Classification of lower limb arterial stenoses from Doppler blood flow signal analysis with time-frequency representation and pattern recognition techniques.

A pattern recognition system was used to classify Doppler blood flow signals for the determination of lower limb arterial stenoses. The diagnostic features were extracted from time-frequency representations of Doppler signals. Three techniques were tested to estimate time-frequency representations: the short-time Fourier transform, the autoregressive (AR) modeling, and the Bessel distribution. A boundary tracking algorithm was proposed to extract the frequency contour of the Doppler time-frequency representations. Based on the characteristics of the Doppler frequency contour, shape descriptors from an autoregressive analysis were proposed as diagnostic features. Simple algorithms were proposed to normalize these autoregressive shape descriptors. Amplitude distribution of the Doppler time-frequency representation was also found useful for stenosis classification. A total of 379 arterial segments from the aorta to the popliteal artery were classified by the pattern recognition system into three categories of diameter reduction (0-19%, 20-49%, and 50-99%). The short-time Fourier transform provided an overall accuracy of 80% (kappa = 0.38); AR modeling, 81% (kappa = 0.42); and the Bessel distribution, 82% (kappa = 0.43). All these results are better than those based on visual interpretation (accuracy = 76%, kappa = 0.29) performed by a trained technologist. The AR modeling and the Bessel distribution improved the performance of the pattern recognition system in comparison with the short-time Fourier transform. It is likely that with further improvement, the pattern recognition approach will be a useful clinical tool to quantify stenoses and to follow the disease progression with more reliability and less bias than visual interpretation as done currently in clinical practice.

[1]  R. Cobbold,et al.  Comparison of four digital maximum frequency estimators for Doppler ultrasound. , 1988, Ultrasound in medicine & biology.

[2]  G Cloutier,et al.  Computer evaluation of Doppler spectral envelope area in patients having a valvular aortic stenosis. , 1990, Ultrasound in medicine & biology.

[3]  Jacob Cohen.  A Coefficient of Agreement for Nominal Scales , 1960 .

[4]  A. Creutzig,et al.  Duplex scanning of the peripheral arteries: correlation of the peak velocity ratio with angiographic diameter reduction. , 1992, Ultrasound in medicine & biology.

[5]  G. Cloutier,et al.  Change in amplitude distributions of Doppler spectrograms recorded below the aortic valve in patients with a valvular aortic stenosis. , 1991, Ultrasound in medicine & biology.

[6]  G. Moneta,et al.  Accuracy of lower extremity arterial duplex mapping. , 1992, Journal of vascular surgery.

[7]  P Robillard,et al.  Computer analysis and pattern recognition of Doppler blood flow spectra for disease classification in the lower limb arteries. , 1991, Ultrasound in medicine & biology.

[8]  D. Strandness,et al.  Can duplex scanning replace arteriography for lower extremity arterial disease? , 1990, Annals of vascular surgery.

[9]  Anil K. Jain.  Advances in statistical pattern recognition , 1987 .

[10]  R. Cobbold,et al.  On the Doppler signal from a steady flow asymmetrical stenosis model: effects of turbulence. , 1993, Ultrasound in medicine & biology.

[11]  Jechang Jeong,et al.  Kernel design for reduced interference distributions , 1992, IEEE Trans. Signal Process..

[12]  S.M. Kay,et al.  Spectrum analysis—A modern perspective , 1981, Proceedings of the IEEE.

[13]  D J Phillips,et al.  Noninvasive mapping of lower limb arterial lesions. , 1985, Ultrasound in medicine & biology.

[14]  A. Oliviero,et al.  A new approach to contour coding , 1981 .

[15]  William J. Williams,et al.  Improved time-frequency representation of multicomponent signals using exponential kernels , 1989, IEEE Trans. Acoust. Speech Signal Process..

[16]  Godfried T. Toussaint,et al.  Bibliography on estimation of misclassification , 1974, IEEE Trans. Inf. Theory.

[17]  Peter N. Burns,et al.  Clinical applications of Doppler ultrasound , 1988 .

[18]  Louis-Gilles Durand,et al.  A Bayes model for automatic detection and quantification of bioprosthetic valve degeneration , 1988 .

[19]  Rama Chellappa,et al.  Stochastic models for closed boundary analysis: Representation and reconstruction , 1981, IEEE Trans. Inf. Theory.

[20]  Zhenyu Guo,et al.  The time-frequency distributions of nonstationary signals based on a Bessel kernel , 1994, IEEE Trans. Signal Process..

[21]  R S Cobbold,et al.  A comparative study and assessment of Doppler ultrasound spectral estimation techniques. Part II: Methods and results. , 1988, Ultrasound in medicine & biology.

[22]  King-Sun Fu,et al.  Shape Discrimination Using Fourier Descriptors , 1977, IEEE Trans. Syst. Man Cybern..

[23]  Thomas S. Huang,et al.  Image processing , 1971 .

[24]  D E Strandness,et al.  Duplex Scanning for Diagnosis of Aortoiliac and Femoropopliteal Disease: A Prospective Study , 1987, Circulation.

[25]  G Cloutier,et al.  Limitations of ultrasonic duplex scanning for diagnosing lower limb arterial stenoses in the presence of adjacent segment disease. , 1994, Journal of vascular surgery.

[26]  S R Cannon,et al.  Digital fourier techniques in the diagnosis and quantification of aortic stenosis with pulsed‐Doppler echocardiography , 1982, Journal of clinical ultrasound : JCU.

[27]  P. Wintz,et al.  An efficient three-dimensional aircraft recognition algorithm using normalized fourier descriptors , 1980 .