Introduction:

The monitoring of cardiorespiratory signals (acquisition, processing and analysis) in different scenarios facilitates the diagnosis but faces the loss of information due to disturbances and the lack of robust processing and analysis methods. In this research, we used 3 cardiorespiratory signals: electrocardiogram (ECG), phonocardiogram (PCG) and seismocardiogram (SCG), and their corresponding time series.

Methods:

We characterized the source of disturbances and the performance of ECG processing and analysis methods using international and in-house databases. Twenty-one novel approaches were proposed and evaluated with 1 or more of the aforementioned signals. Moreover, we presented 4 tools that showcase the practical contribution.

Results:

The 6 findings derived from the characterization suggest how to extract meaningful information from the electrocardiographic signal. The proposed methods performed similarly or better than the state-of-the-art and proved to be robust to disturbances in the signals and in their time series. This opens new alternatives to avoid errors during the processing. These innovative monitoring tools allow for ubiquitous, reliable, friendly, fast, and robust diagnosis in presence of disturbances in the signals and in their time series. As some conclusions was the obtained findings, methods and tools improve the diagnosis and life quality of the patients by using portable, non-invasive, safe, reusable, non-polluting, easily generalized and low-cost devices. This allows for extending and enhancing cardiology services in Cuba.

        · resumen en Español     · texto en Español     · Español ( pdf )