Method for localizing a site of origin of an electrical heart activity

• US 5,634,469 A
• Filed: 03/26/1996
• Issued: 06/03/1997
• Est. Priority Date: 03/29/1995
• Status: Expired due to Term

• Current Assignee: Siemens AG
• Original Assignee: Siemens AG
• Inventors: Bruder, Herbert, Killmann, Reinmar
• Primary Examiner(s): Kamm, William E.
• Assistant Examiner(s): Getzow, Scott M.
• Application Number:
  US08/622,688
• Time in Patent Office: 434 Days
• Field of Search: 128/696, 128/699
• US Class Current: 600/512
• CPC Class Codes: A61B 5/282 Holders for multiple electrodes for introduction into the body A61B5/287
• CPC Class Codes: A61B 5/349 Detecting specific parameters of the electrocardiograph cycle

Abstract

In a method for localizing an origination site of electrical heart activity, body surface potentials generated by the heart activity are measured at a number of measuring points with a multi-channel measuring system, and values are stored that characterize the body surface potentials at the measuring points. These values are then compared to comparison values stored in a data bank, the comparison values representing comparison surface potentials that arise from modeled comparison heart activities whose position in the heart is known. The position of that comparison heart activity whose comparison values exhibit the most similarity to the characteristic values is emitted as a localization result. The comparison values are determined using a heart model embedded in a thorax model.

Claims

1. A method for localizing an origination site of electrical heart activity comprising the steps of:

providing a heart model and providing a thorax model and embedding said heart model in said thorax model, and storing said comparison values in a data bank;

using said heart model embedded in said thorax model, generating a plurality of comparison values representing modeled heart activities respectively originating at a plurality of known positions;

measuring body surface potentials produced by heart activity of a subject at a plurality of measuring points using a multi-channel measuring system and obtaining measured values respectively representing the body surface potentials at said measuring points, generating a characteristic value from said measured values and storing said characteristic value;

comparing said characteristic value to each of said comparison values in said data bank and identifying a comparison value having a highest similarity to said characteristic value; and

identifying the known position of said comparison value having the highest similarity to said characteristic value and emitting said position as the site of origin of the heart activity which produced said body surface potentials.

2. A method as claimed in claim 1 wherein the step of comparing said comparison values and said characteristic value comprises:

forming a first set of normalized correlation coefficients from each of said comparison values and forming a second set of normalized correlation coefficients from said measured values as said characteristic value;

correlating each of said first sets with said second set of correlation coefficients to identify correlation coefficients in a first set having a highest correlation with correlation coefficients in said second set; and

using said correlation coefficients in said first set having the highest correlation with correlation coefficients in the second set as said comparison values having the highest similarity to said characteristic value.

3. A method as claimed in claim 1 wherein the step of generating said characteristic value from said measured values comprises generating a characteristic value identifying the QRS complex of said subject from said body surface potentials, and wherein the step of generating said comparison values comprises generating comparison values respectively representing a modeled QRS complex from said heart model embedded in said thorax model.

4. A method as claimed in claim 1 wherein the step of generating said characteristic values comprises generating a characteristic value reproducing a chronological curve of said body surface potentials at the respective measuring points.

5. A method as claimed in claim 4, comprising the further steps of:

selecting successive discrete values of said chronological curve of said body surface potentials at the respective measuring points and forming a measured value vector therefrom;

generating a plurality of comparison vectors respectively for each known position of heart activity from said heart model embedded in said thorax model;

comparing said measured value vector to each comparison vector by forming a plurality of scalar products respectively of said measured value vector and each comparison vector;

identifying a highest scalar product of said plurality of scalar products and identifying the position of the heart activity associated with the comparison vector which produced said highest scalar product; and

using the position of the activity associated with said comparison vector which produced the highest scalar product as said side of origination of said heart activity which produced said body surface potentials.

6. A method as claimed in claim 1 comprising the further steps of:

generating a chronological curve of modeled comparison surface potentials from heart activity in said heart model embedded in said thorax model and identifying successive discrete values of said chronological curve of modeled comparison surface potentials;

extracting eigenvectors from said discrete values of said time curve of modeled comparison surface potentials;

generating a chronological curve of said body surface potentials measured at the respective measuring points and identifying successive discrete values of said chronological curve of said body surface potentials;

conducting a spectral decomposition of said discrete values of said chronological curve of said body surface potentials to obtain a plurality of measured development coefficients;

conducting a spectral decomposition of said discrete values of said chronological curve of said comparison surface potentials to obtain a plurality of comparison expansion coefficients; and

comparing said expansion coefficients associated with the measurement signal and comparison expansion coefficients end identifying comparison expansion coefficients having a greatest similarity to said expansion coefficients associated with the measurement signal and using the known position of the modeled position surface potentials which produced said comparison expansion coefficients having the highest similarity to said expansion coefficient associated with the measurement signal as the site of origin of said heart activity which produced said body surface potentials.