Noninvasive identification of three-dimensional myocardial infarctions from inversely reconstructed equivalent current density

Zhaoye Zhou, Chengzong Han, Bin He

Research output: Contribution to journalConference articlepeer-review

Abstract

The study presents a new approach to non-invasively identifY the 3-dimensional MI substrate from the equivalent current densities (ECDs) that is inversely reconstructedfrom body surface potential maps (BSPMs). The MI substrate was characterized using a threshold determined from the ECD magnitude. A total of 114 sites of transmural infarctions, 91 sites of epicardial infarctions, and 36 sites of endocardial infarctions were simulated. The results show that: 1)With 205 BSPM electrodes and 10 f.1 V Gaussian white noise, the averaged accuracies for transmural MI are sensitivity=83.4%, specificity=82.2%, and the distance between the centers of gravity (DCG)=6.5mm. Epicardial infarctions (sensitivity=81.6%, specificity=75.8%, and DCG=7.5mm) obtained similar accuracies to endocardial infarctions (sensitivity=80.0%, specificity=77.0%, and DCG=10.4 mm). A reasonably good imaging performance was obtained under a higher noise level, fewer BSPM electrodes, and mild volume conductor modeling error, respectively. The results suggest that this method is capable of imaging the transmural and surface infarction.

Original languageEnglish (US)
Article number7043172
Pages (from-to)833-836
Number of pages4
JournalComputing in Cardiology
Volume41
Issue numberJanuary
StatePublished - 2014
Event41st Computing in Cardiology Conference, CinC 2014 - Cambridge, United States
Duration: Sep 7 2014Sep 10 2014

Fingerprint

Dive into the research topics of 'Noninvasive identification of three-dimensional myocardial infarctions from inversely reconstructed equivalent current density'. Together they form a unique fingerprint.

Cite this