Software Open Access
Adriana N. Vest; Giulia Da Poian; Qiao Li; Chengyu Liu; Shamim Nemati; Amit Shah; Gari D Clifford
The PhysioNet Cardiovascular Signal Toolbox is a a cardiovascular dynamics analysis package, designed to meet the need in the clinical and scientific community for a validated, standardized, well-documented open-source toolkit to evaluate the relationships between physiological signals and disease. The package not only includes standard HRV tools to generate time and frequency domain metrics from ECG or pulsatile waveforms (like the blood pressure or photoplethysmographic waveforms), but more recent metrics such as acceleration and deceleration capacity and pulse transit time. The package is designed to accommodate a variety of input data, from raw unprocessed and unannotated waveforms, to fully annotated tachogram data. In general, abnormal beat and noise removal and methods for dealing with the missing data are poorly described and highly variant in most of the literature. Therefore, we have included signal processing methods that include state of the art peak detectors, signal quality processing units, and beat/rhythm phenotyping. The package can also analyze the interactions between multiple physiological signals.