JPH0217034A - Electrocardiogram signal processor - Google Patents

Abstract

PURPOSE:To diagnose a two-time atrioventricular block Mobitz II type, a two- time atrioventicular block Wenckebach type and a sinoatrial block with high accuracy by easily separating the ham mixed in an electrocardiogram signal, the noise component of an electromyogram and drift or the like by developing a time series electrocardiogram data on a frequency region by FFT processing. CONSTITUTION:The electrocardiogram signal amplified by an electrocardiogram amplifier 1 is digitalized, for example, at a sampling interval of 16mec with resolving power of + or -10mV (12-bit) by an A/D converting part 2 and the position of QRS is detected by a QRS group detection part 3 as shown by a drawing. In an arhythmicity detection part 4, the appearing position of a QRS group (post QRS) immediately before a heartbeat falls out from the output of the QRS group detection part 3 and the position of the fallen-out QRS group (defect QRS) are detected to be informed to an FFT processing part 5. in a diagnostic part 7, the diagnosis of a two-time atrioventricular block Mobitz II type, a two-time atrioventricular block Wenckeback type and a sinoatrial block is performed on the basis of the amplitude characteristics and phase characteristics Ap(n), thetap(n), Ad(n), thetad(n) being the output of the FFT processing part and gamma0, alpha0, alpha1 being the output of a statistical processing part 6.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention relates to an electrocardiogram signal processing device that automatically analyzes electrocardiogram signals, and particularly relates to an electrocardiogram signal processing device that can accurately recognize segmentation points. .

[Prior art] An electrocardiogram reflects the electrical phenomena that occur in the heart, and depicts Korean waves such as P waves, QR8 complexes, T waves, and U waves during one cycle of heartbeat. ing. By deciphering the patterns of these various types of sacrificial waves, it is possible to obtain diagnostic information about the living body. Explain the diagnosis of sinoatrial block.

In these diagnoses, as shown in FIG. 4, it is necessary to accurately recognize the presence or absence of P waves and the location where they appear. That is,
As shown in Figure 4 (a), 2nd degree atrioventricular block Mobi
In the Lel type, the QRS complex is absent while the interval between the QRS complex and the P wave (hereinafter referred to as the PR interval) remains constant, and as shown in Figure 4 (b), in the Wenckebach type with 2nd degree atrioventricular block, the PR interval remains constant. As the sinoatrial block gradually lengthens, the QRS complex is missing, and as shown in FIG. 4(C), sinoatrial block takes the form of both the P wave and the QRS complex missing.

Thus, 2nd degree atrioventricular block Mobjtzll type, 2
Cortical ventricular block Wenckebach type and sinoatrial block have exactly the same heart rhythm, and as shown in Figure 4,
It is extremely difficult to recognize because it takes the form of one heartbeat dropping from the regular rhythm, and the QR8 group does not change.

Conventional techniques for making this type of diagnosis include detecting frog waves above a certain threshold on the original electrocardiogram data or on data that has been filtered to emphasize P waves, and detecting the amplitude and time width of the broad waves. The method used was to determine whether it was a P wave or not, and to make a diagnosis based on the location where the wave appeared.

[Problems to be Solved by the Invention] However, the above-mentioned conventional technology has the disadvantage that the recognition rate of P waves is low, and when noise components such as hum, electromyogram, and drift are mixed into the electrocardiogram signal, the electrocardiogram signal is significantly degraded. This has the disadvantage that the recognition rate of P waves decreases. Therefore, there is a problem in that diagnosis of second-degree atrioventricular block Mobitzll type, second-degree cortical ventricular block Wenckebach type, and sinoatrial block cannot be performed with high accuracy.

The present invention has been made in view of such problems, and includes:
It is possible to perform segmentation point recognition with high accuracy.
It is an object of the present invention to provide an electrocardiogram signal processing device that facilitates the diagnosis of degree atrioventricular block Mobitzl type I, bicortical ventricular block Wenckebach type, and sinoatrial block.

[Means for Solving the Problems] An electrocardiogram signal processing device according to the present invention includes an A/D conversion means that converts an electrocardiogram signal into a digital quantity, and detects the QR8 group in the electrocardiogram signal to detect an irregular state of heart rhythm. and means for detecting said A based on the output of said means.
an FFT processing unit which performs FFT processing on a signal in a range including the P wave in each heartbeat among the outputs of the /D conversion means to obtain frequency amplitude characteristics and frequency phase characteristics of the signal within this range; After statistical processing, 2nd degree atrioventricular block (04obitz) type If, 2nd degree atrioventricular block (W
enckebach) type and a diagnostic means for diagnosing sinoatrial block.

[Operation] In the present invention, only the heart rhythm arrhythmia portion is subjected to FFT processing, and the electrocardiogram signal of this portion is expanded into the frequency domain. Therefore, the hum mixed into the ECG signal,
Noise components such as electromyogram and drift can be easily separated,
Demarcation points near P waves can be recognized with high accuracy.

Thereby, diagnosis of second-degree atrioventricular block Mobitzl type, second-degree cortical ventricular block Wenckebach type, and sinoatrial block can be performed with high accuracy.

[Example] Next, an example of the present invention will be described with reference to the accompanying drawings.

FIG. 1 is a block diagram showing an embodiment of the present invention. 1st
As shown in the figure, in this embodiment, an electrocardiogram amplifier 1 inputs an electrocardiogram signal induced by an induction electrode (not shown) and amplifies it to a level suitable for subsequent processing, and converts the output of the electrocardiogram amplifier 1 into a digital signal. QRSR3 ratio detection that detects the QRS complex in the electrocardiogram waveform of one heartbeat shown in FIG. 2, and an irregular part of the heart rhythm shown in FIG. a rhythm irregularity detection unit 4 that detects an FFT processing range in order to detect
An FFT processing unit 5 that performs FFT (fast Fourier transform) processing on the FFT processing range in the figure; a statistical processing unit 6 that performs statistical processing of correlation coefficients and regression line coefficients on the output of the FFT processing unit 5; The diagnosis section 7 executes the processing flow shown in FIG. 3 based on the output of the diagnosis section 7 to perform diagnosis.

Next, the operation of the apparatus of this embodiment configured as described above will be explained. That is, the electrocardiogram signal amplified by the electrocardiogram amplifier 1 is processed by the A/D converter 2 at a sampling interval of 16"l, a second (msec), and a resolution S±1.
It is digitized by 0 mV (12 bits) and the QRS is detected by QRSR3 ratio detection as shown in Figure 2.
The position of is detected.

The rhythm irregularity detection unit 4 detects the QRS complex immediately before the heartbeat falls (hereinafter referred to as postQ) from the output of the QRSR3 ratio detection.
The appearance position of the QRS complex (hereinafter referred to as defecLQ RS) and the position of the QRS complex (hereinafter referred to as defecLQ RS) are detected, and the F
The FT processing unit 5 is notified.

The FFT processing section 5 performs FFT processing on the electrocardiogram data output from the A/D conversion section 2. This FFT processing section 5
The range X in which FFT processing is performed in
The end point is a point 60 milliseconds away from the position of 300
”! Perform FFT processing of 256 boins 1 to 1 second for the range of 1 second, and calculate the amplitude characteristics of postQ RS AP (n) (n
=0.1,2.・-255>, phase characteristic θp (n) of postQR8 (n=0.1.2-・255)
, the amplitude characteristics of defectQRs Ad(n)(n-0
, 1, 2, = -255) and the phase characteristics θd (n) (n = 0.1, 2. -255) of the defect Q RS.

In the statistical processing unit 6, pos which is the output of the FFT processing unit 5
tQ R,S phase characteristic θp (n) and defectQ
Difference between phase characteristics θd(n) of R8 θp (n) θd (
n) and correlation coefficient YO and regression line θp(n) regarding n
−θd(n)=α. Coefficient α of n+α1. , α1 are calculated. However, γ0. When determining α0α, it is necessary to use n in the P-wave frequency band (4 to 8 Hz).The upper and lower limits of n in this embodiment are determined as follows.

However, 62.5Hz is the sampling frequency.

In the diagnosis unit 7, Ap(n) which is the output of the FFT processing unit 5
, θp (n), Ad (n), θd(n) and γ0. which is the output of the statistical processing unit 6. α0. Based on α1, 2nd degree atrioventricular block Mobitzll type, 2nd degree atrioventricular block 1
FIG. 3 shows a flowchart of the diagnostic section 7 for diagnosing llenckebach type and sinoatrial block.

As shown in FIG. 4(c), in the sinoatrial block, both the P wave and the QRS complex are missing, so Ad(n) is set to a predetermined value for n=16 to 32 by processing SL and S2, as shown in FIG. 3(g). A diagnosis is made if the value is below the set value. Figure 4 (a
), in second-degree atrioventricular block MobitZII, the QRS complex and PR interval remain constant and the QRS complex is missing, so the treatments Sl , S3 , S4 , S
5, Ap (n) and A for n=16 to 32
d < n) are approximately equal, and θp(n) and θd(n)
The diagnosis is made when the two conditions are approximately equal. As shown in FIG. 4(b), the two-cortical chamber block Wenckebach type is P
Since the QRS complex is missing as the R interval gradually increases, the processes s, s, s4 . S6. S7゜S
8, a diagnosis is made when γ0 shows a value close to 1, α1 shows a value close to O, and Ap (n) and Ad (n) are approximately equal.

As described above, according to this embodiment, the QRS complex detection section 3 and the rhythm irregularity detection section 4 perform FFT processing on the vicinity of the P wave in each heartbeat, so analysis can be performed using data free of noise components. I can do it. Therefore, after recognizing the presence or absence of P waves and their appearance position, if the heart rhythm is exactly the same,
2nd degree atrioventricular block MobiLz where one heartbeat drops from the regular rhythm and the QR8 complex does not change
Type I, Wenckebach type 2 cortical chamber block, and sinoatrial block can be diagnosed with high accuracy.

[Effects of the Invention] As explained above, the present invention expands time-series electrocardiogram data into the frequency vA range by FFT processing, thereby achieving
Since it is possible to easily separate the noise components such as hum, electromyogram, and drift mixed in the electrocardiogram signal, it is possible to easily separate the noise components such as hum, electromyogram, and drift that are mixed in the electrocardiogram signal.
This method has the advantage that diagnosis of Ebach type and sinoatrial block can be performed with high accuracy.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing an embodiment of the electrocardiogram signal processing device of the present invention, FIG. 2 is a schematic diagram of an electrocardiogram waveform cited to show the FFT processing range in the same embodiment, and FIG.
A flowchart diagram showing the configuration of the diagnostic section 7 in the figure, and FIG. 4 is a schematic diagram of an electrocardiogram waveform cited in the conventional technique. 1; Electrocardiogram amplifier; 2; A/D conversion unit; 3; QRS complex ratio detection unit; rhythm irregularity detection unit; 5; FFT processing unit; 6
;Statistical Processing Department, 7;Diagnosis Department Applicant: NEC Corporation

Claims (1)

[Claims] (1) A/D conversion means for converting an electrocardiogram signal into a digital quantity; means for detecting an irregular state of heart rhythm by detecting the QRS complex in the electrocardiogram signal; and the A/D conversion means based on the output of this means. Among the outputs of the conversion means, a signal in a range including the P wave in each heartbeat is subjected to FFT processing to obtain the frequency amplitude characteristics and frequency phase characteristics of the signal within this range. 1. An electrocardiogram signal processing device comprising diagnostic means for diagnosing second-degree atrioventricular block Movitz type II, second-degree atrioventricular block Behnke buffer type, and sinoatrial block.