the Constructing a compensation coefficient for at least one signal characteristic of the sequence; wherein the stability coefficient is constructed based on at least one of a spectral energy ratio deviation, a spectral blood oxygen deviation, a base frequency group state, and a reasonable number of spectral peaks;
  • the compensation coefficient is an energy ratio deviation coefficient or a blood oxygen deviation coefficient calculated based on the spectral energy sequence, or a position coefficient statistically obtained based on a sequence of spectral position changes with time.
  • step S14 in FIG. 5 For the specific process of constructing the stability coefficient according to the spectral peak energy ratio order and constructing the compensation coefficient, refer to the foregoing description of step S14 in FIG. 5;
  • step S24 in FIG. 6 For the specific process of constructing the stability coefficient of the spectrum peak position sequence and constructing the compensation coefficient, refer to the foregoing description of step S24 in FIG. 6; details are not described herein.
  • Step S36 using the compensation coefficient to compensate the time domain signal and/or the frequency domain signal;
  • the compensated time domain signal and/or the frequency domain signal are further calculated to obtain physiological parameters, the physiological parameters being at least one or more of a blood oxygen parameter, a pulse rate parameter, a waveform area parameter, and a perfusion index parameter.
  • the step includes at least one of: using the compensation coefficient, the selected time domain signal is subjected to gain processing to implement compensation; or
  • the selected frequency domain signal is filtered by the compensation coefficient to achieve compensation.
  • the present invention also provides a medical device for calculating physiological parameters.
  • the calculation method of a physiological parameter disclosed by the present invention can be realized by functional modularization, and is integrated as a plug-in in other auxiliary diagnostic equipment (such as monitoring equipment, defibrillator, AED, automatic resuscitation instrument, electrocardiograph, etc.). It can also be used as a single-parameter medical device for monitoring physiological parameters.
  • the physiological parameters include at least one of blood oxygen parameters, pulse rate parameters, waveform area parameters, and perfusion index parameters.
  • FIG. 21 a schematic structural view of a medical device disclosed in the present invention is shown, in which a single parameter medical device is shown; in this embodiment, the medical device includes:
  • At least one blood oxygen sensor for measuring a subject's fingers, forehead, earlobe, toes, soles, and the like.
  • the oximetry sensor comprises at least one illuminating tube and a receiving tube, the illuminating tube emitting at least two optical signals of different wavelengths for transmitting through the body tissue of the subject, in one example, the illuminating tube can emit a red light and All the way to infrared light signals.
  • the receiving tube receives at least two optical signals transmitted through the body tissue of the subject and converts to at least two electrical signals. In one example, the two electrical signals are red and infrared light signals;
  • An analog to digital converter coupled to the sensor to convert the electrical signal into a digital signal comprising at least a portion of a characteristic of a physiological tissue
  • a digital processor coupled to the analog to digital converter, the digital processor performing the following processing:
  • the compensation signal is used to compensate the digital signal of the segment and/or the corresponding frequency domain signal, and the physiological parameter is calculated based on the compensated digital signal and/or the frequency domain signal.
  • a communication unit connected to the digital processor, and transmitting the physiological parameter calculated by the digital processor.
  • the blood oxygen sensor includes two types of transmissive and reflective, and can be generally worn on the finger, forehead, earlobe, toe, and sole of the subject for physiological parameter measurement, including blood oxygen parameters, pulse rate parameters, waveform area parameters, At least one of the perfusion index parameters.
  • the display unit may be a local display unit, or may be remotely communicated to the remote display unit by wired/wireless means.
  • the display unit provides the user with a perceptible parameter representation through characters, values, waveforms, bar graphs, voice prompts, and the like.
  • At least a portion of the physiological tissue is characterized by one or more of the optical characteristics of oxygenated hemoglobin, deoxyhemoglobin, trioxyhemoglobin, total hemoglobin or carbon monoxide in the blood.
  • the reasonable spectral peak satisfies at least one of a spectral energy relationship, a spectral amplitude relationship, a spectral position relationship, and a spectral peak shape relationship.
  • the stability factor is constructed based on a deviation statistic of the red light and infrared light energy ratio, or is an empirical coefficient.
  • the stability factor adjusts its stability weight according to at least one of a fundamental frequency group characteristic and a spectral peak shape rationality.
  • the digital processor uses the spectral peak energy ratio sequence to construct a compensation coefficient, which is specifically:
  • the denominator of the mean value converted into the compensation coefficient calculation formula is selected, and the (mean + standard deviation) or (mean-standard deviation) is selected to be converted into the numerator of the compensation coefficient calculation formula by the coefficient table, and the ratio is obtained to obtain the compensation coefficient.
  • the physiological parameter is at least one of blood oxygenation, pulse rate, waveform area, perfusion index, and the like.
  • FIG. 8 there is shown a schematic structural diagram of an embodiment of a digital processor used in a medical device for calculating physiological parameters provided by the present invention, which uses Venous Oxygen Compensation (Venous Oxygen Compensation, VOC) is compensated, please refer to FIG. 9 together.
  • the digital processor includes:
  • the time-frequency domain conversion unit 11 is configured to select a time domain signal of a section corresponding to at least one signal of red light and/or infrared light obtained by sampling, and perform time-frequency domain conversion on the time domain signal of the interval to obtain Corresponding frequency domain signal;
  • the energy ratio sequence constructing unit 12 is configured to select all the reasonable spectral peak information in the frequency domain signal, calculate energy information of the selected reasonable spectral peak, and constitute a spectral peak energy ratio of the infrared light and the red light. a sequence; wherein the reasonable spectral peak information is to satisfy a spectral energy relationship, a spectrum At least one of an amplitude relationship, a spectral positional relationship, and a spectral peak shape relationship;
  • a compensation coefficient construction unit 13 configured to construct a stability coefficient according to the spectral peak energy ratio sequence, and if the stability coefficient is low, construct a compensation coefficient by using a spectral peak energy ratio sequence; the stability coefficient is based on the red light And a deviation statistic of the infrared light energy ratio or a deviation statistic based on the blood oxygen parameter value, or an empirical coefficient; comparing the stability coefficient with a set threshold to determine whether the stability coefficient is low.
  • the compensation processing unit 14 is configured to compensate the time domain signal corresponding to the red light and/or the infrared light by using the compensation coefficient, and further calculate the physiological parameter by using at least one compensated time domain signal.
  • the compensation coefficient construction unit 13 further includes:
  • the stability coefficient construction sub-unit 130 is configured to obtain the stability coefficient according to a specific algorithm by using a mean, standard deviation, maximum value, and minimum value of a statistical peak energy ratio deviation sequence or a blood oxygen deviation sequence;
  • the determining unit 131 is configured to determine whether the stability coefficient constructed by the stability coefficient construction sub-unit 130 is low;
  • the numerator and denominator selecting unit 132 is configured to select the mean value of the blood oxygen parameter as the denominator input source of the compensation coefficient calculation formula; and select the mean + standard deviation or the mean value - standard deviation as the molecular input source of the compensation coefficient calculation formula;
  • the calculating unit 134 is configured to substitute the molecular input source and the denominator input source selected by the numerator denominator selecting unit into the following compensation coefficient calculation formula, and calculate a corresponding compensation coefficient:
  • tabR is the R coefficient value obtained by querying the mapping table of a predetermined blood oxygen and R curve coefficient with the molecular input source as the blood oxygen value
  • tabR (Denomnator) is the molecular input source.
  • the blood oxygen value the R coefficient value obtained by querying the mapping table of the predetermined blood oxygen and R curve coefficients
  • Factor compensation is the compensation coefficient.
  • the present invention also provides a medical device for calculating physiological parameters, in this embodiment, the medical device includes:
  • a sensor comprising at least one light emitting tube and at least one receiving tube, the light emitting tube emitting at least two optical signals of different wavelengths transmitted through the physiological tissue, the receiving tube receiving at least two optical signals transmitted through the physiological tissue, and converting to electric signal;
  • An analog to digital converter coupled to the sensor to convert the electrical signal into a digital signal comprising at least a portion of a characteristic of a physiological tissue
  • a digital processor coupled to the analog to digital converter, the digital processor performing the following processing:
  • the compensation coefficient is used to compensate at least one of the time domain signal and/or the frequency domain signal, and the physiological parameter is calculated based on the compensated time domain signal and/or the frequency domain signal.
  • a display unit coupled to the digital processor to display physiological parameters calculated by the digital processor;
  • a communication unit connected to the digital processor, and transmitting the physiological parameter calculated by the digital processor.
  • the two optical signals are red light and infrared light
  • At least a portion of the physiological tissue is characterized by one or more of the optical characteristics of oxygenated hemoglobin, deoxyhemoglobin, trioxyhemoglobin, total hemoglobin or carbon monoxide in the blood.
  • the reasonable spectral peak is at least one or more of a spectral energy relationship, a spectral amplitude relationship, a spectral position relationship, and a spectral peak shape relationship.
  • the stability coefficient is constructed based on at least one of a spectral energy ratio deviation, a spectral blood oxygen deviation, a base frequency group state, a reasonable spectral peak number, and a spectral peak shape rationality. to make.
  • the stability factor adjusts its stability weight according to at least one of a fundamental frequency group characteristic and a spectral peak shape rationality.
  • the stability coefficient is stable or not based on the number of stability factors and/or the weight value.
  • the physiological parameter is at least one of blood oxygenation, pulse rate, waveform area, perfusion index, and the like.
  • FIG. 10 there is shown a schematic structural diagram of another embodiment of a digital processor employed in a medical device for calculating physiological parameters provided by the present invention, which uses a spectral array diagram method (Power Spectrum Array, PSA) for compensation, please refer to FIG. 11 together.
  • the digital processor includes:
  • the time-frequency domain conversion unit 11 is configured to select a time domain signal of a section corresponding to at least one signal of red light and/or infrared light obtained by sampling, and perform time-frequency domain conversion on the time domain signal of the interval to obtain At least one frequency domain signal;
  • the spectral peak position sequence obtaining unit 15 is configured to select all the reasonable spectral peak information in the frequency domain signal, calculate position information of the selected reasonable spectral peak, and form a sequence of spectral peak positions, the reasonable
  • the spectral peak information is at least one or more of a spectrum energy relationship, a spectral amplitude relationship, a spectral position relationship, and a spectral peak shape relationship;
  • a compensation coefficient construction unit 16 configured to construct a stability coefficient according to the sequence of spectral peak positions, and if the stability coefficient is low, construct a time-varying array map by using the spectral peak position sequence, and pass the array The figure calculates a compensation coefficient, and the stability coefficient is constructed based on at least one of a spectral energy ratio deviation, a spectral blood oxygen deviation, a base frequency group state, and a reasonable number of spectral peaks;
  • the compensation processing unit 17 is configured to compensate at least one of the time domain signal and/or the frequency domain signal by using the compensation coefficient, and obtain a physiological parameter based on the compensated time domain signal and/or the frequency domain signal.
  • the physiological parameter is at least one of a blood oxygen parameter, a pulse rate parameter, a waveform area parameter, and a perfusion index parameter.
  • the compensation coefficient construction unit 16 includes:
  • the buffer unit 160 is configured to establish a buffer to store related information of a spectrum peak, where the correlation information of the spectrum peak includes: each spectral peak position information; each spectral peak position weight information; and quantity information stored by the spectrum peak;
  • the spectral peak information recording unit 162 is configured to filter a predetermined number of spectral peaks in the frequency domain signal, fill the buffer into the buffer, fill in the position information of each spectral peak, and add one to the corresponding position weight information.
  • the total length quantity information is incremented by one;
  • the traversing processing unit 164 is configured to: in the cache, assume that each spectral peak is used as a suspected fundamental frequency peak, traverse all other spectral peaks, and determine whether other spectral peaks and the suspected fundamental frequency peak satisfy a multiple of the frequency, if satisfied And eliminating related information of the frequency doubling peak in the buffer, and converting the position weight coefficient of the frequency doubling peak according to the frequency doubling times and accumulating the position weighting coefficient of the corresponding suspected fundamental frequency peak; Information about the spectral peaks;
  • the compensation coefficient obtaining unit 166 is configured to select, in the last remaining suspected spectral peaks, three spectral peaks whose position weight coefficients are ranked in the front row, and determine one of them as the final reasonable spectral peak, and the reasonable spectral peak The information is used as the compensation factor.
  • the present invention also provides a medical device for calculating physiological parameters, in this embodiment, the medical device includes:
  • a sensor comprising at least one light emitting tube and at least one receiving tube, the light emitting tube emitting at least two optical signals of different wavelengths transmitted through the physiological tissue, the receiving tube receiving at least two optical signals transmitted through the physiological tissue, and converting to electric signal;
  • An analog to digital converter coupled to the sensor to convert the electrical signal into a digital signal comprising at least a portion of a characteristic of a physiological tissue
  • a digital processor coupled to the analog to digital converter, the digital processor performing the following processing:
  • the compensation coefficient is used to compensate at least one of the time domain signal and/or the frequency domain signal, and the physiological parameter is calculated based on the compensated time domain signal and/or the frequency domain signal.
  • the two optical signals are red light and infrared light.
  • At least a portion of the physiological tissue is characterized by one or more of the optical characteristics of oxygenated hemoglobin, deoxyhemoglobin, trioxyhemoglobin, total hemoglobin or carbon monoxide in the blood.
  • the reasonable spectral peak is at least one of a spectral energy relationship, a spectral amplitude relationship, a spectral position relationship, and a spectral peak shape relationship.
  • the stability factor is constructed based on at least one of a spectral energy ratio deviation, a spectral blood oxygen deviation, a base frequency group state, and a reasonable number of spectral peaks.
  • the compensation coefficient is an energy ratio deviation coefficient calculated based on a sequence of spectral peak energy ratios, or a position coefficient statistically obtained based on a sequence of spectral peak positions over time.
  • the processing of using the compensation coefficient to compensate at least one of the time domain signal and/or the frequency domain signal is specifically:
  • the selected frequency domain signal is filtered by the compensation coefficient to achieve compensation.
  • the physiological parameter is blood oxygenation, pulse rate, waveform area, irrigation Note at least one of the index, etc.
  • FIG. 12 there is shown a schematic structural diagram of another embodiment of a digital processor employed in a medical device for calculating a physiological parameter, which is combined with a venous oxygen compensation method (Venous Oxygen). Compensation, VOC) and the Spectrum Array (PSA) are compensated.
  • VOC venous oxygen compensation method
  • PSA Spectrum Array
  • the time-frequency domain conversion unit 11 is configured to select a time domain signal of a section corresponding to at least one signal of red light and/or infrared light obtained by sampling, and perform time-frequency domain conversion on the time domain signal of the interval to obtain At least one frequency domain signal;
  • Corresponding sequence construction unit 18 is configured to select all reasonable spectral peak information among the frequency domain signals, calculate energy and/or position information of the selected reasonable spectral peaks, and form a spectral peak energy ratio sequence and/or Or a sequence of spectral peak positions; the reasonable spectral peak information is at least one of a spectrum energy relationship, a spectral amplitude relationship, a spectral position relationship, and a spectral peak shape relationship; it can be understood that the corresponding sequence construction unit 18 has a map at the same time. 8 medium energy ratio sequence construction unit 13 and the function of spectrum peak position sequence obtaining unit 15 in FIG. 10, the specific details may refer to the corresponding descriptions mentioned above;
  • the compensation coefficient construction unit 16 is configured to construct a stability coefficient according to at least one signal characteristic of the sequence, and determine whether the stability coefficient of at least one path of the sequence is low. If the stability coefficient is low, at least one channel of the sequence is used.
  • the signal characteristic constructs a compensation coefficient; the stability coefficient is constructed based on at least one of a spectral energy ratio deviation, a spectral blood oxygen deviation, a base frequency group state, and a reasonable number of spectral peaks; the compensation coefficient constructing unit 16 may be based on the spectrum
  • the compensation coefficient is constructed by calculating an energy ratio deviation coefficient or a blood oxygen deviation coefficient of the energy sequence or a position coefficient statistically obtained based on a sequence of spectral position changes with time. It can be understood that the compensation coefficient construction unit 16 has the functions of the compensation coefficient construction unit 16 of FIG. 8 and the compensation coefficient construction unit 16 of FIG. 10, and specific details can be referred to the foregoing corresponding description;
  • a compensation processing unit 17 for compensating the time domain signal and/or the frequency domain signal by using the compensation coefficient; and further calculating a physiological parameter based on the compensated time domain signal and/or the frequency domain signal, wherein
  • the physiological parameter is at least one of a blood oxygen parameter, a pulse rate parameter, a waveform area parameter, and a perfusion index parameter.
  • the compensation processing unit 17 further includes:
  • the time domain compensation unit 170 is configured to perform gain processing by using the compensation coefficient and the selected time domain signal to implement compensation; or
  • the frequency domain compensation unit 172 is configured to filter the selected frequency domain signal by using the compensation coefficient. Processing to achieve compensation.
  • VOC Venous Oxygen Compensation
  • PSA Power Spectrum Array
  • VOC venous oxygen compensation method
  • venous blood flows relatively slowly due to its physiological characteristics and can be considered as part of the DC flux.
  • Venous oxygen saturation does not have any effect on normal blood oxygen measurement.
  • the venous blood is affected by the disturbance, and a venous pulsation is generated, and the AC exchange amount formed by the pulsation of the vein is mixed into the AC exchange amount formed by the arterial blood pulsation.
  • the oxygen saturation calculated at this time must be deviated from the true value; from the physiological point of view, it can be understood that the venous oxygen saturation is mixed in the arterial oxygen saturation, resulting in the final oxygen saturation deviation. Blood gas value.
  • a schematic diagram of venous bloodbeat interference with arterial bloodbeat is given.
  • the time domain algorithm has almost no way to accurately calculate the blood oxygen saturation value; in addition, it is very difficult to obtain random interference source characteristics.
  • each spectrum segment in the frequency domain signal uniquely corresponds to blood oxygen saturation, and theoretically, each spectrum segment can calculate blood oxygen saturation; and at the same time, due to the randomness of the vein interference, it is superimposed on Interference in different frequency bands is also a matter of severity.
  • the venous oxygen compensation method (VOC) proposed in the present application is constructed based on these two hypothetical models.
  • VOC venous oxygen compensation method
  • VOC intravenous oxygen compensation
  • the spectral peaks satisfying the conditions are obtained from the spectrum signals, and the blood oxygen saturation of each spectral peak is calculated to obtain a blood oxygen saturation sequence.
  • the spectrum peak that satisfies the condition refers to the range that is included in the statistical analysis in terms of amplitude, energy, width, and morphology.
  • the judgment criterion is established based on the characteristics of the physiological signal and the basic method of digital signal processing. Based on the blood oxygen saturation sequence, the mean (vMean), standard deviation (vStd), maximum value (vMax), minimum value (vMin), and the like of blood oxygen are statistically obtained.
  • the compensation coefficient calculation formula is constructed, and the compensation coefficient is calculated based on the difference distribution characteristics.
  • the difference between vMax and vMin is compared with a first threshold (Threshold1), and the vStd value is compared with a second threshold (Threshold2) to determine whether the normal mode or the interference mode is selected.
  • the selection of the first threshold and the second threshold is an empirical coefficient obtained according to physiological parameter characteristics and characteristics of a blood oxygen system. For example, in one example, the first threshold (Threshold 1) may take 15% blood oxygen deviation, and the second threshold (Threshold2) can take 5% blood oxygen deviation.
  • the second step if it is the normal mode, indicates that the fluctuation caused by venous oxygen is relatively small, and the mean and standard deviation can be selected as the input source of the compensation coefficient calculation formula.
  • the mean is used as the input source of the denominator. If the number of blood oxygen in the blood oxygen sequence exceeding the mean is at least half of the total number of sequences, vMean+vStd is used as the molecule, and vMean–vStd is used as the molecule.
  • the third step, and the second step is to select the relationship. If it is an abnormal mode, it is necessary to introduce a repetition factor.
  • repetition factors There are two sources of repetition factors: 1) statistically collect the maximum blood oxygen concentration in a blood oxygen sequence that satisfies a certain value (eg, ⁇ 2%), and take the mean of the set as a repetition factor, where ⁇ 2% is The empirical coefficient can be adjusted according to the actual change; 2) the stable segment of the historical trend of blood oxygen is selected, for example, the stable trend of 4s to 8s, and the mean of the set is calculated as the repetition factor, and the 4s to 8s time period is also the empirical coefficient. Adjust according to actual changes.
  • the repetition factor is used as the molecular input source and the mean is used as the denominator input source.
  • the numerator and denominator parameters are input into the following formula 2, and the compensation coefficient is calculated.
  • the calculation formula is as follows, where tabR is a mapping table of blood oxygen and R curve coefficients (as described above), and the corresponding R coefficient value can be obtained by inversely checking the input blood oxygen value.
  • the compensation coefficient is the ratio of the R coefficient value of the molecular blood oxygen to the R coefficient value of the denominator blood oxygen.
  • TabR is the R coefficient value obtained by using the numerator as the input source and entering the mapping table of blood oxygen and R curve coefficients.
  • TabR (Denominator) is the denominator as the input source and then into the blood oxygen and The R coefficient value found after the mapping table of the R curve coefficients.
  • VOC venous oxygen compensation method
  • Figure 16 a schematic diagram of the blood oxygen distribution of the spectrum segment in the interference state is given. It is assumed that the spectrum signal has four spectral peaks of A, B, C, and D in the physiological bandwidth of 0.3 Hz to 5 Hz, and the spectrum is assumed. The signal is affected by the random pulsation of venous oxygen. The blood oxygen saturation values of each spectral peak converted by the ratio of infrared light and red light are 96%, 85%, 90%, 87%, respectively. There is a deviation between the saturation values.
  • the first threshold eg, 15%
  • vStd is less than the second threshold (eg, 5%)
  • the normal mode is selected.
  • Query the mapping table of blood oxygen and R curve coefficients respectively obtain the R value corresponding to the numerator and the denominator, and substitute the formula 2 to calculate the correction factor of 1.172.
  • the compensation coefficient is used to compensate for the loss of the time domain signal (the frequency domain signal is obtained based on the signal transformation of this segment) due to interference. That is, the time domain signal used for the frequency domain transform is multiplied by the compensation coefficient to obtain a compensation signal.
  • compensation is only made for the red light signal as an example.
  • the compensation coefficient can also be differentiated to compensate for each signal.
  • the time domain signal that completes the compensation is used to transform to the frequency domain signal again, and then the frequency domain signal is used to calculate accurate parameters such as blood oxygen.
  • the embodiment of the present invention is an example based on a frequency domain algorithm. In the actual application, the frequency domain method may also be omitted. Based on the compensated time domain signal, the time domain algorithm is used to obtain accurate parameters such as blood oxygen.
  • a schematic diagram of the framework of the venous oxygen compensation method is given.
  • the whole process is: selecting a time domain signal interval, performing time-frequency domain conversion, and converting the time domain signal to the frequency domain signal.
  • physiological signal characteristics and basic knowledge of digital signal processing eg, physiological pulse wave frequency range, fundamental frequency principle, morphological characteristics, etc.
  • select a reasonable spectral peak and calculate the blood oxygen saturation parameter of the spectral peak.
  • the blood oxygen distribution series is statistically analyzed, and a series of characteristic information is obtained, and according to the characteristic information, whether the blood oxygen saturation has a deviation (or a small deviation) is determined.
  • the compensation coefficient is calculated based on the feature information and compensated into the time domain signal. Finally, the compensated time domain signal is used to perform time-frequency domain conversion, and a reasonable spectrum peak is selected based on the new frequency domain signal and the final blood oxygen saturation parameter is calculated.
  • VOC venous oxygen compensation method
  • the interferences Considering the physiological characteristics, most of the interferences appear as random white noise distribution; a few regular interferences (such as Parkinson's disease, etc.), because of the relatively low amplitude of vibration, the vibration frequency is relatively high, and the blood is relatively high. Oxygen sampling signal There is a substantial impact, that is, does not affect the measurement of blood oxygen parameters. Although the interference signal disturbs the spectrum signal, the physiological spectrum is not recognized, but regardless of how the interference signal changes, the characteristics of the physiological spectrum peak always exist at a certain frequency point and do not change or change relatively slowly within a certain period of time (physiological characteristics) At the same time, the interference noise is randomly distributed, and the spectral characteristics change with time.
  • spectrum peak position information that is, the frequency point (PeakArray); spectrum peak position weight information (WeUNEdArray); spectrum peak storage quantity information (ArrayIndex), where the position information and the weight information share the quantity information.
  • spectrum peak position information that is, the frequency point (PeakArray); spectrum peak position weight information (WeUNEdArray); spectrum peak storage quantity information (ArrayIndex), where the position information and the weight information share the quantity information.
  • the number of spectral peaks examined should not be too large. Too much will increase the complexity and computational complexity of the algorithm identification. For example, 20 can be selected as the upper limit by default, which is beyond the analysis. It can be understood that in practical applications, the appropriate number of analysis can also be selected according to the requirements of the system.
  • the reasonable spectral peaks in the frequency domain signal are filtered and filled into the above buffer.
  • the criteria for screening are based on the combination of information such as the energy, amplitude, shape, and location of the spectral peaks.
  • the buffer needs to accumulate spectrum information for a certain period of time before the physiological spectrum peak can be identified.
  • This time needs to be set according to the actual needs of the system. For example, it can be set to 10 calculations, that is, the time trend of 20S.
  • 10 times of information storage is satisfied, the relevant trend analysis is started, and at the same time, after the trend analysis is completed, the earliest stored spectrum information is reduced.
  • the true physiological spectral peak is identified based on the cached information.
  • the frequency doubling peak in the PeakArray is eliminated, and the weight coefficient of the octave peak is converted according to the frequency doubling times and added to the weight coefficient of the fundamental frequency peak (for example, the fundamental frequency peak weight coefficient is 3, 2 times frequency doubling)
  • the peak weight coefficient is 2, and the weighting coefficient 2 is divided by the frequency multiplication number 2, and the adjusted weight coefficient is obtained, and the weight coefficient of the fundamental frequency peak is accumulated, that is, the fundamental frequency peak weight coefficient is adjusted to 4), and the frequency doubling peak is eliminated.
  • the information stored in PeakArray and WewokedArray is initialized to 0. The detailed flow is shown in Figure 18 below.
  • each spectral peak is a fundamental frequency peak, traversing all other spectral peaks, and judging whether other spectral peaks and the assumed fundamental frequency peak satisfy the frequency.
  • the multiple relationship if it is satisfied, is its frequency doubling peak, eliminating the related information of the octave peak, and vice versa.
  • the general multiplication frequency is more than 4 times, and the influence is relatively small.
  • the setting can be selected according to the needs of the system.
  • each spectrum peak is traversed to complete the elimination of the relevant information of the frequency doubled peak. Finally, the first three peaks with relatively large weight coefficients are selected as the input information for the next step.
  • the threshold can be set to 8, indicating that there is a stable spectrum peak for 16s in succession, and the actual application can be set according to the needs of the system. If two or more spectral peaks satisfy the weighting coefficient greater than or equal to the set threshold, it is necessary to further make a judgment based on the physiological characteristics. For example, in the interference state, the physiological pulse rate cannot be too high or low. Select the most reasonable spectral peak that meets the set threshold, calculate the pulse rate parameter and output it, and optimize the weight coefficients of other spectral peaks according to the state.
  • a schematic diagram of the flow spectrum diagram method is given. It can be seen that the spectrum peak information cache is first established, and the cache is filled in chronological order. When the buffer fill is full, the stored spectral peaks are simplified according to the fundamental frequency multiplication principle, and the spectral peaks satisfying the set threshold in the simplified spectral peaks are selected. If the spectral peaks satisfying the condition are greater than or equal to 2, the criteria for conversion based on physiological characteristics are excluded, and finally a reasonable spectral peak is selected, and the pulse rate parameter is calculated based on the spectral peak.
  • the Spectrum Array Method can eliminate the pulse rate measurement deviation caused by interference, and can accurately identify the physiological spectrum information even in the long-term interference state, which greatly provides the accuracy of the calculation of the pulse rate parameter under the interference state. .
  • the venous oxygen correction method can recognize the interference of venous oxygen and compensate for the blood oxygen deviation caused by the interference.
  • the spectral array map method (PSA) can accurately identify the pulse rate information in continuous interference.
  • Each of the two methods has the ability to identify and process the interference. Therefore, when the two methods are combined, the recognition and suppression of the interference can be significantly improved, thereby obtaining the accuracy of the calculation of the blood oxygen parameters and the accuracy of the pulse rate parameters. Increase in range.
  • Fig. 20 an example of a comprehensive application of two schemes in the method provided by the present invention is shown. In practical applications, these steps can be adaptively added, deleted, and adjusted according to system requirements.
  • the general process is as follows: select a specified time domain signal, perform time-frequency conversion, and perform spectral peak search and identification based on the converted spectral signal, and calculate the blood oxygen value of each reasonable spectral peak.
  • the blood oxygen value of the statistically reasonable peak is obtained by preparing relevant statistical information for the VOC, and also records the information of all the spectral peaks obtained in this calculation for the PSA.
  • the fundamental frequency peak is directly selected to calculate the pulse rate, and the blood oxygen and pulse rate values of the final result are output. If the base octave pair is not satisfied, enter the PSA method and calculate a reasonable spectral peak.
  • the pulse rate parameter is calculated and output; otherwise if the blood oxygen level is stable, then The pulse rate value is calculated and the blood oxygen and pulse rate values of the final result are output; if it is the first calculation (the time domain signal does not cancel the noise/compensate for venous oxygen) and the blood oxygen level is unstable, the VOC identification branch is entered. A specific filter for the spectral peaks obtained based on the PSA method is added to the VOC identification branch. Combining the compensation coefficient with a specific filter, the time domain signal is compensated and noise canceled, and then the time-frequency domain transform is performed again and the relevant parameters are calculated.
  • the embodiment of the invention can greatly improve the calculation accuracy of the blood oxygen and pulse rate parameters under the condition of weak perfusion and exercise, and brings an excellent clinical performance experience to the customer. , can greatly improve the application and promotion of blood oxygen parameters;
  • the present invention is implemented by combining the characteristics of the time-frequency domain signal with the characteristics of the human physiological parameters, and using the venous oxygen compensation method and the spectral array map method to improve the calculation under weak irrigation and motion states in the presence of interference.
  • Pulse rate value and accuracy of blood oxygen parameters can eliminate the deviation of blood oxygen measurement caused by interference, infinitely approach the true physiological blood oxygen value, and greatly provide the accuracy of the calculation of blood oxygen parameters under the interference state;
  • the spectrum array map method can eliminate the interference caused by the interference Pulse rate measurement deviation, even in the long-term interference state, can accurately identify the physiological spectrum information, greatly providing the accuracy of the calculation of the pulse rate parameter in the interference state;
  • the method provided by the embodiment of the present invention has low computational complexity and low demand for computing resources.

Landscapes

  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • Surgery (AREA)
  • Public Health (AREA)
  • Pathology (AREA)
  • Veterinary Medicine (AREA)
  • Biomedical Technology (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Medical Informatics (AREA)
  • Molecular Biology (AREA)
  • Biophysics (AREA)
  • Animal Behavior & Ethology (AREA)
  • General Health & Medical Sciences (AREA)
  • Physiology (AREA)
  • Signal Processing (AREA)
  • Cardiology (AREA)
  • Artificial Intelligence (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Psychiatry (AREA)
  • Spectroscopy & Molecular Physics (AREA)
  • Optics & Photonics (AREA)
  • Power Engineering (AREA)
  • Mathematical Physics (AREA)
  • Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)

Abstract

一种生理参数的计算方法及相应的医疗设备，该计算方法包括：选择通过采样获得的红光和/或红外光的至少一路信号所对应一段区间的时域信号，进行时频域转换，获得相应的频域信号（S10）；选择其中所有合理的频谱峰信息，计算所选择的合理的频谱峰的能量信息，并构成频谱峰能量比值序列（S12）；根据所述频谱峰能量比值序列构建稳定系数，如果所述的稳定系数偏低，则利用频谱峰能量比值序列构建补偿系数（S14）；使用所述的补偿系数补偿所述的时域信号和/或频域信号的至少一路，基于已补偿的时域信号和/或频域信号，计算获得生理参数（S16）。该方法可以在弱灌性和运动状态下提升计算脉率值以及血氧参数计算准确度，计算简单，花费少，适应性广。

Description

一种生理参数的计算方法及相应的医疗设备 技术领域
本发明涉及生理参数监测领域，尤其涉及一种生理参数的计算方法及相应的医疗设备。
背景技术
人体的新陈代谢过程是生物氧化过程，而新陈代谢过程所需要的氧，是通过呼吸系统进入人体血液，与血液红细胞中的还原血红蛋白(Hb)，结合成氧合血红蛋白(HbO2)，再输送到人体各部分组织细胞中。当机体氧供与氧耗之间出现不平衡，就意味着缺氧情况的发生。缺氧对机体有着巨大的影响。故，动脉血氧浓度的实时监测在临床救护中非常重要。
血氧饱和度(SpO2)是血液中氧合血红蛋白(HbO2)容量占总血红蛋白(Hb+HbO2)容量的百分比，即血液中血氧的浓度。血氧饱和度(SpO2)作为动脉血氧饱和度(SaO2)的间接反映，以其无创、简单、准确、快速等特点，迅速被各大应用领域采纳和使用。
脉搏血氧技术应用的如此广泛，随之而来，对脉搏血氧准确性的要求也水涨船高。本领域人员可知，评价脉搏血氧性能的优劣性，有两大关键因素：弱灌注性能和运动性能。如果被测患者存在弱灌注性能差或及运动性能差的情形，则对脉搏血氧性能的准确性和稳定性要求就更加苛刻。
自1998年Masimo公司推出“运动和弱灌注下准确测量病人血氧饱和度技术”以来，Masimo公司迅速确立了其血氧行业的领先地位。其技术的核心思想是确立系数Ra和Rv,其中Ra与动脉血氧饱和度相关，Rv与静脉血氧饱和度相关。通过R系数、红外光和红光，可以构建出一路参考信号，该参考信号的能量与系数相关，在系数Ra和Rv处，可以达到能量最大值。因此，遍历0～100％血氧饱和度的系数，寻找系数Ra和Rv，从而计算动脉血氧饱和度。该方法对运算量要求很高，其实现的硬件成本相对很高。
可以理解的是，基于Lambert-Bear定律可知，氧合血红蛋白(HbO2)和去氧血红蛋白(HB)在不同的光波段下吸收特性不同，如图1所示，为现有技术中氧合血红蛋白和还原血红蛋白吸收光谱曲线图；利用该吸收特性，可以进一步评估血氧饱和度参数。
如图2所示，为现有技术中血氧测量工作原理图；利用发光管发射两路光(例 如：红光和近红外光)透射机体组织，并利用接收管接收透射机体组织的光信号，从而获得血液搏动成分AC交流量(例如：动脉血)和非搏动成分DC直流量(例如：静脉血、肌肉、骨骼、皮肤等)。分别利用红光和近红外光的AC交流量和DC直流量，即可获得与动脉血氧饱和度(SaO2)的映射曲线比值，即R系数表(如公式1所示方法)。再基于查表法即可获得动脉血氧饱和度(SaO2)的无限逼近值脉搏血氧饱和度(SpO2)。
Figure PCTCN2016085777-appb-000001
其中，ACRed＝检测到红光的波动量，即交流量；DCRed＝检测到红光的最大透射量，即直流量；ACIred＝检测到的红外光的交流量；DCIred＝检测到的红外光直流量。
然而如何能够准确识别采集信号中AC交流成分和DC交流成分，获得准确的R值，却不是那么容易的事情。现有技术中存在两种技术：时域技术和频域技术。
其中，时域技术具有响应速度快、相位信息清晰等特点。由于时域信号是有用信号和噪声信号的混合，生理带宽外的噪声可以通过高通/低通滤波器轻松滤除，而当出现生理带宽内的噪声时，由于噪声的变化性、先验知识的匮乏，时域方法几乎没有办法滤除这部分噪声。因此时域技术在抗运动性能上存在着天然的缺陷。
而频域技术理论上可以分离噪声和有用信号的频带，进而达到区分识别真实信号的目的。按照数字信处理的定义，任何一个波形都是由多个正弦波构成，生理参数的脉搏波也不例外。因此，当生理脉搏波信号转换为频域信号时，该生理参数就会出现基倍频特征。当干扰出现时，干扰频谱和生理参数的基倍频谱混叠在一起，很难识别到真实频谱是哪一个。因此，尽管频域技术存在着先天性的优势，但是想在干扰状态下准确计算血氧相关参数，也非常棘手。
由Parseval定理可知，信号时域的总能量等于频域的总能量。因此，公式1也适用于频域信号，其中AC交流量对应于每个频率点的能量变化。在理想条件下，经过规一化后，脉搏波的红光及红外光频谱如图3所示，主频段及各个倍频段的比值可以看作是此频段下对应的能量比值。该能量比值的意义与公式1中对应的R值的意义完全一致，也即与血氧饱和度唯一对应。在不考虑噪声的理想状态下，各个频段的比值一致，都与当前的血氧饱和度唯一对应，理论上在任何频段都可以得到血氧饱和度参数。
同理，根据信号的构成理论，可知图3中基频峰所在频率即为脉率值。常规检测基频峰频率的方法可以通过识别基倍频峰筛选获得，所依据的理论是基频 峰与谐波峰存在能量和频率的比例关系。通过这种比例关系即可识别到生理脉率值。
综上所述，通过检索频域信号中基倍频峰的位置信息和红光与红外光的能量比值，即可获得脉率参数和血氧参数。然而在干扰情况下，红光和红外光的频谱峰信息可能被噪声混淆、湮灭，基频峰频率信息因无法识别或错误识别而导致脉率计算异常；同时红光和红外光的能量比值也因噪声的混入而导致血氧计算偏差较大。如图4所示，给出了一种干扰下的频谱分布示意。从图4中可以看到，红外光和红光的频谱峰能量比值忽大忽小，基频峰几乎被湮没。这种情况下，利用现有的时频域技术正确的识别频谱峰并计算准确的血氧和脉率参数，几乎不可能实现。
发明内容
本发明实施例所要解决的技术问题在于提供一种生理参数的计算方法及相应的医疗设备，可以在弱灌性和运动状态下提升计算生理参数(如脉率值以及血氧参数)的准确度，且计算复杂度低，对运算资源需求低。
为了解决上述技术问题，本发明实施例提供了一种生理参数的计算方法，该方法包括如下步骤：
选择通过采样获得的红光和/或红外光的至少一路信号所对应的一段区间的时域信号，对所述区间的时域信号进行时频域转换，获得相应的频域信号；
在所述频域信号中，选择其中所有合理的频谱峰，计算所选择的合理的频谱峰的能量信息，并构成频谱峰能量比值序列；
根据所述频谱峰能量比值序列构建稳定系数，如果所述的稳定系数偏低，则利用频谱峰能量比值序列构建补偿系数；
使用所述的补偿系数补偿所述的时域信号和/或频域信号的至少一路，基于已补偿的时域信号和/或频域信号计算获得生理参数。
相应地，本发明的另一方面，还提供一种计算生理参数的医疗设备，其包括：
传感器，其包括至少一个发光管和至少一个接收管，所述发光管发射透射生理组织的至少两路不同波长的光信号，所述接收管接收透射生理组织的至少两路光信号，并转为电信号；
模数转换器，与所述传感器连接，将所述电信号转换为数字信号，该数字信号包含了生理组织的至少部分特征；
数字处理器，与所述模数转换器连接，所述数字处理器执行下述处理：
对一段区间的数字信号进行时频域转换，获得对应的频域信号；
在所述频域信号中，选择其中所有合理的频谱峰，计算所选择的合理的频谱峰的能量信息，并构成频谱峰能量比值序列；
根据所述频谱峰能量比值序列构建稳定系数，如果所述的稳定系数偏低，则利用频谱峰能量比值序列构建补偿系数；
使用所述的补偿系数补偿所述一段区间的数字信号和/或对应的频域信号，基于已补偿的数字信号和/或频域信号，计算获得生理参数。
相应地，本发明的另一方面，还提供一种生理参数的计算方法，该方法包括如下步骤：
选择通过采样获得的红光和/或红外光的至少一路信号所对应的一段区间的时域信号，对所述区间的时域信号进行时频域转换，获得至少一路频域信号；
在所述频域信号中，选择其中所有合理的频谱峰，计算所选择的合理的频谱峰的位置信息，并形成频谱峰位置序列；
根据所述的频谱峰位置序列，构建随时间变化的阵列图，随时间变化的每个位置点构建至少一个稳定因子，从而形成稳定因子阵列图；
基于稳定因子阵列图构建稳定系数，如果所述稳定系数偏低，则利用所述的稳定因子阵列图计算得到补偿系数；
使用所述的补偿系数补偿所述的时域信号和/或频域信号的至少一路，基于已补偿的时域信号和/或频域信号计算获得生理参数。
相应地，本发明的另一方面，还提供一种计算生理参数的医疗设备，其包括：
传感器，其包括至少一个发光管和至少一个接收管，所述发光管发射透射生理组织的至少两路不同波长的光信号，所述接收管接收透射生理组织的至少两路光信号，并转为电信号；
模数转换器，与所述传感器连接，将所述电信号转换为数字信号，该数字信号包含了生理组织的至少部分特征；
数字处理器，与所述模数转换器连接，所述数字处理器执行下述处理：
对一段区间的数字信号进行时频域转换，获得对应的频域信号；
选择其中所有合理的频谱峰，计算所选择的合理的频谱峰的位置信息，并形成频谱峰位置序列；
根据所述的频谱峰位置序列，构建随时间变化的阵列图，随时间变化的每个位置点构建至少一个稳定因子，从而形成稳定因子阵列图；
基于稳定因子阵列图构建稳定系数，如果所述稳定系数偏低，则利用所述的稳定因子阵列图计算得到补偿系数；
使用所述的补偿系数补偿所述的所述一段期间的数字信号和/或对应的频 域信号，基于已补偿的数字信号和/或频域信号，计算获得生理参数。
相应地，本发明的另一方面，还提供一种生理参数的计算方法，该方法包括如下步骤：
选择通过采样获得的红光和/或红外光的至少一路信号所对应的一段区间的时域信号，对所述区间的时域信号进行时频域转换，获得至少一路频域信号；
在所述频域信号中，选择其中所有合理的频谱峰，计算所选择的合理的频谱峰的能量和/或位置信息，并形成频谱峰能量比值序列和/或频谱峰位置序列；
根据所述频谱峰能量比值序列和/或频谱峰位置序列，构建稳定系数，如果稳定系数偏低，使用所述频谱峰能量比值序列和/或频谱峰位置序列构建补偿系数；
使用所述的补偿系数补偿所述的时域信号和/或频域信号的至少一路，基于已补偿的时域信号和/或频域信号计算获得生理参数。
相应地，本发明的另一方面，还提供一种计算生理参数的医疗设备，其包括：
传感器，其包括至少一个发光管和至少一个接收管，所述发光管发射透射生理组织的至少两路不同波长的光信号，所述接收管接收透射生理组织的至少两路光信号，并转为电信号；
模数转换器，与所述传感器连接，将所述电信号转换为数字信号，该数字信号包含了生理组织的至少部分特征；
数字处理器，与所述模数转换器连接，所述数字处理器执行下述处理：
对一段区间的数字信号进行时频域转换，获得对应的频域信号；
在所述频域信号中，选择其中所有合理的频谱峰，计算所选择的合理的频谱峰的能量和/或位置信息，并形成频谱峰能量比值序列和/或频谱峰位置序列；
根据所述频谱峰能量比值序列和/或频谱峰位置序列构建稳定系数，偏低如果稳定系数偏低偏低，使用所述频谱峰能量比值序列和/或频谱峰位置序列构建补偿系数；
使用所述的补偿系数补偿所述一段区间的数字信号和/或对应的频域信号，基于已补偿的数字信号和/或频域信号，计算获得生理参数。
实施本发明实施例，具有如下的有益效果：
首先，本发明实施例基于频域技术并结合时域技术，能够在弱灌注和运动状态下，大幅度提升了生理参数的计算准确性，为客户带来了卓越的临床性能体验，可极大地提高生理参数(如血氧参数)的应用和推广；
其次，实施本发明，通过将时频域信号的特点与人体生理参数的特征相结合，在出现干扰情况下，采用静脉氧补偿法以及频谱阵列图法，在弱灌性和运 动状态下提升计算生理参数(如脉率值以及血氧参数)的准确度。其中，静脉氧补偿法可以消除干扰引起的血氧测量偏差，无限逼近真实生理血氧值，极大的提供了干扰状态下血氧参数计算的准确性；频谱阵列图法方法可以消除干扰引起的脉率测量偏差，即使在长时间的干扰状态下，也能够准确的识别生理频谱信息，极大的提供了干扰状态下脉率参数计算的准确性；
另外，本发明实施例提供的方法，计算复杂度低，对运算资源需求低。
附图说明
为了更清楚地说明本发明实施例或现有技术中的技术方案，下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本发明的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动性的前提下，根据这些附图获得其他的附图仍属于本发明的范畴。
图1为现有技术中氧合血红蛋白和还原血红蛋白吸收光谱曲线图；
图2为现有技术中血氧测量工作原理图；
图3为不存在干扰情形时的一种频谱分布图；
图4为存在干扰情形时的一种频谱分布图；
图5为本发明提供的生理参数的计算方法一个实施例的主流程示意图；
图6为本发明提供的生理参数的计算方法另一个实施例的主流程示意图；
图7为本发明提供的生理参数的计算方法又一个实施例的主流程示意图；
图8为本发明提供的计算生理参数的医疗设备中所采用的数字处理器的一个实施例的结构示意图；
图9为图8中补偿系数构建单元的结构示意图；
图10为本发明提供的计算生理参数的医疗设备中所采用的数字处理器的另一个实施例的结构示意图；
图11为图10中补偿系数构建单元的结构示意图；
图12为本发明提供的计算生理参数的医疗设备中所采用的数字处理器的又一个实施例的结构示意图；
图13为图12中补偿处理单元的结构示意图；
图14是本发明一个实施例中涉及的一种静脉血搏动干扰动脉血搏动的示意图；
图15是本发明一个实施例中补偿系数的计算流程示意图；
图16是本发明一个实施例中在干扰状态下的频谱段血氧分布示意图；
图17本发明一个实施例中静脉氧补偿法框架流程示意图；
图18是本发明一个实施例中进行基倍频率筛选流程示意图；
图19示出了本发明一个实施例中频谱阵列图法的流程示意图。
图20示出了本发明一个实施例中同时采用静脉氧补偿法和频谱阵列图法的流程示意图；
图21为本发明提供的一种计算生理参数的医疗设备的一个实施例的结构示意图。
具体实施方式
为了使本发明的目的、技术方案及优点更加清楚明白，以下结合附图及实施例，对本发明进行进一步详细说明。应当理解，此处所描述的具体实施例仅仅用以解释本发明，并不用于限定本发明。
如图5所示，示出了本发明提供的一种生理参数计算准确度的方法的一个实施例的主流程示意图，在该实施例中，示出了静脉氧补偿法(Venous Oxygen Compensation,VOC)的补偿流程，具体地，该方法包括如下的步骤：
步骤S10，选择通过采样获得的红光和/或红外光的至少一路信号所对应的一段区间的时域信号，对所述区间的时域信号进行时频域转换，获得相应的频域信号；
步骤S12，在所述频域信号中，选择其中所有合理的频谱峰信息，计算所选择的合理的频谱峰的能量信息，并构成红外光和红光的频谱峰能量比值序列；其中，所述合理的频谱峰信息为满足频谱能量关系(如，否存在基倍频关系)、频谱幅度关系(幅度是否未低于一个定值)、频谱位置关系、频谱峰形态关系(如，频谱峰的形状对于频点是否左右对称)等因素中的至少一个；
步骤S14，根据所述频谱峰能量比值序列构建稳定系数，如果所述的稳定系数偏低，则利用频谱峰能量比值序列构建补偿系数；其中，所述的稳定系数为基于所述红光和红外光能量比值的偏差统计或基于血氧参数值的偏差统计构建而成，或者为一经验系数。经验系数根据生理状态以及信噪比特征抽象化形成，例如：识别为运动干扰时，生理血氧不可能过低；长时间运动下生理血氧会出现下降趋势；运动干扰剧烈时，当噪声成分远大于信号成分时，血氧趋近于80％，等等。结合血氧系统，即可形成数字化的经验系数值。具体地，将所述稳定系数与一设定阈值对比判断，以确定所述稳定系数是否偏低。在一个例子中，可以利用统计谱峰能量比值偏差序列或血氧偏差序列的均值、标准差、最大值、最小值信息，按照特定算法构建获得所述稳定系数；
在一个例子中，可以采用如下的方法来构建补偿系数：
选取生理参数(如血氧参数)均值作为补偿系数计算公式的分母输入源；选取均值+标准差或均值-标准差作为补偿系数计算公式的分子输入源；
将所述分子输入源和分母输入源代入下述补偿系数计算公式，计算得到相应的补偿系数：
Factorcompensation＝tabR(Numerator)/tabR(Denomnator)
其中，tabR(Numerator)为以所述分子输入源作为血氧值，在一预定的血氧与R曲线系数的映射表查询获得的R系数值，而tabR(Denomnator)为以所述分子输入源作为血氧值，在一预定的血氧与R曲线系数的映射表查询获得的R系数值；Factorcompensation为补偿系数。
步骤S16，使用所述的补偿系数补偿所述的时域信号和/或频域信号的至少一路，基于所述的已补偿的时域信号和/或频域信号计算获得生理参数所述生理参数为血氧参数、脉率参数、波形面积参数、灌注指数参数中的至少一个。
下述以生理参数为血氧饱和度参数为例，举例说明图5中的进一步的详细流程图。
首先在步骤S10中，选择通过采样获得的红光和/或红外光的至少一路信号所对应的一段区间的时域信号，对所述区间的时域信号进行时频域转换，获得相应的频域信号；
接着在步骤S12中，在所述频域信号中，选择其中所有合理的频谱峰信息，计算所选择的合理的频谱峰的能量信息，并构成红外光和红光的频谱峰能量比值序列；具体地；根据每一所述合理的频谱峰的血氧饱和度参数，统计分析血氧分布序列，获得多个特征信息；具体地，根据每一所述合理的频谱峰的血氧饱和度参数，至少计算获得所述血氧饱和度的均值vMean、标准差vStd、最大值vMax、最小值vMin；
然后在步骤S14中，根据所述频谱峰能量比值序列构建稳定系数，如果该稳定系数偏低，则利用频谱峰能量比值序列构建补偿系数。具体地，将前述步骤中获得的多个特征信息中，最大值与最小值的差值作为稳定系数，如果所述稳定系统小于或等于一个设定的第一阈值，则确定该稳定系数偏低，需要构建补偿系数；
具体地，在一些实施例中，构建补偿系数的过程可以通过如下方法获得：
如果此时标准差vStd小于一预设的第二阈值，则选取均值作为补偿系数计算公式的分母输入源；如果所述血氧序列中超过均值的血氧个数占序列总数的至少一半，则以最大值+标准差作为补偿系数计算公式的分子输入源，反之以最 大值–标准差作为补偿系数计算公式的分子输入源；
如果此时标准差大于一预设的第二阈值，则以预定方式获得重复因子，以所作为补偿系数计算公式的分子输入源，并以所述均值作为补偿系数计算公式的分母输入源；
将所述分子输入源和分母输入源代入下述补偿系数计算公式，计算得到相应的补偿系数：
Factorcompensation＝tabR(Numerator)/tabR(Denomnator)...................公式1
其中，tabR(Numerator)为以所述分子输入源作为血氧值，在一预定的血氧与R曲线系数的映射表查询获得的R系数值，而tabR(Denomnator)为以所述分子输入源作为血氧值，在一预定的血氧与R曲线系数的映射表查询获得的R系数值；Factorcompensation为补偿系数。
其中，所述以预定方式获得重复因子的步骤具体为：
统计血氧分布序列中，获得满足一定偏差值(如+2％)的最大血氧参数集合，计算所述最大血氧参数集合的均值，作为重复因子；或者
选取血氧历史趋势中的一定时间范围(如4s～8s)内的稳定段，计算所述稳定段中血氧参数集合的均值，作为重复因子；
最后在步骤S16中，使用所述补偿系数补偿所述红光和/或红外光至少一路所对应的时域信号，使用至少一路已补偿的时域信号进一步计算生理参数。所述生理参数为血氧参数、脉率参数、波形面积参数、灌注指数参数中的至少一个。具体地可以通过如下的两种方式：
一、根据所述补偿后的时域信号，根据时域算法，计算获得最终血氧饱和度参数和脉率值；或者
二、对所述补偿后的时域信号进行时频域变换，获得频域信号，选择频域信号中一个合理的频谱峰，并根据频域算法对所述合理的频谱峰进行计算，获得准确的最终血氧饱和度参数。
如图6所示，示出了本发明提供的一种生理参数计算准确度的方法的另一个实施例的主流程示意图，在该实施例中，示出了频谱阵列图法(Power Spectrum Array,PSA)补偿流程，具体地，该方法包括如下的步骤：
步骤S20，选择通过采样获得的红光和/或红外光的至少一路信号所对应的一段区间的时域信号，对所述区间的时域信号进行时频域转换，获得至少一路频域信号；
步骤S22，在所述频域信号中，选择其中所有合理的频谱峰信息，计算所选 择的合理的频谱峰的位置信息，并形成或频谱峰位置序列；其中，所述合理的频谱峰信息为满足频谱能量关系、频谱幅度关系、频谱位置关系、频谱峰形态关系中的至少一个或多个；
步骤S24，根据所述的频谱峰位置序列，构建稳定系数，如果所述稳定系数偏低，则利用所述的频谱峰位置序列构建随时间变化的阵列图，并通过该阵列图计算得到补偿系数；其中，所述稳定系数为基于频谱能量比值偏差、频谱血氧偏差、基倍频组状态、合理的频谱峰个数中至少一个构建而成，可以根据稳定因子的数量和/或权重值判断所述稳定系数是否偏低；
在一个例子中，根据频谱峰位置序列中的基倍频组状态来确定稳定系数；例如，在该频谱峰位置序列中不能识别出基倍频组对时，则认为稳定系数偏低；
具体地，用所述的频谱峰位置序列构建随时间变化的阵列图，并通过该阵列图计算得到补偿系数的步骤包括：
建立缓存以存储频谱峰的相关信息，所述频谱峰的相关信息包括：每一频谱峰位置信息PeakArray；每一频谱峰位置权重信息WeigtedArray；频谱峰存储的数量信息ArrayIndex；
步骤S161，筛选所述频域信号中预定数量的频谱峰，依次填充到所述缓存中，将每一频谱峰的位置信息填写入PeakArray，在WeigtedArray相应的位置权重信息加一，总长度ArrayIndex加一；其中，如果当前筛选的频谱峰与缓存中的一频谱峰相同，则直接将缓存中所述频谱峰位置权重信息WeigtedArray加一。
步骤S162，在所述缓存中，假定每一个频谱峰作为疑似基频峰，遍历所有其它频谱峰，判断其它频谱峰与所述疑似基频峰是否满足频率的倍数关系，如果满足，则消除所述倍频峰在缓存中的相关信息，并将所述倍频峰的位置权重系数按照倍频次数折算并累加到所述相应疑似基频峰的位置权重系数上；反之保留其它频谱峰的相关信息；
步骤S163，在最后保留的所述疑似频谱峰中，选择位置权重系数排在前列的三个频谱峰，确定其中一个作为最终的合理的频谱峰(在一些例子中可以认为是基频峰)，并根据所述确定的最终的合理的频谱峰获得脉率值，同样，可以理解的是，可以根据该最终的合理的频谱峰通过能量计算获得对应的血氧饱和度。具体地，该步骤进一步包括：
将所选择的每一频谱峰的位置权重系数分别与一预定的第三阈值进行比较；
如果位置权重系数大于所述的第三阈值的频谱峰个数为一个，则将所述频谱峰确定为最终的合理的频谱峰；
如果权重系数大于所述的第三阈值的频谱峰个数为至少两个，则进一步根据 生理特性，确定其中一个频谱峰作为最终的合理的频谱峰，并以后述最终的合理的频谱峰的信息作为补偿系数；
步骤S26，使用所述的补偿系数补偿所述的时域信号和/或频域信号的至少一路，基于所述的已补偿的时域信号和/或频域信号计算获得生理参数；其中，所述生理参数为血氧参数、脉率参数、波形面积参数、灌注指数参数中的至少一个；其中，在一个例子中，使用所述的补偿系数补偿所述的时域信号和/或频域信号的至少一路可以具体为：根据所述补偿系数构建一个带通滤波器，对所述的时域信号和/或频域信号的至少一路进行滤波处理，以实现补偿功能。
如图7所示，示出了本发明提供的一种生理参数计算准确度的方法的另一个实施例的主流程示意图，在该实施例中，综合了静脉氧补偿法(Venous Oxygen Compensation,VOC)和频谱阵列图法(Power Spectrum Array,PSA)补偿流程，具体地，该方法包括如下的步骤：
步骤S30，选择通过采样获得的红光和/或红外光的至少一路信号所对应的一段区间的时域信号，对所述区间的时域信号进行时频域转换，获得至少一路频域信号；
步骤S32，在所述频域信号中，选择其中所有合理的频谱峰信息，计算所选择的合理的频谱峰的能量和或位置信息，并形成频谱峰能量比值序列和频谱峰位置序列；其中，所述合理的频谱峰信息为满足频谱能量关系、频谱幅度关系、频谱位置关系、频谱峰形态关系中的至少一个；所述形成频谱峰能量比值序列和频谱峰位置序列的过程可分别参见前述对图5中步骤S12和对图6中步骤S22的描述；
步骤S34，根据所述频谱峰能量比值序列和频谱峰位置序列的至少一路的信号特征，构建稳定系数，同时判断述序列的至少一路的稳定系数是否偏低，如果稳定系数偏低，使用所述序列的至少一路的信号特征构建补偿系数；其中，所述稳定系数为基于频谱能量比值偏差、频谱血氧偏差、基倍频组状态、合理的频谱峰个数中至少一个构建而成；所述补偿系数为基于频谱能量序列计算得到的能量比值偏差系数或血氧偏差系数，或者为基于频谱位置序列随时间变化而统计得到的位置系数。
其中，根据频谱峰能量比值序构建稳定系数，并构建补偿系数的具体流程可参见前述对图5中步骤S14的描述；
其中，频谱峰位置序列构建稳定系数，并构建补偿系数的具体流程可参见前述对图6中步骤S24的描述；在此不进行详述。
步骤S36,利用所述补偿系数补偿所述的时域信号和/或频域信号；基于所述的 已补偿的时域信号和/或频域信号，进一步计算获得生理参数，所述生理参数为血氧参数、脉率参数、波形面积参数、灌注指数参数中的至少一个或多个。
具体地，该步骤包括下述至少一个：利用所述补偿系数，所选择的时域信号做增益处理，以实现补偿；或者
利用所述补偿系数，对所选择的频域信号进行滤波处理，以实现补偿。
相应地，本发明还提供了一种计算生理参数的医疗设备。
可以理解的是，本发明所公开的一种生理参数的计算方法可以功能模块化来实现，作为插件集成于其他辅助诊断设备(如监护设备，除颤仪，AED，自动复苏仪器，心电图机等)；也可以做成为单参数医疗器械，用于相关生理参数监测，生理参数包含血氧参数、脉率参数、波形面积参数、灌注指数参数中的至少一个。
如图21所示，示出了本发明公开的一种医疗设备的结构示意图，在该实施例中示出了一个单参数医疗器械；在该实施例中，该医疗设备包括：
至少一个血氧传感器，其用于对受试者的手指、额头、耳垂、脚趾、脚掌等部位的测量。该血氧传感器至少包含一个发光管和一个接收管，发光管发射用于透过受试者人体组织的至少两路不同波长的光信号，在一个例子中，该发光管可发射一路红光和一路红外光信号。接收管接收透过受试者人体组织的至少两路光信号，并转为至少两路电信号，在一个例子中，该两路电信号为红光和红外光信号；
模数转换器，与所述传感器连接，将所述电信号转换为数字信号，该数字信号包含了生理组织的至少部分特征；
数字处理器，与所述模数转换器连接，所述数字处理器执行下述处理：
对一段区间的数字信号进行时频域转换，获得对应的频域信号；
在所述频域信号中，选择其中所有合理的频谱峰，计算所选择的合理的频谱峰的能量信息，并构成频谱峰能量比值序列；
根据所述频谱峰能量比值序列构建稳定系数，如果所述的稳定系数偏低，则利用频谱峰能量比值序列构建补偿系数；
使用所述的补偿系数补偿所述一段区间的数字信号和/或对应的频域信号，基于已补偿的数字信号和/或频域信号计算获得生理参数。
进一步包括：与所述数字处理器连接，显示所述数字处理器计算获得的生理参数；和/或
通信单元，与所述数字处理器连接，发送所述数字处理器计算获得的生理参数。
其中，血氧传感器包括透射式和反射式两种，一般可佩戴于受试者手指、额头、耳垂、脚趾、脚掌等部位进行生理参数测量，包含血氧参数、脉率参数、波形面积参数、灌注指数参数中的至少一个。其中，显示单元可以是本地显示单元，也可以是通过有线/无线方式远程通讯到远程显示单元。显示单元通过字符、数值、波形、棒图、声音提示等为用户提供一个可知觉的参数呈现。
在本发明的其中一个实施例中，所述生理组织的至少部分特征为血液中含氧血红蛋白、去氧血红蛋白、铁氧血红蛋白、总血红蛋白或一氧化碳的光学特征中的一项或多项。
在本发明的其中一个实施例中，所述合理的频谱峰满足频谱能量关系、频谱幅度关系、频谱位置关系、频谱峰形态关系中的至少一个。
在本发明的其中一个实施例中，所述稳定系数为基于所述红光和红外光能量比值的偏差统计构建而成，或者为一经验系数。
在本发明的其中一个实施例中，所述稳定因子依据基倍频组特性、频谱峰形态合理性的至少一个，随时间变化调整其稳定性权重。
在本发明的其中一个实施例中，所述数字处理器利用频谱峰能量比值序列构建补偿系数的处理具体为：
选取所述均值经系数表转换为补偿系数计算公式的分母，选取(均值+标准差)或(均值-标准差)经系数表转换为补偿系数计算公式的分子，求取其比值得到补偿系数。
在本发明的其中一个实施例中，所述生理参数为血氧、脉率、波形面积、灌注指数等的至少一个。
为了更好地理解本实施例中数字处理器的功能及工作原理，下面将结合一个具体的例子进行说明。
如图8所示，示出了本发明提供的一种计算生理参数的医疗设备中所采用的数字处理器的一个实施例的结构示意图，该数字处理器采用静脉氧补偿法(Venous Oxygen Compensation,VOC)进行补偿，请一并结合图9所示，具体地，该数字处理器包括：
时频域转换单元11，用于选择通过采样获得的红光和/或红外光的至少一路信号所对应的一段区间的时域信号，对所述区间的时域信号进行时频域转换，获得相应的频域信号；
能量比值序列构建单元12，用于在所述频域信号中，选择其中所有合理的频谱峰信息，计算所选择的合理的频谱峰的能量信息，并构成红外光和红光的频谱峰能量比值序列；其中，所述合理的频谱峰信息为满足频谱能量关系、频谱 幅度关系、频谱位置关系、频谱峰形态关系中的至少一个；
补偿系数构建单元13，用于根据所述频谱峰能量比值序列构建稳定系数，如果所述的稳定系数偏低，则利用频谱峰能量比值序列构建补偿系数；所述稳定系数为基于所述红光和红外光能量比值的偏差统计或基于血氧参数值的偏差统计构建而成，或者为一经验系数；将所述稳定系数与一设定阈值对比判断，以确定所述稳定系数是否偏低。
补偿处理单元14，用于使用所述补偿系数补偿所述红光和/或红外光至少一路所对应的时域信号，使用至少一路已补偿的时域信号进一步计算生理参数。
所述补偿系数构建单元13进一步包括：
稳定系数构建子单元130，用于利用统计谱峰能量比值偏差序列或血氧偏差序列的均值、标准差、最大值、最小值信息，按照特定算法构建获得所述稳定系数；
判断单元131，用于判断稳定系数构建子单元130所构建的稳定系数是否偏低；
分子分母选取单元132，用于选取血氧参数均值作为补偿系数计算公式的分母输入源；以及选取均值+标准差或均值-标准差作为补偿系数计算公式的分子输入源；
计算单元134，用于将所述分子分母选取单元所选取的分子输入源和分母输入源代入下述补偿系数计算公式，计算得到相应的补偿系数：
Factorcompensation＝tabR(Numerator)/tabR(Denomnator)
其中，tabR(Numerator)为以所述分子输入源作为血氧值，在一预定的血氧与R曲线系数的映射表查询获得的R系数值，而tabR(Denomnator)为以所述分子输入源作为血氧值，在一预定的血氧与R曲线系数的映射表查询获得的R系数值；Factorcompensation为补偿系数。
更多的细节，可参照前述对图5的描述。
相应地，本发明还提供了一种计算生理参数的医疗设备，在该实施例中，该医疗设备包括：
传感器，其包括至少一个发光管和至少一个接收管，所述发光管发射透射生理组织的至少两路不同波长的光信号，所述接收管接收透射生理组织的至少两路光信号，并转为电信号；
模数转换器，与所述传感器连接，将所述电信号转换为数字信号，该数字信号包含了生理组织的至少部分特征；
数字处理器，与所述模数转换器连接，所述数字处理器执行下述处理：
对一段区间的数字信号进行时频域转换，获得对应的频域信号；
在所述频域信号中，选择其中所有合理的频谱峰，计算所选择的合理的频谱峰的位置信息，并形成频谱峰位置序列；
根据所述的频谱峰位置序列，构建随时间变化的阵列图，随时间变化的每个位置点构建至少一个稳定因子，从而形成稳定因子阵列图；
基于稳定因子阵列图构建稳定系数，如果所述稳定系数偏低，则利用所述的稳定因子阵列图计算得到补偿系数；
使用所述的补偿系数补偿所述的时域信号和/或频域信号的至少一路，基于已补偿的时域信号和/或频域信号计算获得生理参数。
进一步包括：显示单元，与所述数字处理器连接，显示所述数字处理器计算获得的生理参数；和/或
通信单元，与所述数字处理器连接，发送所述数字处理器计算获得的生理参数。
在本发明的其中一个实施例中，所述两路光信号，是红光和红外光；
在本发明的其中一个实施例中，所述生理组织的至少部分特征为血液中含氧血红蛋白、去氧血红蛋白、铁氧血红蛋白、总血红蛋白或一氧化碳的光学特征中的一项或多项。
在本发明的其中一个实施例中，所述合理的频谱峰为满足频谱能量关系、频谱幅度关系、频谱位置关系、频谱峰形态关系中的至少一个或多个。
在本发明的其中一个实施例中，所述稳定系数为基于频谱能量比值偏差、频谱血氧偏差、基倍频组状态、合理的频谱峰个数、频谱峰形态合理性中的至少一个构建而成。
在本发明的其中一个实施例中，所述稳定因子依据基倍频组特性、频谱峰形态合理性的至少一个，随时间变化调整其稳定性权重。
在本发明的其中一个实施例中，根据稳定因子的数量和/或权重值判断所述稳定系数是否稳定。
在本发明的其中一个实施例中，所述生理参数为血氧、脉率、波形面积、灌注指数等的至少一个。
更多细节，可以参照前述对图21的描述，同时为了更好地理解本实施例中数字处理器的功能及工作原理，下面将结合一个具体的例子进行说明。
如图10所示，示出了本发明提供的一种计算生理参数的医疗设备中所采用的数字处理器的另一个实施例的结构示意图，该数字处理器采用频谱阵列图法 (Power Spectrum Array,PSA)进行补偿，请一并结合图11所示，具体地，该数字处理器包括：
时频域转换单元11，用于选择通过采样获得的红光和/或红外光的至少一路信号所对应的一段区间的时域信号，对所述区间的时域信号进行时频域转换，获得至少一路频域信号；
频谱峰位置序列获得单元15，用于在所述频域信号中，选择其中所有合理的频谱峰信息，计算所选择的合理的频谱峰的位置信息，并形成频谱峰位置序列，所述合理的频谱峰信息为满足频谱能量关系、频谱幅度关系、频谱位置关系、频谱峰形态关系中的至少一个或多个；
补偿系数构建单元16，用于根据所述的频谱峰位置序列，构建稳定系数，如果所述稳定系数偏低，则利用所述的频谱峰位置序列构建随时间变化的阵列图，并通过该阵列图计算得到补偿系数，所述稳定系数为基于频谱能量比值偏差、频谱血氧偏差、基倍频组状态、合理的频谱峰个数中至少一个构建而成；
补偿处理单元17，用于使用所述的补偿系数补偿所述的时域信号和/或频域信号的至少一路，基于所述的已补偿的时域信号和/或频域信号计算获得生理参数，所述生理参数为血氧参数、脉率参数、波形面积参数、灌注指数参数中的至少一个。
其中，补偿系数构建单元16包括：
缓存单元160，用于建立缓存以存储频谱峰的相关信息，所述频谱峰的相关信息包括：每一频谱峰位置信息；每一频谱峰位置权重信息；频谱峰存储的数量信息；
频谱峰信息记录单元162，用于筛选所述频域信号中预定数量的频谱峰，依次填充到所述缓存中，将每一频谱峰的位置信息填写入，在相应的位置权重信息加一，总长度数量信息加一；
遍历处理单元164，用于在所述缓存中，假定每一个频谱峰作为疑似基频峰，遍历所有其它频谱峰，判断其它频谱峰与所述疑似基频峰是否满足频率的倍数关系，如果满足，则消除所述倍频峰在缓存中的相关信息，并将所述倍频峰的位置权重系数按照倍频次数折算并累加到所述相应疑似基频峰的位置权重系数上；反之保留其它频谱峰的相关信息；
补偿系数获得单元166，用于在最后保留的所述疑似频谱峰中，选择位置权重系数排在前列的三个频谱峰，确定其中一个作为最终的合理的频谱峰，以所述合理的频谱峰的信息作为补偿系数。
更多的细节，可参照前述对图6的描述。
相应地，本发明还提供了一种计算生理参数的医疗设备，在该实施例中，该医疗设备包括：
传感器，其包括至少一个发光管和至少一个接收管，所述发光管发射透射生理组织的至少两路不同波长的光信号，所述接收管接收透射生理组织的至少两路光信号，并转为电信号；
模数转换器，与所述传感器连接，将所述电信号转换为数字信号，该数字信号包含了生理组织的至少部分特征；
数字处理器，与所述模数转换器连接，所述数字处理器执行下述处理：
对一段区间的数字信号进行时频域转换，获得对应的频域信号；
在所述频域信号中，选择其中所有合理的频谱峰，计算所选择的合理的频谱峰的能量和/或位置信息，并形成频谱峰能量比值序列和/或频谱峰位置序列；
根据所述频谱峰能量比值序列和/或频谱峰位置序列构建稳定系数，偏低如果所述稳定系数偏低，使用所述频谱峰能量比值序列和/或频谱峰位置序列构建补偿系数；
使用所述的补偿系数补偿所述的时域信号和/或频域信号的至少一路，基于已补偿的时域信号和/或频域信号计算获得生理参数。
在本发明的其中一个实施例中，所述两路光信号，是红光和红外光。
在本发明的其中一个实施例中，所述生理组织的至少部分特征为血液中含氧血红蛋白、去氧血红蛋白、铁氧血红蛋白、总血红蛋白或一氧化碳的光学特征中的一项或多项。
在本发明的其中一个实施例中，所述合理的频谱峰为满足频谱能量关系、频谱幅度关系、频谱位置关系、频谱峰形态关系中的至少一个。
在本发明的其中一个实施例中，所述稳定系数为基于频谱能量比值偏差、频谱血氧偏差、基倍频组状态、合理的频谱峰个数中的至少一个构建而成。
在本发明的其中一个实施例中，所述补偿系数为基于频谱峰能量比值序列计算得到的能量比值偏差系数，或者为基于频谱峰位置序列随时间变化而统计得到的位置系数。
在本发明的其中一个实施例中，使用所述的补偿系数补偿所述的时域信号和/或频域信号的至少一路的处理具体为：
利用所述补偿系数，对所选择的时域信号做增益处理，以实现补偿；或者
利用所述补偿系数，对所选择的频域信号进行滤波处理，以实现补偿。
在本发明的其中一个实施例中，所述生理参数为血氧、脉率、波形面积、灌 注指数等的至少一个。
更多细节，可以参照前述对图21的描述，同时为了更好地理解本实施例中数字处理器的功能及工作原理，下面将结合一个具体的例子进行说明。
如图12所示，示出了本发明提供的一种计算生理参数的医疗设备中所采用的数字处理器的另一个实施例的结构示意图，该数字处理器结合了静脉氧补偿法(Venous Oxygen Compensation,VOC)以及频谱阵列图法(Power Spectrum Array,PSA)进行补偿，请一并结合图13所示，具体地，该数字处理器包括：
时频域转换单元11，用于选择通过采样获得的红光和/或红外光的至少一路信号所对应的一段区间的时域信号，对所述区间的时域信号进行时频域转换，获得至少一路频域信号；
相应序列构建单元18，用于在所述频域信号中，选择其中所有合理的频谱峰信息，计算所选择的合理的频谱峰的能量和/或位置信息，并形成频谱峰能量比值序列和/或频谱峰位置序列；所述合理的频谱峰信息为满足频谱能量关系、频谱幅度关系、频谱位置关系、频谱峰形态关系中的至少一个；可以理解的是，该相应序列构建单元18同时具有图8中能量比值序列构建单元13和图10中频谱峰位置序列获得单元15的功能，具体的细节可参照前述相应的描述；
补偿系数构建单元16，用于根据所述序列的至少一路的信号特征，构建稳定系数，同时判断述序列的至少一路的稳定系数是否偏低，如果稳定系数偏低，使用所述序列的至少一路的信号特征构建补偿系数；所述稳定系数为基于频谱能量比值偏差、频谱血氧偏差、基倍频组状态、合理的频谱峰个数中至少一个构建而成；补偿系数构建单元16可以基于频谱能量序列计算得到的能量比值偏差系数或血氧偏差系数，或者基于频谱位置序列随时间变化而统计得到的位置系数而构建所述补偿系数。可以理解的是，该补偿系数构建单元16同时具有图8中补偿系数构建单元16和图10中补偿系数构建单元16的功能，具体的细节可参照前述相应的描述；
补偿处理单元17，用于利用所述补偿系数补偿所述的时域信号和/或频域信号；基于所述的已补偿的时域信号和/或频域信号，进一步计算获得生理参数，其中，所述生理参数为血氧参数、脉率参数、波形面积参数、灌注指数参数中的至少一个。
其中，补偿处理单元17进一步包括：
时域补偿单元170，用于利用所述补偿系数，所选择的时域信号做增益处理，以实现补偿；或者
频域补偿单元172，用于利用所述补偿系数，对所选择的频域信号进行滤波 处理，以实现补偿。
更多的细节，可参照前述对图7的描述。
为了便于更好地理解本发明，下文中将结合实际的例子来进一步说明前文中涉及的静脉氧补偿法(Venous Oxygen Compensation,VOC)以及频谱阵列图法(Power Spectrum Array,PSA)的原理及实现过程。
一、对于静脉氧补偿法(VOC)
申请人认为，在非干扰状态下，静脉血液由于其生理特性流动相对比较缓慢，可以认为是DC直流量的一部分，静脉血氧饱和度不会对正常的血氧测量产生任何影响。在干扰状态下，静脉血液受到干扰的影响，产生了静脉搏动，该静脉搏动形成的AC交流量就会混入动脉血液搏动所形成的AC交流量中。按照前文的公式1可知，此时计算得到血氧饱和度必然是偏离真实值的；从生理角度，可以理解为动脉波氧饱和度中混入了静脉氧饱和度，导致最终的血氧饱和度偏离血气值。
如图14所示，给出了一种静脉血搏动干扰动脉血搏动的示意图。在无法获取随机干扰源特征的情况下，时域算法几乎没有任何办法准确计算血氧饱和度值；另外，想要获取随机干扰源特征是非常困难的。
在背景技术中提到，频域信号中的每个频谱段都与血氧饱和度唯一对应，理论上每个频谱段都可以计算血氧饱和度；同时由于静脉干扰的随机性，其叠加在不同频段上的干扰也存在轻重之分。本申请所提出的静脉氧补偿法(VOC)即基于这两种假设模型而构建。当每个频谱段之间的血氧值偏差较大时，说明静脉搏动已经干扰到真实血氧结果，通过统计每个频谱段的血氧值偏差变化，得到差异分布特征，并以此构建补偿系数，以消除血氧采样信号中的干扰，最后再以消除了干扰的采样信号重新计算，即可获得无限逼近真实生理血氧的参数结果。
故静脉氧补偿法(VOC)的流程如下：
首先，在完成时频域变换后，统计分析频域信号中每个频谱段血氧饱和度之间的差异变化，从而得到差异分布特征。即从频谱信号中获取满足条件的频谱峰，进而计算每个频谱峰的血氧饱和度，得到血氧饱和度序列。其中，满足条件的频谱峰是指在幅度、能量、宽度、形态等符合纳入统计分析的范围，判断准则是依据生理信号的特征以及数字信号处理的基本方法建立的。基于血氧饱和度序列，统计得到血氧的均值(vMean)、标准差(vStd)、最大值(vMax)、最小值(vMin)等。
其次，构建补偿系数计算公式，基于差异分布特征计算得到补偿系数。
如图13所示，给出了补偿系数的计算流程。
第一步，将vMax与vMin的差值与第一阈值(Threshold1)进行比较，并将vStd值与第二阈值(Threshold2)进行比较，判断选择正常模式还是干扰模式。该第一阈值和第二阈值的选择是根据生理参数特征以及血氧系统特点所得到的经验系数，例如在一个例子中，第一阈值(Threshold1)可以取15％的血氧偏差，第二阈值(Threshold2)可以取5％的血氧偏差。
第二步，如果是正常模式，说明静脉氧引起的波动相对比较小，可以选取均值和标准差作为补偿系数计算公式的输入源。其中，均值作为分母的输入源，如果血氧序列中超过均值的血氧个数占序列总数的至少一半，则以vMean+vStd作为分子，反之以vMean–vStd作为分子。
第三步，与第二步是选择关系，如果是异常模式，需要引入重复因子。重复因子的来源有两种：1)统计血氧序列中满足一定数值(如，±2％)偏差的最大血氧集合，取该集合的均值作为重复因子，其中，此处的±2％为经验系数，可根据实际变化调整；2)选取血氧历史趋势的稳定段，例如4s～8s的稳定趋势，计算该集合的均值作为重复因子，同样该4s～8s时间段也为经验系数，可根据实际变化调整。将该重复因子作为分子输入源，将均值作为分母输入源。
第四步，将分子和分母参数输入下述公式2，计算得到补偿系数。计算公式如下所示，其中tabR是血氧与R曲线系数的映射表(如前文所述)，通过输入的血氧值即可反查得到对应的R系数值。补偿系数就是分子血氧的R系数值与分母血氧的R系数值的比值。
Factorcompensation＝tabR(Numerator)/tabR(Denominator)......................公式2
Factorcompensation是补偿系数，tabR(Numerator)就是将分子作为输入源后代入血氧与R曲线系数的映射表后查得的R系数值，tabR(Denominator)就是将分母作为输入源后代入血氧与R曲线系数的映射表后查得的R系数值。
其中，下述将结合一个实际的例子来说明如何在静脉氧补偿法(VOC)中如何计算获得补偿系数。如图16所示，给出了一种干扰状态下的频谱段血氧分布示意图，假设频谱信号在生理带宽0.3Hz～5Hz中有A、B、C、D四段频谱峰，并假设该频谱信号受到静脉氧随机搏动的影响，每段频谱峰通过红外光和红光比值换算得到的血氧饱和度值分别为96％、85％、90％、87％，该计算出的四个血氧饱和度值之间存在偏差。通过统计偏差分布特征，分别可以得到：vMean＝88.3％，vStd＝4.8％，vMax＝96％，vMin＝85％。判断vMax与vMin的偏差小于 第一阈值(例如，15％)，且vStd小于第二阈值(例如，5％)，故选择正常模式。此时统计大于均值血氧的个数为2，占总血氧序列的50％，故选择分子为vMean+vStd＝93.1％，分母为vMean＝88.3％。查询血氧与R曲线系数的映射表，分别获得分子与分母对应的R值，代入公式2，即可计算得到修正因子约为1.172。
再次，利用该补偿系数补偿时域信号(频域信号是基于此段信号变换得到)因干扰而受到的损失。即用于频域变换的时域信号乘以补偿系数，得到补偿信号。在本申请的一个例子中，仅针对红光一路信号做补偿，作为示例。但在实际应用中也可以分化补偿系数，实现补偿每一路信号。
最后，利用完成补偿的时域信号再次变换到频域信号，再利用频域信号计算准确的血氧等参数。本发明实施例是基于频域算法给出的示例，实际应用中也可以省略频域方法，基于完成补偿的时域信号，利用时域算法获取准确的血氧等参数。
如图17所示，给出了静脉氧补偿法(VOC)的框架流程示意图。其整个过程为：选择时域信号一段区间，进行时频域转换，将时域信号转换到频域信号。利用生理信号特征以及数字信号处理的基本知识(例如：生理脉搏波频率范围、基倍频原理、形态特征等)，选择合理的频谱峰并计算该频谱峰的血氧饱和度参数。按照前述步骤统计分析血氧分布系列，得到一系列特征信息，再根据这些特征信息判断血氧饱和度是否存在偏差(或者小的偏差)。如果没有，则输出血氧参数结果；反之，根据这些特征信息计算补偿系数，并补偿到时域信号中。最后再利用补偿后的时域信号重新做时频域转换，并基于新的频域信号选择合理的频谱峰并计算输出最终血氧饱和度参数。
从中可以看出，静脉氧补偿法(VOC)能够消除干扰引起的血氧测量偏差，无限逼近真实生理血氧值，极大的提供了干扰状态下血氧参数计算的准确性。
二、频谱阵列图法(PSA)
如背景技术中所述，当干扰存在并严重扰乱频谱信号时，从频谱上很难识别出生理频谱信息。例如在图4示出的频谱图中，只依据基倍频原理的能量和频率关系是没有办法识别到基频峰信息的。如何在干扰状态下识别到生理频谱峰？本申请基于噪声随机分布的特性，提出了一种假设理论。任何类型的干扰，例如：弱灌注、肢体擦碰、手指晃动等，在血氧采样信号中，均表现为一种随机干扰成分，干扰的强度与测量端的运动强度正相关。考虑到生理特征状态，绝大多数的干扰都呈现为随机白噪声分布；少数较为规律的干扰(例如：帕金森症等)，由于其震动幅度相对较低，震动频率相对较高，而对血氧采样信号没 有实质性的影响，即不影响血氧参数的测量。尽管干扰信号扰乱了频谱信号，使得生理频谱无法识别出来，但无论干扰信号怎么变化，生理频谱峰的特征始终存在某一个频点上并且在一定时间内不会发生变化或变化相对缓慢(生理特征)；与此同时，干扰噪声是随机分布的，随着时间的变化，频谱特征也随之而发生变化。将每一时刻的频谱按照时间顺序依次排列，并纵向观察频谱信号的特征，就能够发现，绝大部分的频谱峰幅度和位置都在发生变化时，总存在至少一个频谱峰的位置是相对稳定不发生偏移的，这个频谱峰就是生理频谱峰(基频峰)。如果能够采用一种识别算法找出该频谱峰，也就意味找到了正确的脉率值。这就是频谱阵列图法的核心思想。
在本申请中，频谱阵列图法(PSA)的一般流程如下：
首先，建立缓存存储频谱峰的相关信息。例如：频谱峰位置信息，也即所在频率点(PeakArray)；频谱峰位置权重信息(WeigtedArray)；频谱峰存储的数量信息(ArrayIndex)，其中位置信息和权重信息共用数量信息。一般而言，考察的频谱峰数量不能太多，过多会增加算法识别的复杂性和运算量，例如，可以默认选取20作为上限，超出不进行分析。可以理解的是，在实际应用中也可根据系统的需求，选取合适的分析数量。
其次，筛选频域信号中合理的频谱峰，填充到上述缓存中。筛选的准则是根据频谱峰的能量、幅度、形态、位置等信息综合而得到的。记录信息时，将频谱峰的位置信息填写入PeakArray，在WeigtedArray相应的位置权重加一，总长度ArrayIndex加一。以此类推，将一次计算识别到的所有频谱峰添加入缓存中。如果是≧2次记录信息，需要考虑待添加的频谱峰是否已经存于缓存中，如果存在，则WeigtedArray相应位置权重递增，反之按常规方式增加新的频谱峰。假设每一次计算间隔2S，缓存需要累积一定时间的频谱信息，才可以启动生理频谱峰的识别。该时间需要根据系统的实际需求设定，例如可以设定为10次计算，即20S的时间趋势。满足10次信息存储，则启动相关的趋势分析，同时在完成趋势分析之后，消减最早存储的频谱信息。
再次，根据缓存的信息识别真实生理频谱峰。根据基倍频原理，消除PeakArray中的倍频峰，同时倍频峰的权重系数按照倍频次数折算并累加到基频峰的权重系数(例如：基频峰权重系数是3，2次倍频峰权重系数是2，取权重系数2除以倍频次数2，得到调整后的权重系数1，累积到基频峰权重系数，即基频峰权重系数调整为4)，同时消除该倍频峰存储在PeakArray和WeigtedArray中的信息，即初始化为0。详细流程如下图18所示，此时可以假定每一个频谱峰为基频峰，遍历所有其它频谱峰，判断其它频谱峰与该假定的基频峰是否满足频率的 倍数关系，如果满足则是其倍频峰，消除该倍频峰的相关信息，反之保留。根据基倍频能量衰减的特性以及生理特性，一般倍频超过4倍以上影响相对较小，实际应用时可根据系统的需要选择设置。按照类似操作，遍历每个频谱峰，完成倍频峰的相关信息的消除。最后选择权重系数相对较大的前三个峰作为下一步的输入信息。
最后，判断频谱峰的最大权重系数是否大于或等于设定的阈值(例如：该阈值可设定为8，表示连续16s都存在稳定频谱峰，实际应用时可根据系统的需要选择设置)。如果有2个或以上的频谱峰满足权重系数大于或等于设定阈值，则需要进一步根据生理特性做出判断。例如：在干扰状态下，生理脉率不可能过高也不可能偏低。选择满足设定阈值的最为合理的频谱峰，计算脉率参数并输出，同时根据状态适当优化其它频谱峰的权重系数。
如图19所示，给出了频谱阵列图法(PSA)的流程示意图。从中可以看出，首先建立频谱峰信息缓存，按照时间顺序，填充缓存。当缓存填充满时，按照基倍频原则简化存储的频谱峰，再选择简化的频谱峰中满足设定阈值的频谱峰。如果满足条件的频谱峰大于或等于2个，则根据生理特性转化的准则排除，并最终选择1个合理的频谱峰，基于此频谱峰并计算脉率参数。频谱阵列图法(PSA)能够消除干扰引起的脉率测量偏差，即使在长时间的干扰状态下，也能够准确的识别生理频谱信息，极大的提供了干扰状态下脉率参数计算的准确性。
综上所述，静脉氧修正法(VOC)可以识别静脉氧的干扰并补偿干扰引起的血氧偏差，频谱阵列图法(PSA)可以在连续干扰中准确识别脉率信息。这两种方法各自具备干扰的识别和处理能力，因此，当两种方法结合运用时，可以显著的提升对干扰的识别和抑制，从而获得血氧参数计算准确性和脉率参数准确性的大幅度提升。
如图20所示，示出了本发明提供的方法中两种方案综合应用的示例。在实际应用中可以根据系统需要自适应的增加、删减、调整这些步骤。其大致流程如下：选择指定的一段时域信号，进行时频转换，并基于转换后的频谱信号进行频谱峰的检索和识别，同时计算每个合理的频谱峰的血氧值。统计合理峰的血氧值得到相关的统计信息为VOC做准备，同时也记录本次计算所得到的所有频谱峰的信息为PSA做准备。如果识别到基倍频组对并且基倍频组对中每个频谱峰血氧值相对偏差较小，则直接选择基频峰计算脉率，并输出最终结果的血氧和脉率值。如果基倍频组对不满足，则进入PSA方法并计算得到合理的频谱峰。如果是基于已消除噪声/补偿静脉氧的时域信号，但是血氧值不稳定(例如：和历史趋势结果不匹配)，则只计算并输出脉率参数；反之如果血氧值稳定，则计 算脉率值并输出最终结果的血氧和脉率值；如果是第一次计算(时域信号没有消除噪声/补偿静脉氧)且血氧值不稳定，则进入VOC识别分支。在VOC识别分支增加了一步基于PSA方法得到的频谱峰的特定滤波器。结合补偿系数和特定滤波器，对时域信号进行补偿和噪声消除，然后重新做时频域变换并计算相关参数。
综上，实施本发明实施例提供的方法以及医疗设备，具有如下的有益效果：
首先，本发明实施例基于频域技术并结合时域技术，能够在弱灌注和运动状态下，大幅度提升了血氧和脉率参数的计算准确性，为客户带来了卓越的临床性能体验，可极大地提高血氧参数的应用和推广；
其次，实施本发明，通过将时频域信号的特点与人体生理参数的特征相结合，在出现干扰情况下，采用静脉氧补偿法以及频谱阵列图法，在弱灌性和运动状态下提升计算脉率值以及血氧参数的准确度。其中，静脉氧补偿法可以消除干扰引起的血氧测量偏差，无限逼近真实生理血氧值，极大的提供了干扰状态下血氧参数计算的准确性；频谱阵列图法方法可以消除干扰引起的脉率测量偏差，即使在长时间的干扰状态下，也能够准确的识别生理频谱信息，极大的提供了干扰状态下脉率参数计算的准确性；
另外，本发明实施例提供的方法，计算复杂度低，对运算资源需求低。
本领域普通技术人员可以理解实现上述实施例方法中的全部或部分步骤是可以通过程序来指令相关的硬件来完成，程序可以存储于一计算机可读取存储介质中，的存储介质，如ROM/RAM、磁盘、光盘等。
以上仅为本发明的较佳实施例而已，并不用以限制本发明，凡在本发明的精神和原则之内所作的任何修改、等同替换和改进等，均应包含在本发明的保护范围之内。

Claims (45)

  1. 一种生理参数的计算方法，其特征在于，该方法包括如下步骤：
    选择通过采样获得的红光和/或红外光的至少一路信号所对应的一段区间的时域信号，对所述区间的时域信号进行时频域转换，获得相应的频域信号；
    在所述频域信号中，选择其中所有合理的频谱峰，计算所选择的合理的频谱峰的能量信息，并构成频谱峰能量比值序列；
    根据所述频谱峰能量比值序列构建稳定系数，如果所述的稳定系数偏低，则利用频谱峰能量比值序列构建补偿系数；
    使用所述的补偿系数补偿所述的时域信号和/或频域信号的至少一路，基于已补偿的时域信号和/或频域信号计算获得生理参数。
  2. 如权利要求1所述的计算方法，其特征在于，所述合理的频谱峰满足频谱能量关系、频谱幅度关系、频谱位置关系、频谱峰形态关系中的至少一个。
  3. 如权利要求1所述的计算方法，其特征在于，所述稳定系数为基于所述红光和红外光能量比值的偏差统计构建而成，或者为一经验系数；
    如果所述稳定系数小于或等于一设定阈值，则确定所述稳定系数偏低。
  4. 如权利要求3所述的计算方法，其特征在于，根据所述频谱峰能量比值序列构建稳定系数的步骤，具体为：利用统计谱峰能量比值偏差序列的均值、标准差、最大值、最小值信息，按照预定算法构建获得所述稳定系数。
  5. 如权利要求4所述的计算方法，其特征在于，所述如果偏低，则利用频谱峰能量比值序列构建补偿系数的步骤具体为：
    选取所述均值经系数表转换为补偿系数计算公式的分母，选取(均值+标准差)或(均值-标准差)经系数表转换为补偿系数计算公式的分子，求取其比值得到补偿系数。
  6. 如权利要求4所述的计算方法，其特征在于，所述生理参数为血氧、脉率、波形面积、灌注指数的至少一个。
  7. 一种计算生理参数的医疗设备，其特征在于，所述医疗设备包括：
    传感器，其包括至少一个发光管和至少一个接收管，所述发光管发射透射生理组织的至少两路不同波长的光信号，所述接收管接收透射生理组织的至少两路光信号，并转为电信号；
    模数转换器，与所述传感器连接，将所述电信号转换为数字信号，该数字信号包含了生理组织的至少部分特征；
    数字处理器，与所述模数转换器连接，所述数字处理器执行下述处理：
    对一段区间的数字信号进行时频域转换，获得对应的频域信号；
    在所述频域信号中，选择其中所有合理的频谱峰，计算所选择的合理的频谱峰的能量信息，并构成频谱峰能量比值序列；
    根据所述频谱峰能量比值序列构建稳定系数，如果所述的稳定系数偏低，则利用频谱峰能量比值序列构建补偿系数；
    使用所述的补偿系数补偿所述一段区间的数字信号和/或对应的频域信号，基于已补偿的数字信号和/或频域信号，计算获得生理参数。
  8. 如权利要求7所述的医疗设备，其特征在于，所述两路光信号是红光和红外光。
  9. 如权利要求7所述的医疗设备，其特征在于，所述生理组织的至少部分特征为血液中含氧血红蛋白、去氧血红蛋白、铁氧血红蛋白、总血红蛋白或一氧化碳的光学特征中的一项或多项。
  10. 如权利要求7所述的医疗设备，其特征在于，还包括：
    显示单元，与所述数字处理器连接，显示所述数字处理器计算获得的生理参数；和/或
    通信单元，与所述数字处理器连接，发送所述数字处理器计算获得的生理参数。
  11. 如权利要求7所述的医疗设备，其特征在于，所述合理的频谱峰满足频谱能量关系、频谱幅度关系、频谱位置关系、频谱峰形态关系中的至少一个。
  12. 如权利要求8所述的医疗设备，其特征在于，所述稳定系数为基于所述红光和红外光能量比值的偏差统计构建而成，或者为一经验系数。
  13. 如权利要求12所述的医疗设备，其特征在于，所述数字处理器根据所述频谱峰能量比值序列构建稳定系数的处理具体为：利用统计谱峰能量比值偏差序列的均值、标准差、最大值、最小值信息，按照预定算法构建获得所述稳定系数。
  14. 如权利要求13所述的医疗设备，其特征在于，所述数字处理器利用频谱峰能量比值序列构建补偿系数的处理具体为：
    选取所述均值经系数表转换为补偿系数计算公式的分母，选取(均值+标准差)或(均值-标准差)经系数表转换为补偿系数计算公式的分子，求取其比值得到补偿系数。
  15. 如权利要求7所述的医疗设备，其特征在于，所述生理参数为血氧、脉率、波形面积、灌注指数等的至少一个。
  16. 一种生理参数的计算方法，其特征在于，该方法包括如下步骤：
    选择通过采样获得的红光和/或红外光的至少一路信号所对应的一段区间的时域信号的，对所述区间的时域信号进行时频域转换，获得至少一路频域信号；
    在所述频域信号中，选择其中所有合理的频谱峰，计算所选择的合理的频谱峰的位置信息，并形成频谱峰位置序列；
    根据所述的频谱峰位置序列，构建随时间变化的阵列图，随时间变化的每个位置点构建至少一个稳定因子，从而形成稳定因子阵列图；
    基于稳定因子阵列图构建稳定系数，如果所述稳定系数偏低，则利用所述的稳定因子阵列图计算得到补偿系数；
    使用所述的补偿系数补偿所述的时域信号和/或频域信号的至少一路，基于已补偿的时域信号和/或频域信号计算获得生理参数。
  17. 如权利要求16所述的计算方法，其特征在于，所述合理的频谱峰为满足频谱能量关系、频谱幅度关系、频谱位置关系、频谱峰形态关系中的至少一个或多个。
  18. 如权利要求16所述的计算方法，其特征在于，所述稳定系数为基于频谱能量比值偏差、频谱血氧偏差、基倍频组状态、合理的频谱峰个数、频谱峰形态合理性中的至少一个构建而成。
  19. 如权利要求16所述的计算方法，其特征在于，所述稳定因子依据基倍频组特性、频谱峰形态合理性的至少一个，随时间变化调整其稳定性权重。
  20. 如权利要求19所述的计算方法，其特征在于，根据稳定因子的数量和/或权重值判断所述稳定系数是否偏低。
  21. 如权利要求16所述的计算方法，其特征在于，补偿系数选取稳定因子中权重最大的至少一个频谱峰，结合频率特性、频谱峰形态特征的至少一个，计算得到补偿系数。
  22. 如权利要求16所述的计算方法，其特征在于，所述生理参数为血氧、脉率、波形面积、灌注指数等的至少一个。
  23. 一种计算生理参数的医疗设备，其包括：
    传感器，其包括至少一个发光管和至少一个接收管，所述发光管发射透射生理组织的至少两路不同波长的光信号，所述接收管接收透射生理组织的至少两路光信号，并转为电信号；
    模数转换器，与所述传感器连接，将所述电信号转换为数字信号，该数字信号包含了生理组织的至少部分特征；
    数字处理器，与所述模数转换器连接，所述数字处理器执行下述处理：
    对一段区间的数字信号进行时频域转换，获得对应的频域信号；
    选择其中所有合理的频谱峰，计算所选择的合理的频谱峰的位置信息，并形成频谱峰位置序列；
    根据所述的频谱峰位置序列，构建随时间变化的阵列图，随时间变化的每个位置点构建至少一个稳定因子，从而形成稳定因子阵列图；
    基于稳定因子阵列图构建稳定系数，如果所述稳定系数偏低，则利用所述的稳定因子阵列图计算得到补偿系数；
    使用所述的补偿系数补偿所述一段期间的数字信号和/或对应的频域信号，基于已补偿的数字信号和/或频域信号，计算获得生理参数。
  24. 如权利要求23所述的医疗设备，其特征在于，所述两路光信号，是红光和红外光；
  25. 如权利要求23所述的医疗设备，其特征在于，所述生理组织的至少部分特征为血液中含氧血红蛋白、去氧血红蛋白、铁氧血红蛋白、总血红蛋白或一氧化碳的光学特征中的一项或多项。
  26. 如权利要求23所述的医疗设备，其特征在于，还包括：
    显示单元，与所述数字处理器连接，显示所述数字处理器计算获得的生理参数；和/或
    通信单元，与所述数字处理器连接，发送所述数字处理器计算获得的生理参数。
  27. 如权利要求23所述的医疗设备，其特征在于，所述合理的频谱峰为满足频谱能量关系、频谱幅度关系、频谱位置关系、频谱峰形态关系中的至少一个或多个。
  28. 如权利要求23所述的医疗设备，其特征在于，所述稳定系数为基于频谱能量比值偏差、频谱血氧偏差、基倍频组状态、合理的频谱峰个数、频谱峰形态合理性中的至少一个构建而成。
  29. 如权利要求23所述的医疗设备，其特征在于，所述稳定因子依据基倍频组特性、频谱峰形态合理性的至少一个，随时间变化调整其稳定性权重。
  30. 如权利要求29所述的医疗设备，其特征在于，根据稳定因子的数量和/或权重值判断所述稳定系数是否偏低。
  31. 如权利要求23所述的医疗设备，其特征在于，所述生理参数为血氧、脉率、波形面积、灌注指数等的至少一个。
  32. 一种生理参数的计算方法，其特征在于，该方法包括如下步骤：
    选择通过采样获得的红光和/或红外光的至少一路信号所对应的一段区间的时域信号，对所述区间的时域信号进行时频域转换，获得至少一路频域信号；
    在所述频域信号中，选择其中所有合理的频谱峰，计算所选择的合理的频谱峰的能量和/或位置信息，并形成频谱峰能量比值序列和/或频谱峰位置序列；
    根据所述频谱峰能量比值序列和/或频谱峰位置序列构建稳定系数，如果稳定系数偏低，使用所述频谱峰能量比值序列和/或频谱峰位置序列构建补偿系数；
    使用所述的补偿系数补偿所述的时域信号和/或频域信号的至少一路，基于已补偿的时域信号和/或频域信号计算获得生理参数。
  33. 如权利要求32所述的计算方法，其特征在于，所述合理的频谱峰为满足频谱能量关系、频谱幅度关系、频谱位置关系、频谱峰形态关系中的至少一个。
  34. 如权利要求32所述的计算方法，其特征在于，所述稳定系数为基于频谱能量比值偏差、频谱血氧偏差、基倍频组状态、合理的频谱峰个数中的至少一个构建而成。
  35. 如权利要求32所述的计算方法，其特征在于，所述补偿系数为基于频谱峰能量比值序列计算得到的能量比值偏差系数，或者为基于频谱峰位置序列随时间变化而统计得到的位置系数。
  36. 如权利要求32所述的计算方法，其特征在于，使用所述的补偿系数补偿所述的时域信号和/或频域信号的至少一路的步骤具体为：
    利用所述补偿系数，对所选择的时域信号做增益处理，以实现补偿；或者
    利用所述补偿系数，对所选择的频域信号进行滤波处理，以实现补偿。
  37. 如权利要求32所述的计算方法，其特征在于，所述生理参数为血氧、脉率、波形面积、灌注指数中的至少一个或多个。
  38. 一种计算生理参数的医疗设备，其特征在于，其包括：
    传感器，其包括至少一个发光管和至少一个接收管，所述发光管发射透射生理组织的至少两路不同波长的光信号，所述接收管接收透射生理组织的至少两路光信号，并转为电信号；
    模数转换器，与所述传感器连接，将所述电信号转换为数字信号，该数字信号包含了生理组织的至少部分特征；
    数字处理器，与所述模数转换器连接，所述数字处理器执行下述处理：
    对一段区间的数字信号进行时频域转换，获得对应的频域信号；
    在所述频域信号中，选择其中所有合理的频谱峰，计算所选择的合理的频谱峰的能量和/或位置信息，并形成频谱峰能量比值序列和/或频谱峰位置序列；
    根据所述频谱峰能量比值序列和/或频谱峰位置序列构建稳定系数，如果 所述稳定系数偏低，使用所述频谱峰能量比值序列和/或频谱峰位置序列构建补偿系数；
    使用所述的补偿系数补偿所述一段区间的数字信号和/或对应的频域信号，基于已补偿的数字信号和/或频域信号，计算获得生理参数。
  39. 如权利要求38所述的医疗设备，其特征在于，所述两路光信号，是红光和红外光。
  40. 如权利要求38所述的医疗设备，其特征在于，所述生理组织的至少部分特征为血液中含氧血红蛋白、去氧血红蛋白、铁氧血红蛋白、总血红蛋白或一氧化碳的光学特征中的一项或多项。
  41. 如权利要求38所述的医疗设备，其特征在于，所述合理的频谱峰为满足频谱能量关系、频谱幅度关系、频谱位置关系、频谱峰形态关系中的至少一个。
  42. 如权利要求38所述的医疗设备，其特征在于，所述稳定系数为基于频谱能量比值偏差、频谱血氧偏差、基倍频组状态、合理的频谱峰个数中的至少一个构建而成。
  43. 如权利要求38所述的医疗设备，其特征在于，所述补偿系数为基于频谱峰能量比值序列计算得到的能量比值偏差系数，或者为基于频谱峰位置序列随时间变化而统计得到的位置系数。
  44. 如权利要求42所述的医疗设备，其特征在于，使用所述的补偿系数补偿所述的时域信号和/或频域信号的至少一路的处理具体为：
    利用所述补偿系数，对所选择的时域信号做增益处理，以实现补偿；或者
    利用所述补偿系数，对所选择的频域信号进行滤波处理，以实现补偿。
  45. 如权利要求38所述的医疗设备，其特征在于，所述生理参数为血氧、脉率、波形面积、灌注指数等的至少一个。
PCT/CN2016/085777 2016-06-15 2016-06-15 一种生理参数的计算方法及相应的医疗设备 WO2017214870A1 (zh)

Priority Applications (6)

Application Number Priority Date Filing Date Title
CN201680073670.3A CN108471961B (zh) 2016-06-15 2016-06-15 一种生理参数的计算方法及相应的医疗设备
PCT/CN2016/085777 WO2017214870A1 (zh) 2016-06-15 2016-06-15 一种生理参数的计算方法及相应的医疗设备
CN202110094270.3A CN112932473B (zh) 2016-06-15 2016-06-15 一种生理参数的计算方法及相应的医疗设备
EP16904987.1A EP3473171B1 (en) 2016-06-15 2016-06-15 Method for calculating physiological parameters and corresponding medical equipment
US16/203,068 US11154250B2 (en) 2016-06-15 2018-11-28 Methods and systems for calculating physiological parameters
US17/505,155 US11872060B2 (en) 2016-06-15 2021-10-19 Methods and systems for calculating physiological parameters

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2016/085777 WO2017214870A1 (zh) 2016-06-15 2016-06-15 一种生理参数的计算方法及相应的医疗设备

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US16/203,068 Continuation US11154250B2 (en) 2016-06-15 2018-11-28 Methods and systems for calculating physiological parameters

Publications (1)

Publication Number Publication Date
WO2017214870A1 true WO2017214870A1 (zh) 2017-12-21

Family

ID=60662905

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2016/085777 WO2017214870A1 (zh) 2016-06-15 2016-06-15 一种生理参数的计算方法及相应的医疗设备

Country Status (4)

Country Link
US (2) US11154250B2 (zh)
EP (1) EP3473171B1 (zh)
CN (2) CN108471961B (zh)
WO (1) WO2017214870A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115990008A (zh) * 2022-11-22 2023-04-21 森思泰克河北科技有限公司 心率变异性监测方法、装置、雷达及可读存储介质
CN117288129A (zh) * 2023-11-27 2023-12-26 承德华实机电设备制造有限责任公司 一种托盘盛装的辐照物料厚度检测方法

Families Citing this family (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP3473171B1 (en) * 2016-06-15 2024-09-18 Shenzhen Mindray Bio-Medical Electronics Co., Ltd. Method for calculating physiological parameters and corresponding medical equipment
CN110755089A (zh) * 2019-10-17 2020-02-07 浙江荷清柔性电子技术有限公司 血氧检测方法、血氧检测装置和终端设备
CN114422714B (zh) * 2020-10-28 2024-07-02 北京小米移动软件有限公司 闪频光源与无闪频光源之间的切换方法及切换装置
CN112587111B (zh) * 2020-12-01 2022-08-19 清华大学 一种生理信号采集方法及系统
CN112826460B (zh) * 2020-12-29 2022-04-26 武汉联影智融医疗科技有限公司 生理信号频率提取方法、装置、生理信号采集设备和介质
CN113288090B (zh) * 2021-05-06 2022-04-22 广东工业大学 血压预测方法、系统、设备及存储介质
CN113671366B (zh) * 2021-08-25 2024-01-23 西安西电开关电气有限公司 信号处理方法及其应用装置、存储介质
CN114366090B (zh) * 2022-01-13 2024-02-02 湖南龙罡智能科技有限公司 一种集成多种测量机制的血液成分检定方法
CN116491894B (zh) * 2022-11-09 2024-02-27 桂林电子科技大学 基于欧拉影像放大算法的帕金森病症识别方法

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1830386A (zh) * 2005-03-10 2006-09-13 深圳迈瑞生物医疗电子股份有限公司 低灌注下测量血氧的方法
CN101039617A (zh) * 2004-10-15 2007-09-19 普尔塞特拉瑟技术有限公司 用于生理脉冲测量的光学输入信号的运动消除
CN101103921A (zh) * 2007-08-14 2008-01-16 北京麦邦光电仪器有限公司 一种测量血氧饱和度的方法和装置
US20130172760A1 (en) * 2006-05-16 2013-07-04 The Research Foundation Of State University Of New York Photoplethysmography apparatus and method employing high resolution estimation of time-frequency spectra
CN103230267A (zh) * 2013-05-14 2013-08-07 北京理工大学 一种抗运动干扰的脉率提取方法
US20140343385A1 (en) * 2005-03-03 2014-11-20 Covidien Lp Method for enhancing pulse oximetry calculations in the presence of correlated artifacts

Family Cites Families (16)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6594512B2 (en) * 2000-11-21 2003-07-15 Siemens Medical Solutions Usa, Inc. Method and apparatus for estimating a physiological parameter from a physiological signal
US7020507B2 (en) * 2002-01-31 2006-03-28 Dolphin Medical, Inc. Separating motion from cardiac signals using second order derivative of the photo-plethysmogram and fast fourier transforms
EP1611847A1 (en) * 2004-06-28 2006-01-04 Datex-Ohmeda, Inc. Validating pulse oximetry signals in the potential presence of artifact
US7502123B2 (en) * 2007-02-05 2009-03-10 Palo Alto Research Center Incorporated Obtaining information from optical cavity output light
CN201104882Y (zh) * 2007-12-05 2008-08-27 沈阳东软医疗系统有限公司 一种血氧饱和度测量装置
US9757043B2 (en) * 2007-12-06 2017-09-12 Los Angeles Biomedical Research Institute At Harbor-Ucla Medical Center Method and system for detection of respiratory variation in plethysmographic oximetry
US20090247837A1 (en) * 2008-03-27 2009-10-01 Nellcor Puritan Bennett Llc System And Method For Diagnosing Sleep Apnea
WO2009133851A1 (ja) * 2008-04-30 2009-11-05 コニカミノルタセンシング株式会社 酸素飽和度測定装置
JP5562805B2 (ja) * 2010-11-05 2014-07-30 オータックス株式会社 脈拍数測定方法及び血中酸素飽和度測定方法
CN102512178B (zh) * 2011-12-23 2014-04-09 深圳市理邦精密仪器股份有限公司 一种血氧测量装置
CN104095640A (zh) * 2013-04-03 2014-10-15 达尔生技股份有限公司 血氧饱和度检测方法及装置
EP2859913B1 (en) * 2013-10-11 2020-08-19 Peking Union Medical College Hospital, Chinese Academy of Medical Sciences Pulse oximetry-based Cardio-Pulmonary Resuscitation (CPR) quality feedback systems and methods
CN105249939A (zh) * 2014-07-17 2016-01-20 原相科技股份有限公司 具有去噪功能的生理检测模组及其生理检测方法
CN104382569B (zh) * 2014-12-08 2017-04-12 天津工业大学 光纤传感智能服装及其心音参数的处理方法
CN105147301B (zh) * 2015-08-04 2017-09-05 成都云卫康医疗科技有限公司 血氧离散饱和度转换算法的快速实现方法
EP3473171B1 (en) * 2016-06-15 2024-09-18 Shenzhen Mindray Bio-Medical Electronics Co., Ltd. Method for calculating physiological parameters and corresponding medical equipment

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101039617A (zh) * 2004-10-15 2007-09-19 普尔塞特拉瑟技术有限公司 用于生理脉冲测量的光学输入信号的运动消除
US20140343385A1 (en) * 2005-03-03 2014-11-20 Covidien Lp Method for enhancing pulse oximetry calculations in the presence of correlated artifacts
CN1830386A (zh) * 2005-03-10 2006-09-13 深圳迈瑞生物医疗电子股份有限公司 低灌注下测量血氧的方法
US20130172760A1 (en) * 2006-05-16 2013-07-04 The Research Foundation Of State University Of New York Photoplethysmography apparatus and method employing high resolution estimation of time-frequency spectra
CN101103921A (zh) * 2007-08-14 2008-01-16 北京麦邦光电仪器有限公司 一种测量血氧饱和度的方法和装置
CN103230267A (zh) * 2013-05-14 2013-08-07 北京理工大学 一种抗运动干扰的脉率提取方法

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP3473171A4 *

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115990008A (zh) * 2022-11-22 2023-04-21 森思泰克河北科技有限公司 心率变异性监测方法、装置、雷达及可读存储介质
CN117288129A (zh) * 2023-11-27 2023-12-26 承德华实机电设备制造有限责任公司 一种托盘盛装的辐照物料厚度检测方法
CN117288129B (zh) * 2023-11-27 2024-02-02 承德华实机电设备制造有限责任公司 一种托盘盛装的辐照物料厚度检测方法

Also Published As

Publication number Publication date
CN108471961B (zh) 2021-02-09
US11872060B2 (en) 2024-01-16
EP3473171A1 (en) 2019-04-24
EP3473171B1 (en) 2024-09-18
CN108471961A (zh) 2018-08-31
CN112932473A (zh) 2021-06-11
US20190133535A1 (en) 2019-05-09
US11154250B2 (en) 2021-10-26
US20220031252A1 (en) 2022-02-03
CN112932473B (zh) 2024-07-26
EP3473171A4 (en) 2020-06-17

Similar Documents

Publication Publication Date Title
WO2017214870A1 (zh) 一种生理参数的计算方法及相应的医疗设备
US6805673B2 (en) Monitoring mayer wave effects based on a photoplethysmographic signal
US9936886B2 (en) Method for the estimation of the heart-rate and corresponding system
US6393311B1 (en) Method, apparatus and system for removing motion artifacts from measurements of bodily parameters
US7991448B2 (en) Method, apparatus, and system for removing motion artifacts from measurements of bodily parameters
US6702752B2 (en) Monitoring respiration based on plethysmographic heart rate signal
KR101210828B1 (ko) 다중 생체 신호 측정을 이용하여 손목혈압의 정확도를 향상시키는 방법 및 장치
CN1264474C (zh) 剔除反常数据方法和应用该方法的血液成分光谱分析系统
US9119597B2 (en) Systems and methods for determining respiration information from a photoplethysmograph
US8073516B2 (en) Separating motion from cardiac signals using second order derivative of the photo-plethysmogram and fast fourier transforms
US8880576B2 (en) Systems and methods for determining respiration information from a photoplethysmograph
CN103027690B (zh) 一种基于自相关建模法的低灌注血氧饱和度测量方法
US9693709B2 (en) Systems and methods for determining respiration information from a photoplethysmograph
EP1611847A1 (en) Validating pulse oximetry signals in the potential presence of artifact
Yang et al. Estimation and validation of arterial blood pressure using photoplethysmogram morphology features in conjunction with pulse arrival time in large open databases
US20130079657A1 (en) Systems and methods for determining respiration information from a photoplethysmograph
EP2757944A1 (en) Systems and methods for determining respiration information from a photoplethysmograph
CN111000544B (zh) 基于ppg波形的混合式连续血压测量模型构建方法及系统
CN101940476B (zh) 一种血氧饱和度检测方法及系统
He et al. Spectral data quality assessment based on variability analysis: application to noninvasive hemoglobin measurement by dynamic spectrum
CN116869499A (zh) 一种基于ppg及其多阶微分信号的血压连续测量方法
Zhang et al. Non-invasive blood glucose detection using NIR based on GA and SVR
RU2805810C1 (ru) Носимое устройство с функцией определения концентрации гемоглобина, способ и система для определения концентрации гемоглобина
US20240268720A1 (en) Wearable device with function of determining hemoglobin concentration, method and system for determining hemoglobin concentration
Gurzhin et al. On-line monitoring of functional patient status in chronomagnetotherapy complex “Multimag”

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 16904987

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 2016904987

Country of ref document: EP

Effective date: 20190115