Improved Methods for Processing Optical Mapping Signals from Human Left Ventricular Tissues at Baseline and following Adrenergic Stimulation

DIEZ, EMILIANO R; VALLEJO-GIL, JOSE M; MATAMALA-ADELL, MARTA; BALLESTER-CUENCA, CARLOS; PUEYO, ESTHER; RHYINS, JULIA; FRESNEDA-ROLDAN, PEDRO C; SORRIBAS-BERJON, FERNANDO; SEGOVIA-ROLDAN, MARGARITA; PEREZ-ZABALZA, MARIA; MOUNTRIS, KOSTANTINOS A; FANANAS-MASTRAL, JAVIER; VAZQUEZ-SANCHO, MANUEL; OLIVAN-VIGUERA, AIDA

Computing in Cardiology

IEEE Computer Society

Año: 2020 vol. 2020

2325-8861

Optical mapping (OM) allows ex vivo measurement of electrophysiological signals at high spatio-temporal resolution, but the signal-to-roise ratio is commonly low. A variety of software options have been proposed to extract relevant information from OM recordings, being ElectroMap the most advanced tool currently available. In this study, improved methods are presented for processing OM signals of cardiac transmembrane voltage. A software called OMap is developed that incorporates novel techniques into ElectroMap for improved baseline drift removal, spatiotemporal filtering and characterization of action potential duration (APD) maps. In synthetically generated signals contaminated with baseline wander, white noise and the combination of both, the errors in APD maps between noisy and clean signals are remarkably lower for OMap than for ElectroMap, particularly for high noise levels. In OM signals recorded from human ventricular tissue specimens, OMap allows to clearly characterize the APD shortening effect induced by ß-adrenergic stimulation, whereas ElectroMap renders highly overlapped APD distributions for baseline and ß-adrenergic stimulation. In conclusion, improved methods are proposed and tested to characterize human ventricular electrophysiology from noisy OM recordings.