Biological information measurement system and biological information measurement method

Abstract

To provide a biological information measurement system capable of improving accuracy of QRS wave detection in electrocardiograms.SOLUTION: A biological information measurement system includes a detection unit for detecting a QRS candidate waveform from an electrocardiogram of a subject, and an estimation unit for estimating whether or not the QRS candidate waveform is a QRS waveform based on the QRS candidate waveform and information on a cardiac function of the subject.SELECTED DRAWING: Figure 11

Description

[Technical field]

[0001] The present invention relates to a biological information measuring system and a biological information measuring method. [Background technology]

[0002] 2. Description of the Related Art Vital information measuring systems such as vital sign monitors measure the heart rate of a patient to grasp the patient's condition, and if the heart rate falls outside the normal range, alert a medical professional by sounding an alarm or the like.

[0003] Generally, the heart rate is the number of QRS waves (heart beats) that occur in one minute. In order to measure the heart rate in a time shorter than one minute, the number of QRS waves measured in six seconds on an electrocardiogram is multiplied by 10, or the heart rate is calculated based on the RR interval, which is the interval between QRS waves. For this reason, it is important to accurately detect QRS waves when measuring the heart rate. In addition, when attempting to detect atrial fibrillation, in which the occurrence of QRS waves becomes irregular, it is important to accurately capture the timing of the appearance of QRS waves over a period of about 15 seconds to several minutes.

[0004] However, electrocardiograms can contain artifacts such as myoelectric noise. If noise is mixed into an electrocardiogram, the detection of QRS waves becomes inaccurate, causing the following problems:

[0005] If noise is mistakenly recognized as a QRS wave, the number of QRS waves will be measured higher than the actual number, or the RR interval will be measured shorter than the actual number. This will cause the heart rate to be measured higher than the actual number. As a result, the heart rate displayed on the patient monitor may not decrease even though the heart rate is actually decreasing, or the heart rate may increase even though the heart rate is not actually increasing. This may result in a false negative for the lower heart rate alarm or a false positive for the upper heart rate alarm.

[0006] If the QRS wave is mistakenly recognized as noise, the number of QRS waves will be measured lower than the actual number, or the RR interval will be measured longer than the actual number. This will cause the heart rate to be measured lower than the actual number. As a result, the heart rate displayed on the patient monitor may not increase even though the heart rate is actually increasing, or the heart rate displayed on the patient monitor may decrease even though the heart rate is not actually decreasing. This may result in a false positive for the lower heart rate alarm or a false negative for the upper heart rate alarm.

[0007] In addition, when it is determined that the signal quality of the electrocardiogram is low due to noise contamination, the electrocardiogram may be deemed unanalyzable, and the most recently measured heart rate may be displayed without calculating the heart rate. In this case, by deeming the electrocardiogram unanalyzable, analysis for atrial fibrillation detection, etc. is interrupted, which results in interruption of detection of atrial fibrillation, etc., which is undesirable.

[0008] The following prior art is disclosed in the following patent document: An electrocardiogram and a pulse wave are measured, and a QRS wave in the electrocardiogram is detected to determine whether or not the QRS wave is a wave due to a premature ventricular contraction. Then, the occurrence of a premature supraventricular contraction is detected by determining whether or not the amplitude of the pulse wave corresponding to a QRS wave determined not to be a wave due to a premature ventricular contraction is smaller than a set value. [Prior art documents] [Patent documents]

[0009] [Patent Document 1] Japanese Patent Application Publication No. 11-47107 Summary of the Invention [Problem to be solved by the invention]

[0010] However, the above-mentioned prior art cannot improve the accuracy of detecting QRS waves in an electrocardiogram.

[0011] The present invention has been made to solve the above-mentioned problems, and an object of the present invention is to provide a biological information measuring system and a biological information measuring method that can realize high accuracy in detecting QRS waves in an electrocardiogram. [Means for solving the problem]

[0012] The above-mentioned object of the present invention is achieved by the following means.

[0013] (1) A biological information measuring system having a detection unit that detects a QRS candidate waveform from a subject's electrocardiogram, and an estimation unit that estimates whether the QRS candidate waveform is a QRS waveform based on the QRS candidate waveform and information regarding the subject's cardiac function.

[0014] (2) A method for measuring biological information comprising: a detection step of detecting a QRS candidate waveform from a subject's electrocardiogram; and an estimation step of estimating whether the QRS candidate waveform is a QRS waveform based on the QRS candidate waveform and information regarding the subject's cardiac function. Effect of the Invention

[0015] A QRS candidate waveform is detected from the subject's electrocardiogram, and whether or not the QRS candidate waveform is a QRS waveform is estimated based on the QRS candidate waveform and information on the subject's cardiac function. This makes it possible to improve the accuracy of detecting QRS waves in an electrocardiogram. [Brief description of the drawings]

[0016] [Figure 1] FIG. 1 is a schematic configuration diagram of a biological information measuring system. [Diagram 2] FIG. 2 is a block diagram of a hardware configuration of the bedside monitor. [Diagram 3] FIG. 4 is a block diagram showing the functions of a control unit. [Figure 4] FIG. 1 is an explanatory diagram for explaining a general method for classifying a QRS candidate waveform into either a QRS waveform or noise. [Diagram 5]FIG. 11 is an explanatory diagram for explaining an example of a method for determining the presence or absence of a pulse wave corresponding to a QRS candidate waveform. [Figure 6] FIG. 13 is a diagram showing a comparative example in which a QRS waveform is estimated based on a QRS candidate waveform. [Figure 7] FIG. 11 is a diagram showing another comparative example in which a QRS waveform is estimated based on a QRS candidate waveform. [Figure 8] FIG. 13 is a diagram showing an embodiment in which a QRS waveform is estimated based on a candidate QRS waveform. [Figure 9] FIG. 2 is a block diagram of a hardware configuration of a central monitor 200. [Figure 10] 4 is a flowchart showing the operation of the bedside monitor 100. [Figure 11] 11 is a subroutine flowchart of step S103 in the flowchart of FIG. 10. [Figure 12] 11 is a subroutine flowchart of step S103 in the flowchart of FIG. 10. [Figure 13] 11 is a subroutine flowchart of step S103 in the flowchart of FIG. 10. [Figure 14] FIG. 13 is a graph showing a method for estimating whether a candidate QRS waveform is a QRS waveform or noise based on a first probability and a second probability. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0017] Hereinafter, a biological information measuring system and a biological information measuring method according to an embodiment of the present invention will be described in detail with reference to the drawings. In the drawings, the same elements are given the same reference numerals and duplicated explanations will be omitted.

[0018] (First embodiment) (Biometric information measurement system 10) FIG. 1 is a schematic diagram of a biological information measuring system 10. The biological information measuring system 10 includes a bedside monitor 100 and a central monitor 200. One or more bedside monitors 100 may be provided. The bedside monitor 100 includes an electrocardiogram sensor 140 and a pulse wave sensor 150 (see FIG. 2), but the electrocardiogram sensor 140 and the pulse wave sensor 150 may be configured as devices separate from the bedside monitor 100. For example, the electrocardiogram sensor 140 and the pulse wave sensor 150 may be configured by an electrocardiogram measuring device, a non-invasive blood pressure monitor, or a transmitter capable of wirelessly communicating with the bedside monitor 100. The biological information measuring system 10 may be configured by one bedside monitor 100.

[0019] The central monitor 200 and the bedside monitor 100 are connected to each other so as to be able to communicate with each other via a wired or wireless network. The network may be, for example, a local area network (LAN) or a wide area network (WAN). For example, Ethernet (registered trademark), Wi-Fi (registered trademark), Bluetooth (registered trademark), or 5G may be used as the communication standard of the network.

[0020] (Bedside Monitor 100) 2 is a block diagram of the hardware configuration of the bedside monitor 100. The bedside monitor 100 includes a control unit 110, a storage unit 120, a communication unit 130, an electrocardiogram sensor 140, a pulse wave sensor 150, and an output unit 160. These components are connected to each other via a bus. The bedside monitor 100 is provided, for example, for each bed of a patient whose biological information is to be measured, or for each room of the patient.

[0021] The control unit 110 is configured with, for example, a CPU (Central Processing Unit) and a RAM (Random Access Memory), and controls each component of the bedside monitor 100 and performs various calculations. The functions of the control unit 110 will be described in detail later.

[0022] The storage unit 120 is configured, for example, by an SSD (Solid State Drive), and stores various data and various programs.

[0023] The communication unit 130 is an interface for communicatively connecting the bedside monitor 100 to other devices such as the central monitor 200. The communication unit 130 may be configured with, for example, an input terminal, an antenna, a front-end circuit, and the like. The communication unit 130 includes an alarm output unit 130a. The alarm output unit 130a transmits (outputs) each alarm to the central monitor 200. Specifically, the alarm output unit 130a transmits, for example, an alarm signal to notify the central monitor 200 of an abnormality in heart rate.

[0024] The electrocardiogram sensor 140 includes devices, elements, and electrodes for measuring an electrocardiogram, and detects an electrocardiogram.

[0025] Pulse wave sensor 150 includes an SpO2 probe, an invasive blood pressure monitor (including a catheter), or a non-invasive blood pressure monitor (including a cuff), and detects a pulse wave. Pulse wave sensor 150 can measure an SpO2 waveform or a blood pressure waveform, and detect a pulse wave based on these waveforms. Pulse wave sensor 150 can be replaced with a heart sound sensor.

[0026] The function of the control unit 110 will now be described.

[0027] 3 is a block diagram showing the functions of the control unit 110. By executing a program, the control unit 110 functions as a QRS candidate waveform detection unit 111, an estimation unit 112, and a calculation unit 113. The QRS candidate waveform detection unit 111 constitutes a detection unit.

[0028] The QRS candidate waveform detection unit 111 acquires an electrocardiogram of a subject, whose heart rate is to be measured, from the electrocardiogram sensor 140, and detects a QRS candidate waveform from the electrocardiogram. The QRS candidate waveform is a waveform that is a candidate for a QRS waveform, detected from an electrocardiogram detected by the electrocardiogram sensor 140. The QRS candidate waveform is detected by any known method, provided that the method detects only from electrocardiogram information. For example, the QRS candidate waveform is detected by filtering the electrocardiogram with a bandpass filter and then detecting the peak of the waveform. The electrocardiogram detected by the electrocardiogram sensor 140 may contain noise such as electromyographic noise. For this reason, noise (a noise waveform) may be detected as a QRS candidate waveform. In addition, a QRS waveform on which noise is superimposed may not be detected as a QRS candidate waveform due to the superimposition of noise.

[0029] The estimation unit 112 estimates whether the QRS candidate waveform is a QRS waveform based on the QRS candidate waveform and information on the subject's cardiac function. The information on the subject's cardiac function includes a pulse wave, heart sounds, etc. For ease of explanation, the following description will be given taking as an example a case where the information on the subject's cardiac function is a pulse wave.

[0030] The estimation unit 112 classifies the QRS candidate waveform into either a QRS waveform or noise based on the presence or absence of a pulse wave (information related to cardiac function) corresponding to the QRS candidate waveform, thereby estimating whether the QRS candidate waveform is a QRS waveform.

[0031] A common method for classifying a QRS candidate complex into either a QRS complex or noise is to determine whether the QRS candidate complex has the characteristics of noise, such as:

[0032] - The number of peaks around the QRS candidate waveform in the electrocardiogram is large (above a certain threshold) or small (below a certain threshold) The number of times that the waveform crosses the baseline around the QRS candidate waveform in the electrocardiogram is high (above a certain threshold) or low (below a certain threshold) - The waveform variation (standard deviation) around the QRS candidate waveform in the electrocardiogram is large (above a certain threshold) or small (below a certain threshold) The similarity of the QRS candidate waveform to the pre-stored QRS waveform template is low (below a certain threshold) Each of the above-mentioned predetermined threshold values ​​can be set to an appropriate value through experiments in terms of the detection accuracy of the QRS waveform.

[0033] FIG. 4 is an explanatory diagram for explaining a general method for classifying a QRS candidate waveform into either a QRS waveform or noise.

[0034] FIG. 4 shows an electrocardiogram, the detection location of the QRS candidate waveform in the electrocardiogram, the classification result of the QRS candidate waveform, and a pulse wave. The detection location of the QRS candidate waveform in the electrocardiogram and the classification result of the QRS candidate waveform are indicated by a beat label of "N" or "|". The beat label includes "N" indicating a QRS waveform and "|" indicating a noise (not a QRS waveform). In FIG. 4, the beat labels "|" with the round letters "2", "3", and "4" have the above-mentioned noise characteristics, and therefore can be classified as noise when using a general method of classifying QRS candidate waveforms into either a QRS waveform or noise. However, among the QRS candidate waveforms "2", "3", and "4" with the beat label "|", the QRS candidate waveforms "2" and "4" have corresponding pulse waves. Here, "corresponding" means that they are caused by the same heart beat. Therefore, it can be seen that the QRS candidate waveforms "2" and "4" with the beat label "|" are actually QRS waveforms and not noise, and have been erroneously classified as noise.

[0035] The estimation unit 112 classifies the QRS candidate waveform based on the presence or absence of a pulse wave corresponding to the QRS candidate waveform, thereby improving the accuracy of classifying the QRS candidate waveform into a QRS waveform or noise.

[0036] The estimation unit 112 can determine the presence or absence of a pulse wave corresponding to a QRS candidate waveform based on whether a pulse wave is detected within a predetermined delay time range after the detection of a QRS candidate waveform. That is, the estimation unit 112 determines that a pulse wave detected within a predetermined delay time range after the detection of a QRS candidate waveform is a pulse wave corresponding to the QRS candidate waveform. The predetermined delay time range can be set appropriately by experimentation from the viewpoint of the detection accuracy of the QRS waveform.

[0037] The estimation unit 112 may use a pulse wave that satisfies at least one of the following conditions as a pulse wave used to determine the presence or absence of a pulse wave corresponding to a QRS candidate waveform.

[0038] The amplitude and slope of the waveform are within a specified range from the average value of past pulse waves. - The waveform interval is constant (within a specified range) The waveform has a template waveform of the pulse wave or a similarity to adjacent waveforms in the time series that is equal to or greater than a specified value. The above-mentioned predetermined range and predetermined value can be set to appropriate values ​​through experiments from the viewpoint of the detection accuracy of the QRS waveform.

[0039] 5 is an explanatory diagram for explaining an example of a method for determining the presence or absence of a pulse wave corresponding to a QRS candidate waveform, in which an electrocardiogram and a pulse wave are shown on a common time axis.

[0040] The pulse wave transmitted through the blood in the blood vessels is detected later than the QRS waveform. The pulse wave (more specifically, the peak of the pulse wave) is detected with a delay (the length of the single-directional arrow in Figure 5) from the QRS waveform (more specifically, the R wave).

[0041] The range of the predetermined delay time may be a range having a predetermined width (the length of the bidirectional arrow in FIG. 5) centered on the average delay time of each past beat of the same subject (patient, etc.) while the heart rate is being measured. For example, the range of the predetermined delay time may be a range having a width of 40 ms before and after the average delay time of each past beat of the same subject while the heart rate is being measured. The range of the predetermined delay time may also be a range having a width of 10 to 100 ms before and after the average delay time of each past beat of the same subject.

[0042] When the information on the subject's cardiac function is heart sounds, it is preferable to use the first or second sound as the heart sound used to determine the presence or absence of a heart sound corresponding to the QRS candidate waveform. In a phonocardiogram, valve sounds generated with the contraction of the heart are measured. The first sound is a sound associated with the closure of the mitral valve and the tricuspid valve. The second sound is a sound associated with the closure of the aortic valve and the pulmonary valve. The estimation unit 112 can determine the presence or absence of a heart sound corresponding to the QRS candidate waveform based on whether the first or second sound is detected within a predetermined delay time range after the detection of the QRS candidate waveform. When reliability is important, the condition may be that both the first and second sounds are detected.

[0043] Comparative Example 1 Fig. 6 is a diagram showing a comparative example in which a QRS waveform is estimated based on a QRS candidate waveform. Fig. 6 shows an example in which all QRS candidate waveforms are estimated to be QRS waveforms as a result of classifying the QRS candidate waveforms using the general method described above. The detection points of the QRS candidate waveforms in the electrocardiogram and the classification results of the QRS candidate waveforms are shown with the beat label "N". Note that, although a pulse wave is not used in estimating the QRS waveform in this comparative example, a pulse wave is also shown in Fig. 6 for convenience of explanation.

[0044] Noise such as myoelectric noise is mixed into the electrocardiogram. For this reason, there is a possibility that noise that is not a QRS waveform is detected as a QRS candidate waveform. In this comparative example, all QRS candidate waveforms are estimated as QRS waveforms, and as a result, noise that is not a QRS waveform is erroneously estimated as a QRS waveform. Among the beat labels, beat labels that are noise but are erroneously estimated as a QRS waveform are shown surrounded by triangles. In FIG. 6, the correspondence between the QRS candidate waveform and the pulse wave is shown by a bidirectional arrow, and a beat label that is noise but is erroneously estimated as a QRS waveform is a beat label that is estimated as a QRS waveform even though there is no corresponding pulse wave.

[0045] If noise is erroneously estimated as a QRS waveform, as in this comparative example, the heart rate calculated based on the QRS waveform will be higher than the actual heart rate, reducing the accuracy of heart rate measurement.

[0046] Comparative Example 2 Fig. 7 is a diagram showing another comparative example in which a QRS waveform is estimated based on a QRS candidate waveform. Fig. 7 shows a case in which, as a result of classifying QRS candidate waveforms using the above-mentioned general method, some QRS candidate waveforms are estimated as noise even though they are actually QRS waveforms. Note that, although a pulse wave is not used in estimating the QRS waveform in this comparative example, a pulse wave is also shown in Fig. 7 for convenience of explanation.

[0047] Noise such as myoelectric noise is mixed into the electrocardiogram. Therefore, due to the influence of noise, there is a possibility that a QRS candidate waveform is estimated as noise even though it is actually a QRS waveform. In this comparative example, some QRS candidate waveforms are estimated as noise even though they are actually QRS waveforms. Among the beat labels, beat labels that are erroneously estimated as noise even though they are QRS waveforms are shown surrounded by triangles. In FIG. 7, the correspondence between the QRS candidate waveform and the pulse wave is shown by a bidirectional arrow, and a beat label that is erroneously estimated as noise even though it is a QRS waveform is a beat label that is estimated as noise even though there is a corresponding pulse wave.

[0048] If the QRS waveform is erroneously estimated as noise, as in this comparative example, the heart rate calculated based on the QRS waveform will be lower than the actual heart rate, reducing the accuracy of heart rate measurement.

[0049] (Example) FIG. 8 is a diagram showing an embodiment in which a QRS waveform is estimated based on a candidate QRS waveform.

[0050] As shown in Fig. 8, in this embodiment, a beat label of "N" is displayed as a classification result for a QRS candidate waveform with a corresponding pulse wave, and a beat label of "|" is displayed as a classification result for a QRS candidate waveform with no corresponding pulse wave. In other words, a QRS candidate waveform with a corresponding pulse wave is estimated to be a QRS waveform, and a QRS candidate waveform with no corresponding pulse wave is estimated to be noise. In Fig. 8, the correspondence between the QRS candidate waveform and the pulse wave is indicated by a bidirectional arrow.

[0051] As in the embodiment, by classifying QRS candidate waveforms based on the presence or absence of a pulse wave corresponding to the QRS candidate waveform, it is possible to suppress a decrease in the estimation accuracy of the QRS waveform due to noise.

[0052] The calculation unit 113 calculates the heart rate based on the estimated QRS waveform (QRS candidate waveform classified as a QRS waveform). Specifically, the control unit 110 calculates an RR interval from the estimated QRS waveform, and calculates the heart rate from the RR interval. The control unit 110 determines whether the heart rate is abnormal by determining whether the heart rate exceeds a predetermined normal range. The predetermined normal range is set to, for example, 60 to 100. When the control unit 110 determines that the heart rate is abnormal, it issues a heart rate alarm via the output unit 160. When the control unit 110 determines that the heart rate is abnormal, it may issue an alarm by transmitting a heart rate alarm signal to the central monitor 200 via the alarm output unit 130a together with an electrocardiogram, a pulse wave, and information identifying the subject.

[0053] The output unit 160 may be configured with a display and a speaker. The output unit 160 may notify the heart rate alarm by displaying the heart rate alarm in highlighted characters or the like on the display. The output unit 160 may notify the heart rate alarm by outputting the heart rate alarm as voice, warning sound, or the like from the speaker.

[0054] (Central Monitor 200) 9 is a block diagram of the hardware configuration of the central monitor 200. The central monitor 200 includes a control unit 210, a storage unit 220, a communication unit 230, and an output unit 240. These components are connected to each other via a bus. Among these components of the central monitor 200, the basic configurations of the components corresponding to the components of the bedside monitor 100 are similar to those of the bedside monitor 100, so duplicated explanations will be omitted.

[0055] The control unit 210 receives, via the communication unit 230, alarm signals for the electrocardiogram, pulse wave, and heart rate, as well as information for identifying the subject, from each bedside monitor 100.

[0056] The control unit 210 may issue a heart rate alarm by displaying the electrocardiogram, pulse wave, and heart rate alarm signals on the display in highlighted characters or the like for each subject identified from the information identifying the subject through the output unit 240. The output unit 160 may issue a heart rate alarm by outputting the heart rate alarm as a voice, a warning sound, or the like from a speaker.

[0057] The operation of the biological information measuring system 10 will be described.

[0058] 10 is a flowchart showing the operation of the bedside monitor 100. This flowchart can be executed by the control unit 110 of the bedside monitor 100 in accordance with a program.

[0059] The control unit 110 acquires an electrocardiogram and a pulse wave using the electrocardiogram sensor 140 and the pulse wave sensor 150, respectively (S101).

[0060] The control unit 110 detects a candidate QRS waveform from the electrocardiogram (S102).

[0061] The controller 110 classifies the QRS candidate waveform into either a QRS waveform or noise (S103). Details of this step will be described later with reference to the subroutine flow chart of FIG.

[0062] When the control unit 110 classifies the QRS candidate waveform not as a QRS waveform but as noise (S104: NO), the control unit 110 returns to step S101 and continues the process.

[0063] When the control unit 110 classifies the QRS waveform candidate as a QRS waveform (S104: YES), the control unit 110 stores information on the QRS waveform in memory (S105).

[0064] The control unit 110 calculates the RR interval from the stored QRS waveform information (S106), and calculates the heart rate from the RR interval (S107).

[0065] The control unit 110 determines whether the heart rate is an abnormal value (S108). If the control unit 110 determines that the heart rate is not an abnormal value (S108: NO), the process returns to step S101 and continues the process.

[0066] When the control unit 110 determines that the heart rate is an abnormal value (S108: YES), it issues a heart rate alarm via the output unit 160 or the like (S109).

[0067] FIG. 11 is a subroutine flowchart of step S103 in the flowchart of FIG.

[0068] The control unit 110 calculates the detection times (occurrence times) of the QRS candidate waveform and the pulse wave (S201).

[0069] The control unit 110 determines whether or not a pulse wave corresponding to the QRS candidate waveform is present, based on the detection times of the QRS candidate waveform and the pulse wave, depending on whether or not a pulse wave is detected within a predetermined delay time range after the detection of the QRS candidate waveform (S202).When the control unit 110 determines that a pulse wave corresponding to the QRS candidate waveform is not present (S202: NO), it estimates the QRS candidate waveform to be noise (S204).

[0070] When the control unit 110 determines that a pulse wave corresponding to the QRS candidate waveform is present (S202: YES), it estimates the QRS candidate waveform as a QRS waveform (S203).

[0071] (Variation 1) Hereinafter, a modified example of the operation of the biological information measuring system 10 will be described. In this modified example, a QRS candidate waveform is classified into a QRS waveform or noise by the general method described above based only on the QRS candidate waveform, and then the QRS candidate classified into noise is reclassified into a QRS waveform or noise depending on the presence or absence of a corresponding pulse wave.

[0072] FIG. 12 is a subroutine flowchart of step S103 in the flowchart of FIG.

[0073] The control unit 110 calculates the characteristics of the QRS candidate waveform and the pulse wave (S301). The characteristics of the QRS candidate waveform include the detection time (occurrence time) of the QRS candidate waveform, and a characteristic for determining the presence or absence of noise characteristics. The characteristics for determining the presence or absence of noise characteristics specifically include the general method of classifying the QRS candidate waveform into either a QRS waveform or noise, as described above. The characteristics of the pulse wave include the detection time (occurrence time) of the pulse wave.

[0074] The control unit 110 determines whether the QRS candidate waveform has noise characteristics based on the characteristics of the QRS candidate waveform (S302). If the control unit 110 determines that the QRS candidate waveform does not have noise characteristics (S302: NO), it estimates the QRS candidate waveform as a QRS waveform.

[0075] If the control unit 110 determines that the QRS candidate waveform has noise characteristics (S302: YES), it determines whether or not there is a pulse wave corresponding to the QRS candidate waveform based on the detection time of the QRS candidate waveform and the detection time of the pulse wave (S304).

[0076] When the control unit 110 determines that a pulse wave corresponding to the QRS candidate waveform exists (S304: YES), it estimates the QRS candidate waveform as a QRS waveform (S303).When the control unit 110 determines that a pulse wave corresponding to the QRS candidate waveform does not exist (S204: NO), it estimates the QRS candidate waveform as noise (S305).

[0077] (Variation 2) A description will be given of another modified example of the operation of the biological information measuring system 10. In this modified example, a QRS candidate waveform is classified into a QRS waveform or noise by the general method described above based only on the QRS candidate waveform, and then the QRS candidate classified into a QRS waveform is reclassified into a QRS waveform or noise depending on the presence or absence of a corresponding pulse wave.

[0078] FIG. 13 is a subroutine flowchart of step S103 in the flowchart of FIG.

[0079] The control unit 110 calculates the characteristics of the QRS candidate waveform and the pulse wave (S401). The characteristics of the QRS candidate waveform and the pulse wave are the same as those in the first modification example described above.

[0080] The control unit 110 determines whether the QRS candidate waveform has noise characteristics based on the characteristics of the QRS candidate waveform (S402). If the control unit 110 determines that the QRS candidate waveform has noise characteristics (S402: YES), it estimates the QRS candidate waveform as noise.

[0081] If the control unit 110 determines that the QRS candidate waveform does not have noise characteristics (S402: NO), it determines whether or not there is a pulse wave corresponding to the QRS candidate waveform based on the detection time of the QRS candidate waveform and the detection time of the pulse wave (S404).

[0082] When the control unit 110 determines that a pulse wave corresponding to the QRS candidate waveform exists (S404: YES), it estimates the QRS candidate waveform as a QRS waveform (S405).When the control unit 110 determines that a pulse wave corresponding to the QRS candidate waveform does not exist (S404: NO), it estimates the QRS candidate waveform as noise (S403).

[0083] Second embodiment A second embodiment will be described. This embodiment differs from the first embodiment in the following respects. In the first embodiment, whether or not a QRS candidate waveform is a QRS waveform is estimated based on the presence or absence of a pulse wave corresponding to the QRS candidate waveform. On the other hand, in this embodiment, whether or not a QRS candidate waveform is a QRS waveform is estimated based on the probability that the QRS candidate waveform is a QRS waveform based on the QRS candidate waveform and the probability that a pulse wave corresponding to the QRS candidate waveform exists. In other respects, this embodiment is similar to the first embodiment, so duplicated explanations will be omitted or simplified.

[0084] The estimation unit 112 calculates the probability that the QRS candidate waveform is a QRS waveform (hereinafter also referred to as "first probability") based on the QRS candidate waveform. The first probability can be calculated as a value obtained by multiplying the number of conditions satisfied by the QRS candidate waveform among the following conditions by 0.1, with an initial value of 0.5, for example, and adding the result to the initial value.

[0085] The number of peaks in the QRS candidate waveform is less than a certain threshold. The number of times the QRS candidate waveform crosses the baseline is less than a certain threshold. - The waveform variation (standard deviation) of the QRS candidate waveform is less than a specified threshold. The similarity of the QRS candidate waveform to the QRS template is less than a certain threshold. The above-mentioned predetermined thresholds can be set to appropriate values ​​through experiments in terms of the detection accuracy of the QRS waveform. The conditions for calculating the first accuracy may be a part of the above-mentioned conditions, or other conditions may be added.

[0086] The estimation unit 112 calculates the probability that a pulse wave corresponding to the QRS candidate waveform exists (hereinafter also referred to as "second probability") based on the QRS candidate waveform and the pulse wave. The second probability can be calculated, for example, by setting an initial value to 0.5 and adding a number obtained by multiplying the number of conditions satisfied by the QRS candidate waveform or pulse wave among the following conditions by 0.1 to the initial value.

[0087] The time from the detection (occurrence) of the QRS candidate to the detection (occurrence) of the pulse wave is within a specified range centered on the past average value. - The amplitude of the pulse wave is within a specified range centered on the past average value. The slope of the pulse wave (differential value) is within a specified range centered on the past average value The similarity of the pulse wave to the pulse wave template is equal to or greater than a specified threshold. The above-mentioned predetermined ranges and predetermined thresholds can be set to appropriate values ​​through experiments in terms of the detection accuracy of the QRS waveform. The conditions for calculating the second accuracy may be a part of the above-mentioned conditions, or other conditions may be added.

[0088] The estimation unit 112 estimates whether the QRS candidate waveform is a QRS waveform or noise based on the first and second probabilities.

[0089] FIG. 14 is a diagram showing a graph for estimating whether a QRS candidate waveform is a QRS waveform or noise based on the first probability and the second probability.

[0090] Whether the QRS candidate waveform is a QRS waveform or noise is estimated as follows, for example, using the graph in FIG.

[0091] In the graph of FIG. 14, the horizontal axis (x-coordinate) is the first accuracy, and the vertical axis (y-coordinate) is the second accuracy. In the graph, a dashed line is shown as a threshold connecting the point on the horizontal axis where the first accuracy is 1.0 and the point on the vertical axis where the second accuracy is 1.0. The area on the origin side of the graph from the threshold line is the noise area, and the area on the opposite side of the threshold line from the origin is the QRS waveform area. Whether the QRS candidate waveform is noise or a QRS waveform is estimated depending on whether the coordinate (x, y) = (first accuracy, second accuracy) of the combination of the first accuracy and second accuracy belongs to the noise area or the QRS waveform area.

[0092] The embodiment provides the following advantages.

[0093] A QRS candidate waveform is detected from the subject's electrocardiogram, and whether or not the QRS candidate waveform is a QRS waveform is estimated based on the QRS candidate waveform and information on the subject's cardiac function. This makes it possible to improve the accuracy of detecting QRS waves in an electrocardiogram.

[0094] Furthermore, the heart rate is calculated based on the waveform estimated as QRS, which improves the accuracy of measuring the heart rate based on an electrocardiogram.

[0095] Furthermore, the information regarding cardiac function is the pulse wave or heart sounds, which makes it possible to simply and effectively improve the accuracy of detecting QRS waves in an electrocardiogram.

[0096] In addition, whether or not the QRS candidate waveform is a QRS waveform is estimated based on the presence or absence of information about the cardiac function corresponding to the QRS candidate waveform, which makes it possible to more effectively improve the accuracy of detecting QRS waves in an electrocardiogram.

[0097] Furthermore, the presence or absence of information on cardiac function corresponding to the QRS candidate waveform is determined based on whether information on cardiac function is detected within a predetermined delay time range after the detection of the QRS candidate waveform, thereby further improving the accuracy of detecting QRS waves in an electrocardiogram.

[0098] In addition, the information on cardiac function used to determine the presence or absence of information on cardiac function corresponding to the candidate QRS waveform is a pulse wave or heart sound that satisfies at least one of the following: the amplitude and slope of the waveform are within a predetermined range from the past average value, the interval of the waveform is constant, and the similarity of the waveform to a template waveform or to adjacent waveforms in a time series is equal to or greater than a predetermined value. This makes it possible to further improve the accuracy of detection of QRS waves in an electrocardiogram.

[0099] In addition, the heart rate is calculated by calculating the RR interval of the waveform estimated as QRS, which makes it easier to measure the heart rate based on an electrocardiogram.

[0100] In addition, whether or not the QRS candidate waveform is a QRS waveform is estimated based on the probability that the QRS candidate waveform is a QRS waveform and the probability that a pulse wave corresponding to the QRS candidate waveform exists. This makes it possible to further improve the accuracy of detecting QRS waves in an electrocardiogram.

[0101] Although the embodiment of the present invention has been described in detail above, the present invention is not limited to the above-described embodiment.

[0102] For example, some or all of the functions and configurations of the bedside monitor 100 may be included in the functions and configurations of the central monitor 200. In this case, the bedside monitor 100 may be replaced by a transmitter, and an electrocardiogram and a pulse wave may be transmitted from each transmitter to the central monitor 200.

[0103] For example, some or all of the functions realized by the programs in the above-described embodiments may be realized by hardware such as circuits.

[0104] In addition, in the above-described flowchart, some steps may be omitted and other steps may be added. Furthermore, some of the steps may be executed simultaneously, and one step may be divided into multiple steps and executed. [Explanation of symbols]

[0105] 10 Biometric information measurement system, 100 Bedside monitors, 110 control section, 111 QRS candidate waveform detection unit, 112 Estimation Department, 113 Calculation unit, 120 storage section, 130 Communications Department, 140 electrocardiogram sensor, 150 Pulse wave sensor, 160 output section, 200 Central Monitor, 210 control section, 220 storage section, 230 Communications Department, 240 output section.

Claims

1. A detection unit for detecting a candidate QRS waveform from an electrocardiogram of a subject; an estimation unit that estimates whether the QRS candidate waveform is a QRS waveform based on the QRS candidate waveform and information on the subject's cardiac function; A vital sign measuring system having the above structure.

2. The biological information measuring system according to claim 1 , further comprising a calculation unit that calculates a heart rate based on the waveform estimated to be the QRS by the estimation unit.

3. The biological information measuring system according to claim 1 , wherein the information regarding the cardiac function is a pulse wave or a heart sound.

4. The biological information measuring system according to claim 1 , wherein the estimation unit estimates whether the QRS candidate waveform is the QRS waveform based on the presence or absence of information about the cardiac function corresponding to the QRS candidate waveform.

5. The biological information measuring system of claim 4, wherein the estimation unit determines whether or not there is information about the cardiac function corresponding to the QRS candidate waveform based on whether or not information about the cardiac function is detected within a predetermined delay time range after the QRS candidate waveform is detected.

6. The estimation unit estimates a pulse wave or a heart sound as the information on the cardiac function used to determine the presence or absence of information on the cardiac function corresponding to the QRS candidate waveform, The amplitude and slope of the waveform are within a specified range from the past average value, The intervals of the waveform are constant; and The similarity of the waveform to a template waveform or to an adjacent waveform in a time series is equal to or greater than a predetermined value; The biological information measuring system according to claim 4 , wherein the biological information measuring system satisfies at least one of the above.

7. The biological information measuring system according to claim 2 , wherein the calculation unit calculates a heart rate by calculating an RR interval of the waveform estimated to be the QRS by the estimation unit.

8. The biological information measuring system according to claim 1, wherein the estimation unit estimates whether the QRS candidate waveform is the QRS waveform based on a probability that the QRS candidate waveform is the QRS waveform and a probability that a pulse wave corresponding to the QRS candidate waveform exists.

9. A detection step of detecting a candidate QRS waveform from an electrocardiogram of a subject; an estimation step of estimating whether the QRS candidate waveform is a QRS waveform based on the QRS candidate waveform and information on the subject's cardiac function; The biological information measuring method includes the steps of:

10. The biological information measuring method according to claim 9 , further comprising a calculation step of calculating a heart rate based on the waveform estimated to be a QRS in the estimation step.

11. ３ The biological information measuring method according to claim 9 , wherein the information regarding the cardiac function is a pulse wave or a heart sound.

12. 10. The biological information measuring method according to claim 9, wherein in the estimating step, it is estimated whether the QRS candidate waveform is the QRS waveform based on the presence or absence of information on the cardiac function corresponding to the QRS candidate waveform.

13. The biological information measuring method of claim 12, wherein in the estimation step, the presence or absence of information about the cardiac function corresponding to the QRS candidate waveform is determined based on whether information about the cardiac function is detected within a predetermined delay time range after the QRS candidate waveform is detected.

14. In the estimation step, the information on the cardiac function used to determine the presence or absence of information on the cardiac function corresponding to the QRS candidate waveform is a pulse wave or a heart sound, The amplitude and slope of the waveform are within a specified range from the past average value, The intervals of the waveform are constant; and The similarity of the waveform to a template waveform or to an adjacent waveform in a time series is equal to or greater than a predetermined value; The biological information measuring method according to claim 12 , wherein a biological information measuring device that satisfies at least any one of the above conditions is used.

15. 11. The biological information measuring method according to claim 10, wherein in said calculating step, a heart rate is calculated by calculating an RR interval of the waveform estimated to be the QRS in said estimating step.

16. 10. The biological information measuring method according to claim 9, wherein in the estimation step, it is estimated whether the QRS candidate waveform is the QRS waveform based on a probability that the QRS candidate waveform is the QRS waveform and a probability that a pulse wave corresponding to the QRS candidate waveform exists.

Images