Computer based pattern recognition of carotid artery Doppler signals for disease classification: prospective validation.

A computer based pattern recognition method has been developed to classify the percent diameter reduction in nonoccluded internal carotid arteries. Using a combined B-mode/pulsed Doppler unit, the system utilizes spectral waveforms obtained from the low common and proximal internal carotid artery locations. The ECG-R wave is used as a time reference to synchronize the averaging of Doppler spectra from 20 heart cycles. An averaged waveform is generated and represents the spectral data from which features are extracted for analysis. A stepwise selection algorithm identifies a feature subset for partitioning the entire range of disease into two states, less than and greater than a decision point. Three such partitions are made, leading to the following categories: Normal, 1-20, 21-50 and 51-99% dia. reduction. A classifier was trained, tested prospectively against unknown data and the results compared to angiography. Of the 170 vessels tested, 141 (82%) were classified in the same category by angiography and the computer system. Agreement for each category was 93% (27/29) for the normals, 81.5% (44/54) for the 1-20% lesions, 78% (29/37) for the 21-50% lesions and 82% (41/50) for the 51-99% lesions. The computer method and angiography differed by more than one category in only one of the 170 tests. The level of agreement corrected for chance (Kappa +/- SE(K] was 0.769 +/- 0.039. Future efforts will be directed toward dividing classification of disease further (especially in the 51-99% category), developing a dedicated microprocessor for on-line analysis of the signals and using the system for prospective epidemiological studies of various populations.

[1]  D J Farrar,et al.  Noninvasively and invasively measured pulsatile haemodynamics with graded arterial stenosis. , 1979, Cardiovascular research.

[2]  Jacob Cohen,et al.  Weighted kappa: Nominal scale agreement provision for scaled disagreement or partial credit. , 1968 .

[3]  R G Gosling,et al.  Screening for carotid junction disease by spectral analysis of Doppler signals. , 1977, Cardiovascular research.

[4]  R W Gill,et al.  Performance of the mean frequency Doppler modulator. , 1979, Ultrasound in medicine & biology.

[5]  J. Fleiss.  Measuring nominal scale agreement among many raters. , 1971 .

[6]  F. M. Greene,et al.  Computer based pattern recognition of carotid arterial disease using pulsed Doppler ultrasound. , 1982, Ultrasound in medicine & biology.

[7]  K. Beach,et al.  Post-endarterectomy carotid ultrasonic duplex scanning concordance with contrast angiography. , 1983, Ultrasound in medicine & biology.

[8]  D. Strandness,et al.  The Use of Common Carotid Waveform Analysis in the Diagnosis of Carotid Occlusive Disease , 1983, Angiology.

[9]  B. Angelsen,et al.  Transcutaneous measurement of blood flow velocity in the human aorta. , 1976, Cardiovascular research.

[10]  J P Archie,et al.  A Simple, Non‐Dimensional, Normalized Common Carotid Doppler Velocity Wave‐Form Index That Identifies Patients with Carotid Stenosis , 1981, Stroke.

[11]  Donald W. Baker,et al.  Duplex Scanner II: for Simultaneous Imaging of Artery Tissues and Flow , 1974 .

[12]  K W Beach,et al.  Evaluating carotid artery disease. The concordance between pulsed Doppler/spectrum analysis and angiography. , 1983, Ultrasound in medicine & biology.

[13]  Philip J. McCarthy,et al.  The Use of Balanced Half-Sample Replication in Cross-Validation Studies , 1976 .

[14]  D J Phillips,et al.  Carotid Artery Velocity Patterns in Normal and Stenotic Vessels , 1980, Stroke.

[15]  T. R. P. Martin,et al.  Objective feature extraction applied to the diagnosis of carotid artery disease using a Doppler ultrasound technique , 1980 .

[16]  F. M. Greene,et al.  Computer based classification of carotid arterial disease: a prospective assessment. , 1982, Stroke.

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

[18]  M. Anliker,et al.  Noninvasive Measurement of Velocity Profiles and Blood Flow in the Common Carotid Artery by Pulsed Doppler Ultrasound , 1976, Stroke.

[19]  R. Rutherford,et al.  The use of velocity wave form analysis in the diagnosis of carotid artery occlusive disease. , 1977, Surgery.

[20]  P. L. Shreve,et al.  A System for Locating Points, Lines, and Planes in Space , 1977, IEEE Transactions on Instrumentation and Measurement.

[21]  R S Cobbold,et al.  Method for estimating the Doppler mean velocity waveform. , 1982, Ultrasound in medicine & biology.

[22]  P. Chikos,et al.  Correlation of arteriographic findings and symptoms in cerebrovascular disease , 1980, Neurology.

[23]  D. E. Strandness,et al.  Empirical findings relating sample volume size to diagnostic accuracy in pulsed doppler cerebrovascular studies , 1982, Journal of clinical ultrasound : JCU.

[24]  P. Chikos,et al.  Ultrasonic Duplex Scanning for Disease of the Carotid Artery , 1981, Circulation.

[25]  Peter C. Jurs,et al.  Iterative least squares development of discriminant functions for spectroscopic data analysis by pattern recognition , 1972, Pattern Recognit..

[26]  S B Sherriff,et al.  Assessment of directional doppler ultrasound techniques in the diagnosis of carotid artery diseases , 2022 .

[27]  J F Brinkley,et al.  Ultrasonic three-dimensional imaging and volume from a series of arbitrary sector scans. , 1978, Ultrasound in medicine & biology.