VDI Vision - Analysis of Ventricular Electrical Dyssynchrony in Real-Time

0555022 - ÚPT 2022 RIV US eng C - Konferenční příspěvek (zahraniční konf.)
Plešinger, Filip - Viščor, Ivo - Vondra, Vlastimil - Halámek, Josef - Koščová, Zuzana - Leinveber, P. - Čurila, K. - Jurák, Pavel
VDI Vision - Analysis of Ventricular Electrical Dyssynchrony in Real-Time.
2021 Computing in Cardiology (CinC). Vol. 48. New York: IEEE, 2021, č. článku 103. ISBN 978-166547916-5. ISSN 2325-8861. E-ISSN 2325-887X.
[Computing in Cardiology 2021 /48./. Brno (CZ), 12.09.2021-15.09.2021]
Institucionální podpora: RVO:68081731
Klíčová slova: Ultra high frequency ECG * Ventricular Electrical Dyssynchrony * real-time analysis
Obor OECD: Medical engineering
https://ieeexplore.ieee.org/document/9662916

Background: Ventricular electrical dyssynchrony can be examined using ultra-high-frequency (UHF-ECG) analysis. Furthermore, UHF-ECG analysis would allow direct optimization of pacing therapy. Here we introduce VDI vision (Ventricular Dyssynchrony Imaging), a desktop application for the real-time processing of UHF-ECG recordings. Method: Incoming ECG data (5kHz, 26 bits, 24 channels) are processed as follows: QRS detection, pacemaker stimuli elimination, QRS clustering, amplitude envelopes in nine frequency bands, and final combination into the Ventricular Depolarization (VD) map. The VD map is updated whenever a new QRS is detected. Results: We developed the VDI vision using the. NET platform. Until the end of March 2021, the VDI monitor was used to analyze 773 and 4,849 recordings at ICRC-FNUSA hospital (Brno, Czechia) and FNKV hospital (Prague, Czechia), respectively. The median length for ICRC-FNUSA recordings was 124 (IQR 121-139) seconds. The median length for recordings at FNKV hospital was 157 seconds (IQR 127-200). Conclusion: The VDI vision delivers information about electrical ventricular dyssynchrony in real-time. The instant analysis allows using the software during implant procedures for optimizing electrode placement and pacing. The presented real-time solution also significantly minimized measurement duration.
Trvalý link: http://hdl.handle.net/11104/0329641