JP3726825B2 - Heart rate signal correction method - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a heartbeat signal correction method performed using a measured reference value of a heartbeat signal.
[0002]
[Prior art]
In measuring the heartbeat signal and using this heartbeat information, there are many cases where only inaccurate information can be obtained due to electrical noise, subject movement, and the like. For this reason, it is necessary to check whether the measured heartbeat signal is free of noise, but in the past, it was measured using the upper and lower limits that are reasonable for human heartbeat intervals. It only checks the validity of the heart rate signal.
[0003]
Also, a correction that removes noise from the measured heartbeat signal when a continuous heartbeat signal is required, even though it is clear that the measured heartbeat signal has, for example, noise. It is necessary to perform correction to compensate for the missing portion of the heartbeat signal. Conventionally, an average value of heartbeat intervals is used as a reference value for correction at this time.
[0004]
[Problems to be solved by the invention]
However, it is difficult to say that an accurate inspection is performed in a noisy system only by checking validity using the upper limit value and the lower limit value. Further, the correction of the heartbeat signal cannot be dealt with when noise is accidentally mixed in at the timing satisfying the condition of the human heartbeat interval. In addition, there is a possibility that inaccurate information is mixed in the creation of the reference value, and in the correction based on such a reference value, the corrected heart rate signal is also inaccurate. End up.
[0005]
The present invention has been made in view of the above points, and an object of the present invention is to appropriately correct a heartbeat signal using a reference value derived as an appropriate value as a reference value for heartbeat signal inspection. It is to provide a heartbeat signal correction method.
[0006]
[Means for Solving the Problems]
Therefore, the present invention is characterized in that an average value of heartbeat signal intervals when a heartbeat signal having a variation in interval within a predetermined percentage is continuously measured a predetermined number of times or more is used as a reference value for heartbeat signal inspection. are doing. Focusing on the variation in the continuous heartbeat interval, if the variation is not within a predetermined range, even if it is within a reasonable range for the human heartbeat interval, it is canceled and not reflected in the reference value.
[0007]
The heartbeat signal correction method according to the present invention has a first feature in canceling a signal having an interval equal to or less than a predetermined percentage of the reference value among the measured heartbeat signals. A second feature resides in that a signal for complementation is inserted between signals having intervals of a predetermined percentage or more of the reference value obtained by the derivation method.
[0008]
Of course, it is preferable to perform both of the above correction methods. In this case, signals having an interval equal to or less than a predetermined percentage of the reference value are canceled, and then signals having an interval equal to or greater than the predetermined value of the reference value are complemented in the meantime. It is assumed that a signal for is inserted.
[0009]
When inserting the signals, it is preferable to arrange them in such numbers and positions that the intervals generated by inserting the signals are closest to the reference value.
[0010]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, the present invention will be described in detail. FIG. 2 shows an example of a system configuration for implementing a reference value deriving method and a heartbeat signal correcting method. A physical change caused by a heartbeat detected from the subject 1 by the pickup 2 is converted into an electrical signal by the detection circuit 3, converted into a digital signal by the binarization circuit 4, and then input to the microcomputer (CPU) 5. . The derivation method and the correction method are implemented by a method suitable for the CPU 5, and specific work is performed in the CPU 5. A derivation method and a correction method will be described below assuming this system. It is assumed that the system has a system and control program configuration that recognizes the rising edge of a signal as a heartbeat.
[0011]
First, regarding the derivation of the reference value, as shown in FIG. 1, the data detected in step 202 is used as interval data in step 203. Step 204 then checks the validity of the interval data. This is to check whether the human heart rate interval is normal. As described in the conventional example, the upper limit value (H-limit msec) and lower limit value (L-limit msec) Check whether it fits in between. It should be noted that these two values are considered to change somewhat depending on the usage of the system and the state of the target person of the system.
[0012]
If it is determined as abnormal in step 204, the accumulated number of detections is canceled and the next interval data is awaited. If normal, it is checked in step 206 whether the variation in interval is within ± R%. If the same person is in the same state, the variation range of the heart rate interval must be much narrower than the normal range examined in step 204, and it is determined that the continuous signal having a large variation is noise. can do. Since the threshold value R here depends on the subject and the system, it cannot be determined unconditionally, but a value of 20 to 30% can be suitably used.
[0013]
If it is determined in step 206 that there is an abnormality, the same processing as in step 204 is performed, and if normal, 1 is added to the number of detected data (step 207). By repeating the above processing, if X consecutive normal interval data are detected, the average value of them is obtained as a reference value. As for X at this time, if the amount is too small, the reliability of the value is lowered, and if it is too large, it takes too much time to obtain the reference value, so about 3 to 10, preferably about 5 is appropriate.
[0014]
Next, a correction method for correcting the measured heartbeat signal using the reference value will be described. First, as shown in FIG. 3, regarding false alarms such as when noise enters between the original heartbeat signals, it is determined that there is an abnormality when the measured heartbeat signal interval data is smaller than A% of the reference value. It is assumed that the rising edge obtained this time is canceled and not detected. Also, although it is determined that the previous rising edge is normal as the interval data, there is a possibility of a slight forward / backward shift, so cancel the previous rising edge and set the last rising edge as the reliable last edge. Process.
[0015]
Next, as shown in FIG. 4, when the original heartbeat signal is missing, if the measured heartbeat signal interval data is greater than B% of the reference value, as shown in FIG. Assuming that a signal to be detected during the edge is not detected, the following signal complement is performed.
[0016]
FIG. 5 shows a complement routine. In step 502, the interval data that is considered to contain a plurality of undetected signals in between is divided by the reference value. The obtained answer is converted to an integer by an appropriate method such as rounding off in steps 503 and 504, and then 1 is subtracted. The value obtained in this way is used as the number of signals to be complemented, and the number of complement edges (heartbeat signals) are arranged in the interval data. In this arrangement, these are set to be equally spaced.
[0017]
As the values of the thresholds A and B in the correction, values of A = 80% to 60% and B = 125% to 150% can be preferably used. Since it depends on the system to be implemented, it cannot be decided unconditionally. For example, in a system in which the heartbeat interval is predicted to gradually increase (for example, a system that induces sleep), the value of B should be set larger and the value of A should be set smaller. By taking into account, correction with higher accuracy can be performed.
[0018]
Needless to say, the correction for the false alarm and the misreport is preferably performed in combination, and the flow in this case is shown in FIG. After correction for false alarms, correction for false alarms is performed. It is possible to cancel the interval data that is too short compared to the reference value, treat it as interval data that is too wide, and perform complementation, so that simplification and speeding up of the algorithm can be expected.
[0019]
The above system shows a real-time processing system as an example of the description, but the present invention is not limited to this and can be applied to a batch processing system.
[0020]
【The invention's effect】
As described above, since the derivation of the reference value for heartbeat signal inspection sets a variation range, a more accurate reference value can be derived in a short time. In the heartbeat signal correction method of the present invention, signals having an interval equal to or less than a predetermined percentage of the reference value are canceled, and signals having an interval equal to or greater than the predetermined percentage of the reference value are interpolated with signals for complementation derived from the reference value. In insertion, more accurate correction can be performed in order to use an accurate reference value. In particular, interval data that is too short compared to the reference value is not corrected, but is canceled and treated as interval data that is too wide.
[0021]
In addition, when complementing by inserting signals, signals that are closer to the actual heartbeat signal can be complemented by arranging them at the number and position where each interval that can be obtained by inserting the signal is closest to the reference value. it can.
[Brief description of the drawings]
FIG. 1 is a flowchart regarding reference value derivation in the present invention.
FIG. 2 is a system configuration diagram of the above.
FIG. 3 is a flowchart regarding correction for false alarms.
FIG. 4 is a flowchart relating to correction for missing information.
FIG. 5 is a flowchart for complementary correction.
FIG. 6 is a flowchart regarding correction.
[Explanation of symbols]
1 Test subject 2 Pickup 3 Detection circuit 4 Binarization circuit 5 CPU

Claims (4)

The average value of the heartbeat signal intervals when the heartbeat signal whose interval variation is within a predetermined percentage is continuously measured a predetermined number of times or more is used as a reference value for heartbeat signal inspection, and this reference value among the measured heartbeat signals A method for correcting a heartbeat signal, wherein a signal having an interval equal to or less than a predetermined percentage is canceled. The average value of the heartbeat signal interval when the heartbeat signal whose interval variation is within a predetermined percentage is continuously measured a predetermined number of times or more is used as a reference value for heartbeat signal inspection. A heartbeat signal correction method, wherein a signal for complementation is inserted between signals at intervals of a predetermined percentage or more. The average value of the heartbeat signal interval when the heartbeat signal whose interval variation is within a predetermined percentage is continuously measured a predetermined number of times or more is used as a reference value for heartbeat signal inspection. A heart rate signal correction method comprising: canceling a signal having an interval of a predetermined percentage or less; 4. The heartbeat signal correcting method according to claim 2, wherein the number and position of each signal are such that each interval generated by inserting the signal is closest to the reference value.

Images