Original contribution
Assessing spectral algorithms to predict atherosclerotic plaque composition with normalized and raw intravascular ultrasound data

https://doi.org/10.1016/S0301-5629(01)00436-7Get rights and content

Abstract

Spectral analysis of backscattered intravascular ultrasound (IVUS) data has demonstrated the ability to characterize plaque. We compared the ability of spectral parameters (e.g., slope, midband fit and y-intercept), computed via classic Fourier transform (CPSD), Welch power spectrum (WPSD) and autoregressive (MPSD) models, to classify plaque composition. Data were collected ex vivo from 32 human left anterior descending coronary arteries. Regions-of-interest (ROIs), selected from histology, comprised 64 collagen-rich, 24 fibrolipidic, 23 calcified and 37 calcified-necrotic regions. A novel quantitative method was used to correlate IVUS data with corresponding histologic sections. Periodograms of IVUS samples, identified for each ROI, were used to calculate spectral parameters. Statistical classification trees (CT) were computed with 75% of the data for plaque characterization. The remaining data were used to assess the accuracy of the CTs. The overall accuracies for normalized spectra with CPSD, WPSD and MPSD were, respectively, 84.7%, 85.6% and 81.1% (training data) and 54.1%, 64.9% and 37.8% (test data). These numbers were improved to 89.2%, 91.9% and 89.2% (training) and 62.2%, 73% and 59.5% (test) when the calcified and calcified-necrotic regions were combined for analysis. Most CTs misclassified a few fibrolipidic regions as collagen, which is histologically acceptable, and the unnormalized and normalized spectra results were similar. (E-mail: [email protected])

Section snippets

Introduction and literature

Angiography, the traditional method for assessing atherosclerosis, has significant limitations. It allows evaluation of the geometry of the unobstructed part of the lumen only; it cannot provide information on the structure of the arterial wall, which is essential to estimating the extent of atherosclerosis. Intravascular ultrasound (IVUS) is becoming accepted as an imaging modality that allows precise cross-sectional assessment of the coronary anatomy in vivo Gussenhoven et al 1989b, Potkin et

Specimens and preparation

A total of 32 human hearts were obtained at autopsy, from subjects between the ages of 40 and 70 years. The left anterior descending (LAD) coronary artery was dissected from the ostium to the apex, including approximately 40 mm of surrounding fat and myocardial tissue. Inclusion of surrounding tissue ensured the maintenance of proper vessel support and reduced artefactual US reflections from the tissue-air interface. The LAD was harvested because it most often contains significant disease and

One-way ANOVA

The one-way ANOVA tested for significance of separation of the four plaque types. Due to the presence of calcium in both the calcified and the calcified-necrotic regions, there was no significant difference in mean values of the spectral parameters of these two plaque components. Hence, these two were combined as one atherosclerotic component for the ANOVA analysis. The number of ROIs from this category was 60 (23 calcium + 37 calcified-necrosis).

The unnormalized data were analyzed with 14

Discussion

This study aimed at assessing two factors that may have a significant effect on the analysis of IVUS RF spectral signature, and employing a novel approach to achieve accurate IVUS-histology correlation. These factors were: 1. analysis of normalized and un-normalized spectra; and 2. performance of three PSD algorithms toward classification of atherosclerotic tissue with IVUS.

The one-way ANOVA results among the three spectral algorithms were similar for both the unnormalized and normalized

Summary

This study aimed at comparing the ability of different spectral estimators for characterizing atherosclerosis with IVUS backscatter, by constructing classification trees. The approach employed various spectral parameters at the same time to provide the classification. These parameters were calculated with different spectral estimators in an attempt to achieve high spectral resolution. Results demonstrated that WPSD and CPSD performed equally well. The best accuracies for classification of

Acknowledgements

This project was funded in part by a Biomedical Engineering Grant awarded by the Whitaker Foundation (97 to 0332; D. G. Vince, Principal Investigator). The authors thank Jon Klingensmith for use of the IVUS image reconstruction software. Azita Tajaddini helped with data collection, and Dominik Meier was an invaluable source of help with histology image registration and morphing. The authors thank Christine Kassuba for editorial assistance.

References (47)

  • T Spencer et al.

    Characterisation of atherosclerotic plaque by spectral analysis of intravascular ultrasoundAn in vitro methodology

    Ultrasound Med Biol

    (1997)
  • A.F.W van der Steen et al.

    Correlation of histology and acoustic parameters of liver tissue on a microscopic scale

    Ultrasound Med Biol

    (1994)
  • D.G Vince et al.

    Comparison of texture analysis methods for the characterization of coronary plaques in intravascular ultrasound images

    Comput Med Imaging Graphics

    (2000)
  • R.J Watson et al.

    Classification of arterial plaque by spectral analysis of in vitro radio frequency intravascular ultrasound data

    Ultrasound Med Biol

    (2000)
  • L.S Wilson et al.

    Preliminary results from attenuation-slope mapping of plaque using intravascular ultrasound

    Ultrasound Med Biol

    (1994)
  • F.L Bookstein

    Principal warpsThin-plate splines and the decomposition of deformations

    IEEE Trans Pattern Anal Mach Intell

    (1989)
  • F.L Bookstein

    Morphometric tools for landmark dataGeometry and biology

    (1997)
  • L Breiman et al.

    Classification, and regression trees

    (1993)
  • Clark LA, Pregibon D. Tree-based models. In: Statistical sciences, S-Plus guide to statistical and mathematical...
  • C.L de Korte et al.

    Characterization of plaque components with intravascular ultrasound elastography in human femoral and coronary arteries in vitro

    Circulation

    (2000)
  • K.J Dixon et al.

    Characterization of coronary plaque in intravascular ultrasound using histological correlation

    Annu Int Conf IEEE Eng Med Biol Proc

    (1997)
  • F Esposito et al.

    A comparative analysis of methods for pruning decision trees

    IEEE Trans Pattern Anal Machine Intell

    (1997)
  • D.J Hand

    Construction, and assessment of classification rules

    (1997)
  • Cited by (0)

    View full text