| Title |
P-wave morphological variability exacerbation prior to atrial fibrillation episodes |
| Authors |
Noriega, Maikel ; Martín-Yebra, Alba ; Savelev, Aleksei ; Platonov, Pyotr ; Marta, Javier ; Butkuviene, Monika ; Martínez, Juan Pablo ; Laguna, Pablo |
| DOI |
10.22489/CinC.2025.084 |
| Full Text |
|
| Is Part of |
Computing in cardiology (CinC): 52nd international conference, CINC 2025, São Paulo, Brazil: program & final papers.. [S.l.] : "Computing in Cardiology". 2025, vol. 52, p. 1-4.. ISSN 2325-8861. eISSN 2325-887X |
| Abstract [eng] |
This study analyzes the temporal evolution of the short-term P-wave morphological variability (PWMV) within the one-hour interval immediately preceding the onset of atrial fibrillation (AF) episodes. This PWMV is hypothesized to be an index of vulnerable atrial substrate, allowing to identify instability increases preceding the AF episode onset. We analyzed 32 ambulatory ECG recordings from subjects with paroxysmal AF. For each i-th subject and j-th AF episode, the PWMV was characterized in consecutive 2-min windows (k = 1 . . . 30) within the last 60 min before the AF onset. For each 2-min window, the median absolute deviation (MAD) of the P wave signals pi,j,kMAD(n); and the MAD energy normalized to the median of the P wave energy in each subject Eri,j(k) were measured. Then, for each subject, a normalized median across AF episodes was obtained, Ẽri(k). A two-segment linear function fit of Ẽr(k) temporal evolution, by least squares error (LSE) minimization was used to characterize the cohort dynamics. Analyzing the cohort dynamics of Ẽr(k), a relative stability can be observed up to 8 minutes before the AF episode onset, when Ẽr(k) value starts increasing. In conclusion, PWMV can be monitored by quantifying the P-wave short-time MAD relative energy, which shows an increasing trend within the 8 minutes previous to the AF onset. |
| Published |
[S.l.] : "Computing in Cardiology" |
| Type |
Conference paper |
| Language |
English |
| Publication date |
2025 |
| CC license |
|