Methodology for automated QT variability measurement

• US 5,560,368 A
• Filed: 11/15/1994
• Issued: 10/01/1996
• Est. Priority Date: 11/15/1994
• Status: Expired due to Term

• Current Assignee: Ronald D. Berger
• Original Assignee: Ronald D. Berger
• Inventors: Berger, Ronald D.
• Primary Examiner(s): Kamm, William E.
• Assistant Examiner(s): LAYNO, CARL HERNANDZ
• Application Number:
  US08/340,861
• Time in Patent Office: 686 Days
• Field of Search: 128/702, 128/703
• US Class Current: 600/517
• CPC Class Codes: A61B 5/366 Detecting abnormal QRS complex, e.g. widening

Abstract

A method for analyzing electrocardiograph signals to determine risk of malignant arrhythmias, that involves: sensing fluctuations in voltage resulting from electrical activity of a heart over a time period of about 256 seconds as signals having an analog value; converting such signals having an analog value to digital values corresponding substantially to the analog value of the signals; recording the digital values in a record; analyzing the digital values of the record by: identifying a time of each R wave of a heartbeat; defining a template QT interval for a heartbeat by selecting a beginning of a QRS complex and an end of a T wave for the heartbeat; determining an alteration value selected from the group consisting of an elongation of a heartbeat in time and a compression of a heartbeat in time as an error function for the heartbeat; performing a binary search to determine a minimal value for the error function; and assessing changes in QT interval for each heartbeat using the entire T wave.

Claims

1. A method for analyzing electrocardiograph signals to determine beat-to-beat QT interval variability, said method comprising:

a) sensing fluctuations in voltage resulting from electrical activity of a heart over a predetermined time period as signals having an analog value;

b) converting said signals having an analog value to digital values defined as x(n), where n is a sample number by sampling said analog signal with a fixed digitization interval Δ

t;

c) analyzing said digital values, said analyzing comprising;

i) identifying a time of each R wave of a heartbeat using a peak detection algorithm, wherein T_i is defined as a time location for beat i;

ii) defining a template QT interval, φ

(n) for a heartbeat defined as beat number k, by selecting a beginning of a QRS complex, n₀, and an end point of a T wave, n₁, for said heartbeat wherein;

φ

(n)=x(n) from n=n₀ to n=n₁ ;

iii) determining a QT interval for all other heartbeats by finding an optimally time-altered version of a T wave of each beat i that substantially matches a template T wave of beat wherein a QT interval determined for beat i is defined as QT_i.

2. The method of claim 1, wherein said fluctuations in voltage are sensed by leads.

3. The method of claim 2, wherein said leads comprise leads selected from the group consisting of surface leads and sub-surface leads.

4. The method of claim 3, wherein said leads comprise a ventricular leads selected from the group consisting of unipolar ventricular leads and bipolar ventricular leads endocardial ventricular leads, and epicardial ventricular leads.

5. The method of claim 4, wherein said ventricular lead is operably connected to an implanted apparatus.

6. The method of claim 5, wherein said implanted apparatus is selected from the group consisting of a pacemaker, and a defibrillator.

7. The method of claim 3, wherein said leads are surface leads selected from the group consisting of surface no. 1 leads and surface no. 2 leads, said leads being operably connected to an electrocardiograph, and said method comprising obtaining an electrocardiogram showing said fluctuation in voltage.

8. The method of claim 1, where QT_i is taken as α

_i ·

(n₁ -n₀)·

Δ

t, where α

_i is the value of time-alteration factor α

that minimizes an error functions ε

_i (α

) for beat i.

9. The method of claim 8, wherein the error function ε

_i (α

) is defined as;

##EQU3## where n.sub.∇

represents a blanking period, set for a time to avoid inclusion of the QRS complex in the error function calculation, and T_k is a time location of an R wave for a template beat k.

10. The method of claim 8, wherein the error function ε

_i (α

) is minimized by searching over a domain of values for α

as follows;

a) α

is initially set to 1.0, and a search step size Δ

α

is initially set to 0.02;

b) ε

is computed for three values as follows;

space="preserve" listing-type="equation">ε

.sub.a =ε

.sub.i (α

-Δ

α

)

space="preserve" listing-type="equation">ε

.sub.b =ε

.sub.i (α

)

space="preserve" listing-type="equation">ε

.sub.c =ε

.sub.i (α

+Δ

α

)

c) based on a value of ε

_a, ε

_b, and ε

_c which is the lowest, conducting a search as follows;

i) α

is reset to α

-Δ

α

when ε

_a is the lowest,

ii) Δ

α

is reset to Δ

α

/2 when ε

_b is the lowest, and

iii) α

is reset to α

+Δ

α

when ε

_c is the lowest; and

d) if ending the search with α

_i taken as the final value of α

that was used Δ

α

<

0.0001; and

continuing the search if Δ

α

>

0.0001 by returning to step b.

11. The method of claim 2, wherein said leads are surface electrodes attached to body limbs or chest, and operably connected to an electrocardiograph.

12. The method of claim 2, further comprising determining a QT interval variability index (QTVI) based on the series of QT interval values derived from an electrocardiographic signal.

13. The method of claim 12, wherein the QTVI is defined as:

space="preserve" listing-type="equation">QTVI=log.sub.10 (P.sub.QTV /P.sub.HRV)

where P_QTV and P_HRV represent total QT and heart rate variability power, respectively.

14. The method of claim 11, wherein the QTVI is defined as:

##EQU4## where P_QTV and P_HRV represent total QT and heart rate variability power, respectively, and <

QT> and

<

HR>

represent QT interval and heart rate, respectively.

15. The method of claim 1 further comprising determining risk of malignant arrhythmias.

16. A method for analyzing electrocardiograph signals, said method comprising:

a) sensing fluctuations in voltage resulting from electrical activity of a heart over a time period of about 256 seconds as signals having an analog value;

b) converting said signals having an analog value to digital values corresponding substantially to said analog value of said signals;

c) recording said digital values in a record;

d) analyzing said digital values of said record, said analyzing comprising;

i) identifying a time of each R wave of a heartbeat;

ii) defining a template QT interval for a heartbeat by selecting a beginning of a QRS complex and an end of a T wave for said heartbeat;

iii) determining an alteration value selected from the group consisting of an elongation of a heartbeat in time and a compression of a heartbeat in time as an error function for said heartbeat;

iv) performing a search to determine a minimal value for said error function;

v) assessing changes in QT interval for each heartbeat using said entire T wave; and

utilizing a region of said QT template from n=T_k +n.sub.∇

to n=n₁, where n.sub.∇

comprises a preset delay, to match all heartbeats to said QT template wherein;

i) comprises identifying a time for each R wave using a peak detection algorithm wherein T_i are time locations; and

ii) comprises selecting a beginning of a QT interval and an end of a QT interval for said heartbeat, wherein k is said heartbeat, so a QT template is φ

(n), where n is a sample number such that;

space="preserve" listing-type="equation">φ

(n)=x(n) from n=n.sub.0 to n=n.sub.1

where x(n) is one of said signals, and n₀ and n₁ are said beginning of said QT interval and end of said QT interval, respectively;

said QT template comprises N=n₁ -n₀ points and said template QT interval duration comprises N Δ

t where Δ

t is a digitization interval; and

said digitization interval comprises about 1 msec.