Data Mining methods for the stratification of the arrhythmic risk in young and master athletes

The paper describes the data mining procedures concerning ECG signals and other cardiovascular blood parameters collected at the Ospedale Maggiore di Crema to evaluate the cardiovascular response to exercise in young and master athletes, compared with control groups of untrained subjects. After recruiting four groups of healthy athletes and sedentary subjects, with age under and over 40, we analyzed the collected ECG and cardiovascular data by means of clustering techniques and Artificial Neural Networks, obtaining cross-validated classifications and significant variable differences among clusters. We could establish some important relationships between physical activity, age, sex, and cardiovascular parameters. In particular the existence of significant differences in the cardiovascular status of these groups was shown, depending in particular on the MSE1, PNN20, VO and FC variables. This will make it possible to follow the subjects examining the variability of these parameters over time, in order to identify indicators of arrythmic risk that may help to prevent possibly fatal cardiac events.