Literature DB >> 9700218

Spectral broadening of clinical Doppler signals using FFT and autoregressive modelling.

P I Keeton1, F S Schlindwein.   

Abstract

OBJECTIVE: This paper investigates the behaviour of the spectral broadening index (SBI) derived from spectra obtained using autoregressive (AR) modelling compared to that of SBI based on fast Fourier transform (FFT) analysis of clinical Doppler ultrasound scans.
METHODS: Doppler signals from internal carotid arteries of patients with normal and diseased vessels with up to 80% stenosis were analysed. A threshold at -6 dB of the maximum magnitude component of each individual spectrum was implemented to reject low-level noise. The SBI was obtained using the maximum and the mean frequency envelopes extracted from the sonogram.
RESULTS: A qualitative improvement in both the appearance of the AR sonograms and the shape of the individual AR spectra was noticeable. The AR approach consistently produced narrower spectra than the FFT and the shapes of the frequency envelopes derived from the AR sonogram and the FFT sonogram were also rather different. Despite these differences a strong correlation was observed between the value of the FFT-based SBI and the AR-based SBI. The mean value of the FFT-SBI is larger than that of the AR-SBI and the variance of the FFT-SBI is smaller than that of the AR-SBI based on a set of at least 20 sequentially recorded heartbeats.
CONCLUSIONS: It was established that, for all cases where significant stenosis was present, a statistically significant value for SBI could be obtained using four or more heartbeats if five spectra around the peak systole were used to estimate the SBI of each individual heartbeat. No quantitative advantage in using the AR approach over the FFT for the determination of SBI was obtained due to the poorer variance of the AR-SBI and the additional computational complexity of the AR approach. Copyright 1998 Elsevier Science Ireland Ltd. All rights reserved.

Entities:  

Mesh:

Year:  1998        PMID: 9700218     DOI: 10.1016/s0929-8266(98)00032-9

Source DB:  PubMed          Journal:  Eur J Ultrasound        ISSN: 0929-8266


  1 in total

1.  Determination of stenosis and occlusion in arteries with the application of FFT, AR, and ARMA methods.

Authors:  Elif Derya Ubeyli; Inan Güler
Journal:  J Med Syst       Date:  2003-04       Impact factor: 4.460

  1 in total

北京卡尤迪生物科技股份有限公司 © 2022-2023.