The Electrocardiogram (ECG) is a recording of the electrical activity of the heart as detected at the surface of the body. There are several characteristics which comprise a cardiac cycle, the first being the P wave, which represents the initial pacemaker pulse of the cycle and the propagation of this pulse through the atrial tissue mass. The P wave can be difficult to discern in some cases, especially elderly patients, this has lead to it being overlooked in practical cardiology and automated analysis.

The enhancement and automated analysis of the P wave would constitute a significant advance in diagnostic ECG applications. The aim of this project is to research and develop this as a practical tool and to prove that an inexpensive routine test can provide early warnings to major cardiac conditions, thus saving the health care system significant cost. In addition this test can be provided in basic facilities reducing loads on specialist clinics and practitioners.

The ECG provides a temporal-spatial map of the polarization sequence of the heart. Each feature of the ECG is represented by a letter. The P wave can be difficult to see by eye, especially in elderly patients, and thus has been largely ignored by the medical fraternity. Using a combination of modern signal processing techniques, alternate ways of viewing the data allow the P wave to be enhanced. Enhancement of the P wave and further analysis using wavelet based techniques shows marked differences between normal and unhealthy atrial conduction.

Each lead is processed to identify the location of the P waves, the method used relies upon detection of prominent markers in the ECG and then cross correlation between several leads to improve accuracy.

The leads are processed using the CWT and the areas of transform coefficients relating to the P wave are then analysed using the wavelet energy spectrum. The process observes the varying spread of P wave energy, which is intrinsically linked to the atrial conduction. This method is based upon the observation and classification of a signal based upon its entropy; this is a common approach to signal classification. Here the innovation is achieved through the use of wavelet processing and the specific application of these techniques to the analysis and characterisation of the P wave which has not been previously undertaken.

By observing the P wave in this way we reduce the effects of noise and observe characteristics and relationships in the atrial conduction which are not apparent in the normal ECG representation. When applied to a complete pathological set we determine statistical parameters and distribution patterns which are indicative of specific atrial arrhythmia. A trial to collect ECG data from a group of healthy patients and four distinct groups with abnormalities relating to the P wave has provided important developmental data to refine the process which has already been devised. Publication is in press!!Heart disease affects 17% of the total Australian population and 40% of people aged over 65 [ ] and based on figures from 2000 the cost of heart, stroke and vascular drugs was estimated at $1.546 billion to the Public Benefits Scheme (PBS) [ ]. Because our method uses the cheapest of the diagnostic equipment it may be used in place of the more expensive methods (more than 10x cheaper in many cases) in some cases and be used more often as a preventative assessment (which is difficult to quantify), we estimate significant savings could be realised. This increased prevalence of heart disease in the elderly coupled with an aging population means research and development of technologies in the field of cardiology are of vital importance and worth to the community. This work will allow for early detection, on existing commonly available equipment, of atrial arrhythmia, which will help decrease the incidence of stroke in an economic manner.

The Electrocardiogram (ECG) is a recording of the electrical activity of the heart as detected at the surface of the body. There are several characteristics which comprise a cardiac cycle, the first being the P wave, which represents the initial pacemaker pulse of the cycle and the propagation of this pulse through the atrial tissue mass. The P wave can be difficult to discern in some cases, especially elderly patients, this has lead to it being overlooked in practical cardiology and automated analysis.

The enhancement and automated analysis of the P wave would constitute a significant advance in diagnostic ECG applications. The aim of this project is to research and develop this as a practical tool and to prove that an inexpensive routine test can provide early warnings to major cardiac conditions, thus saving the health care system significant cost. In addition this test can be provided in basic facilities reducing loads on specialist clinics and practitioners.

The ECG provides a temporal-spatial map of the polarization sequence of the heart. Each feature of the ECG is represented by a letter. The P wave can be difficult to see by eye, especially in elderly patients, and thus has been largely ignored by the medical fraternity. Using a combination of modern signal processing techniques, alternate ways of viewing the data allow the P wave to be enhanced. Enhancement of the P wave and further analysis using wavelet based techniques shows marked differences between normal and unhealthy atrial conduction.

Each lead is processed to identify the location of the P waves, the method used relies upon detection of prominent markers in the ECG and then cross correlation between several leads to improve accuracy.

The leads are processed using the CWT and the areas of transform coefficients relating to the P wave are then analysed using the wavelet energy spectrum. The process observes the varying spread of P wave energy, which is intrinsically linked to the atrial conduction. This method is based upon the observation and classification of a signal based upon its entropy; this is a common approach to signal classification. Here the innovation is achieved through the use of wavelet processing and the specific application of these techniques to the analysis and characterisation of the P wave which has not been previously undertaken.

By observing the P wave in this way we reduce the effects of noise and observe characteristics and relationships in the atrial conduction which are not apparent in the normal ECG representation. When applied to a complete pathological set we determine statistical parameters and distribution patterns which are indicative of specific atrial arrhythmia. A trial to collect ECG data from a group of healthy patients and four distinct groups with abnormalities relating to the P wave has provided important developmental data to refine the process which has already been devised. Publication is in press!!