Influence of Body-Surface Geometry Accuracy on Noninvasive Reconstruction of Electrical Activation and Recovery in Electrocardiographic Imaging

M.J.M. Cluitmans*, P.G.A. Volders

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference article in proceedingAcademicpeer-review

Abstract

In Electrocardiographic Imaging (ECGI), electrical potentials on a patient's heart surface are noninvasively reconstructed from potential recordings on the body surface. This process depends on a geometrical description of the heart surface and body surface. Here, we investigate the influence of inaccuracies in the body-surface geometry on ECGI reconstruction accuracy.In four canines, ca. 200 electrodes were attached to the body surface and in total 93 beats recorded. A CT scan provided the exact position of these electrodes and the heart-surface. The body geometry was defined by the electrode positions only, and their position in the digital geometry was then varied: 1) for all electrodes collectively in the same direction, 2) or in a random direction (where electrodes on the same strip were taken together). Invasive recordings on the heart surface provided ground truth data to which noninvasive reconstructions with these varying geometries were compared, in terms of correlation coefficients between recorded and reconstructed potentials (CC), localization error (LE) between known beat origin and reconstructed beat origin, and activation and recovery isochrones.Results for displacement of all electrodes in a single direction or in random directions indicate that displacement up to 30 mm has a limited effect on ECGI accuracy. More importantly, consistent displacement of electrodes in one direction mainly results in a shift of the activation or recovery pattern in that direction. Patterns are also preserved when displacement is in random directions. Thus, inaccuracy in body-surface geometry up to 30 mm in ECGI preserves activation and recovery patterns but hampers their exact localization.
Original languageEnglish
Title of host publication2017 COMPUTING IN CARDIOLOGY (CINC)
PublisherIEEE Computer Society
Pages1-4
Number of pages4
Volume44
DOIs
Publication statusPublished - 2017
Event44th Computing in Cardiology Conference - Rennes, France
Duration: 24 Sept 201727 Sept 2017
https://www.cinc.org/2017/ProgramBook/2017CinCProgramBook2.pdf

Publication series

SeriesComputing in Cardiology Conference
Volume44
ISSN2325-8861

Conference

Conference44th Computing in Cardiology Conference
Abbreviated titleCinC
Country/TerritoryFrance
CityRennes
Period24/09/1727/09/17
Internet address

Fingerprint

Dive into the research topics of 'Influence of Body-Surface Geometry Accuracy on Noninvasive Reconstruction of Electrical Activation and Recovery in Electrocardiographic Imaging'. Together they form a unique fingerprint.

Cite this