Recognition of Atrial Pathologies in the Elderly through the study of P wave morphology

The global population aged 60 years or over is more than twice as large as in 1980, and this number will increase in the future. In Italy , after Japan, the elderly constitute an important part of the total population, with a number of people over 60 that is expected to increase in the following years.In our country there is the highest proportion of elderly citizens (22.4% of the total population has more than 65 years in 2017) among European countries and for this reason aging and chronic diseases can be considered two of the major matters that the public health care system has to face up in the next years in order to reduce the health care cost and improve the citizens quality of life.In particular, the Cardiovascular diseases (CVD) in older people can be considered one of the more frequent and obvious problem which comes from this context.One-third of deaths throughout the world is caused to Cardiovascular diseases."Past periods of decline in cardiovascular disease mortality marked a remarkable achievement for public health and medical care around the world," said Dr. Christopher Murray, director of IHME and study co-author. "Governments, advocacy groups, clinicians, and communities should look to this new evidence when developing programs and policies that could reduce the burden of cardiovascular disease and save more lives."The Automatic Analysis of Electrocardiograms (ECGs) provides an important support for diagnostic classification, in ECG databases, to predict some heart disease. A recently study, called 'The Predictor', about the correlation between the P-wave morphology and the heart failure, has demonstrated that it is possible to predict the heart disease insurgence analyzing the ECG signal. Some cardiac abnormalities can be identificated extractioning a series of significant ECG parameters.A research, conducting by an equipe of doctors at the ''San Giovanni Addolorata'' Hospital in Rome, has extracting six principal class in that it is possible to classify P-wave morphology. Also they found a correlation between P-wave duration and arising to the heart pathology. This study wants to realize an automatic measurement algorithm in high-resolution digital electrocardiograms, for the detection and the classification of P-wave, to be a support for the doctors and gives results faster and repeatable. The focus is to intervene early and avoid a sudden heart failure for the patients. The algorithm system has been implemented and simulated in Matlab. The algorithm was implemeted used recordings of electrocardiograms tracing from twenty-four patients, each file, in xml format, includes general information about the patient, some characteristics of the sampling and the traces of the twelve derivations. All files were kindly provided by Dr. Rafal Baranowski, collaborator of Dr. Antonio Bayés De Luna. In this file there are all different kinds of P-wave morphology.Subsequently, 1000 data were provided by a Company in the Biomedical sector, used to test the robustness of the algorithm on a larger datasetThe key points of the algorithm are- Baseline estimation and denoising using sparsity and the Savitzky-Golay filter.- Wavelet Transform - the Levenberg-Marquardt Algorithm and Euclidean distanceOnce the data has been processed, the six classes, in which the P waves are divided, can be recognized.