Surgery – Diagnostic testing – Cardiovascular
Reexamination Certificate
2000-03-24
2002-08-20
Getzow, Scott M. (Department: 3762)
Surgery
Diagnostic testing
Cardiovascular
Reexamination Certificate
active
06438409
ABSTRACT:
FIELD OF THE INVENTION
The present inventions relate to methods of characterising ventricular operation. In particular, but not exclusively, they relate to a system for quantifying abnormalities of an electrocardiogram and to a method and an apparatus for measuring such abnormalities. The present inventions also extend to an operating system for a computer, to a computer program and to media having stored thereon a computer program for putting the inventions into effect. Other applications include use of the algorithms in pacemakers and heart monitors. The inventions share a common link of characterising differences in the wavefront of the repolarisation wave.
BACKGROUND OF THE INVENTION
Electrocardiographic patterns of the heart's movements have been well studied. An electrocardiogram (ECG) records the changes in electrical potential associated with the spread of depolarisation and repolarisation through the heart muscle. In a normal healthy patient, depolarisation starts in an area of the right atrium called the sinoatrial node and spreads through the atrioventricular node and into the ventricular muscle via specialised conduction tissue, causing the two atria and the two ventricles to contract. During repolarisation, the atria and ventricles relax and refill with blood. The depolarisation of the atria is responsible for the P wave of an ECG and depolarisation of the ventricles results in the QRS complex. Repolarisation of the atria coincides with the QRS complex so it is not seen. Repolarisation of the ventricles, however, is seen as the T wave.
ECG's are typically recorded using a standard arrangement of 12 leads, 6 (the I, II, III, VR, VL, VF leads) looking at the heart in different directions in an approximately vertical plane of a body in an upright position and 6 (the V
1
, V
2
, V
3
, V
4
, V
5
and V
6
leads) looking at the heart in different directions in an approximately horizontal plane. Using such an arrangement of leads, the spread of the waves of electrical potential associated with depolarisation and repolarisation through the three dimensional space of the body, can be recorded.
The spread of these waves through the heart is often described by vectors. For example, the average direction of spread of the depolarisation wave through the ventricles as seen from the front of the body is called the cardiac axis and the direction of this axis has long been used to indicate different abnormalities of the heart.
To study abnormalities associated with ventricular repolarisation, a number of data processing techniques have been proposed to measure, for example, the QT interval, i.e. the interval between the beginning of depolarisation and the end of repolarisation of the ventricles. Interlead variability of the QT interval durations in standard 12 lead ECG recordings has also been studied. However, whilst these measurements may provide some diagnostic assistance, concerns have been raised about the poor reproducibility of results.
Studies have also tried to quantify the inhomogeneities in the ventricular repolarisation patterns by evaluating the complexity of the T wave morphology using eigenvalues associated with the principal components of ECG, measured over a period of 24 hours. The direction of the ECG vector during T wave in the 3D physical (x,y,z) has also been shown to have some predictive value.
However, there is still a need for further measurements which may provide a more accurate technique for identifying certain conditions, particularly those which affect repolarisation of the ventricles. A problem with known methods, for example, is that they only quantify global variations in the T wave rather than spatial variations in individual waves, that is the synchronicity of the T wave, as observed from different locations on the body, is not observed.
SUMMARY OF THE INVENTION
Thus, viewed from a first broad aspect, a first invention described herein provides a method of quantifying abnormalities of an electrocardiogram observing repolarisation patterns from different locations on a body, wherein the abnormalities are quantified by a measure of the synchronicity of the repolarisation patterns as observed from those different locations on the body. In other words, this is a measure of the homogenity of the spread of repolarisation waves.
Unlike depolarisation of the heart muscle, the repolarisation of individual cells is not triggered by neighbouring cells but is instead a time dependent process. If repolarisation patterns, as observed in different locations on the body, lack synchronicity, then this can be indicative of certain heart complications.
By quantifying these abnormalities, it may be possible to use the data to assist with diagnosis or to identify patients at most risk or classify them into different categories of risk. This may be of great importance in determining whether certain treatments should be offered to a patient, for example. The data could also be used to trigger an alert in a monitoring device.
The homogenity of the spread of repolarisation waves can be measured by quantifying the spatial variability of the ventricular repolarisation patterns i.e. the spatial variability of the T wave.
Thus viewed from a second aspect, the first invention provides a method of characterising ventricular operation, comprising the steps of:
recording a signal monitoring the propagation of a repolarisation wave;
determining a vector which is representative of the wavefront of the repolarisation wave; and
determining a measure of the spatial variation of the repolarisation wavefront.
In one preferred embodiment, it provideed a method of quantifying abnormalities of ventricular repolarisation by determining a measure of the spatial T wave morphology variation.
Preferably the spatial T wave morphology variation is quantified by measuring the T wave Morphology Dispersion.
Preferably this is achieved by determining vectors describing the contributions which the signals from each lead (often referred to as the channels of an ECG) makes to the T wave. The angles between these vectors are then calculated and a mean value is determined. This mean value of the angles provides a measure of the spatial T wave morphology variation. The smaller the value, the closer the T wave morphologies will be in the signals of the individual leads.
Preferably the ECG signal is morphologically filtered to improve the signal to noise ratio. In one preferred embodiment, this consists of the steps of decomposing the T wave using a technique such as Singular Value Decomposition, filtering by keeping only the two most significant signal components, and applying a DC compensation. A preferred DC compensation is provided by subtracting an average of the start and end signal components during the QRS complex and T wave. The morphologically filtered T wave is then preferably rescaled to equalise energies in the different component directions. The corresponding reconstruction parameters are calculated to determine the vector contributions of each of the ECG leads. The angles between each pair of the vector contributions is then calculated and the mean determined. Most preferably the contribution in respect of lead V1 is ignored because the T wave morphology in this lead is generally different than that of other channels, irrespective of any clinical background, mainly due to the position of the V
1
electrode, and by ignoring this component, it has the effect of enhancing the predictive value of the T wave morphology dispersion descriptor.
The main reason for initially decomposing the data matrix is to find an optimum representation of the ECG signals upon which the measurements can be performed. In this way, the system does not use the standard XYZ axes of the body but finds an optimally constructed orthogonal system to represent the 12 lead ECG. In a preferred embodiment, therefore, the first invention can be seen as providing a method for looking at the vector representation of each of the standard electrocardiographic leads in an optimum dimensional vector space in which the ECG si
Acar Burak
Batchvarov Velislav Nikolaev
Malik Marek
Berry Thomas G.
Getzow Scott M.
Medtronic Inc.
Waldkoetter Eric R.
Woods Thomas F.
LandOfFree
Methods of characterizing ventricular operations and... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Methods of characterizing ventricular operations and..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Methods of characterizing ventricular operations and... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2931687