Ebden M.J.Tarassenko L.Payne S.J.Darowski A.Price J.D.STEPHEN JOHN PAYNE2022-05-242022-05-242004https://www.scopus.com/inward/record.uri?eid=2-s2.0-11044236949&partnerID=40&md5=5146d43030e19d4afc85f4805cd1ef2ahttps://scholars.lib.ntu.edu.tw/handle/123456789/611877The Smoothed Pseudo Wigner-Ville Distribution (SPWVD) is used for the time-frequency analysis of variations in RR interval. A novel technique to determine the smoothing window lengths is implemented, and a new heart rate variability (HRV) metric is developed, instantaneous center frequency variability (ICFV), which uses the time-frequency map generated by the SPWVD. The technique is then applied to 50 patients with unexplained falls and age > 60, under-going head-upright tilt table testing (HUT). Eighteen of the patients were diagnosed with vasovagal syndrome. Attempts at syncope prediction using the new metric is an improvement on traditional techniques: an ICFV less than 0.07 Hz from 90 s to 180 s after tilt is predictive of a negative test (negative predictive value: 0.77). The comorbidity and autonomie degeneration present in elderly patients are thought to be responsible for lowering the negative predictive value.Blood vessel prosthesesCardiovascular systemElectrocardiographyFrequency domain analysisNeurologyTime domain analysisHeart rate variability (HRV)Vasovagal syncopesVasovagal syndromesBiomedical engineeringconference paper2-s2.0-11044236949