Relationship between ultrasound texture classification images and histology of atherosclerotic plaque.

Structure and content of atherosclerotic plaque varies between patients and may be indicative of their risk for embolisation. This study aimed to construct parametric images of B-scan texture and assess their potential for predicting plaque morphology. Sequential transverse in vitro scans of 10 carotid plaques, excised during endarterectomy, were compared with macrohistology maps of plaque content. Multidiscriminant analysis combined the output of 157 statistical and textural algorithms into five separate texture classes, displayed as ultrasound (US) texture classification images (UTCI). Visual comparison between corresponding UTCI and histology maps found the five texture classes matched with the location of fibrin, elastin, calcium, haemorrhage or lipid. However, histology preparation removes calcium and lipid and, so, can affect the structural integrity of atherosclerotic plaques. Soft tissue regions smaller than the UTCI kernel, (0.87 mm x 0.85 mm x 3.9 mm), such as blood clots, are also difficult to detect by UTCI. These factors demonstrate limitations in the use of histology as a "gold standard" for US tissue characterisation.

[1]  T. Noritomi,et al.  Carotid plaque typing by multiple-parameter ultrasonic tissue characterization. , 1997, Ultrasound in medicine & biology.

[2]  J Domjan,et al.  Ultrasonic carotid artery plaque structure and the risk of cerebral infarction on computed tomography. , 1994, Journal of vascular surgery.

[3]  R. Fisher THE USE OF MULTIPLE MEASUREMENTS IN TAXONOMIC PROBLEMS , 1936 .

[4]  R. F. Wagner,et al.  Statistics of Speckle in Ultrasound B-Scans , 1983, IEEE Transactions on Sonics and Ultrasonics.

[5]  P R Detmer,et al.  Echolucent regions in carotid plaque: preliminary analysis comparing three-dimensional histologic reconstructions to sonographic findings. , 1994, Ultrasound in medicine & biology.

[6]  Robert M. Haralick,et al.  Textural Features for Image Classification , 1973, IEEE Trans. Syst. Man Cybern..

[7]  F. Moll,et al.  Ultrasonographic characterization of carotid plaques. , 1998, Ultrasound in medicine & biology.

[8]  T. Elatrozy,et al.  The effect of B-mode ultrasonic image standardisation on the echodensity of symptomatic and asymptomatic carotid bifurcation plaques. , 1998, International angiology : a journal of the International Union of Angiology.

[9]  A new method for studying plaque morphology. , 1995, American journal of cardiac imaging.

[10]  E. Falk Morphologic features of unstable atherothrombotic plaques underlying acute coronary syndromes. , 1989, The American journal of cardiology.

[11]  Benoit B. Mandelbrot,et al.  Fractal Geometry of Nature , 1984 .

[12]  J C Bamber,et al.  Compensation for the signal processing characteristics of ultrasound B-mode scanners in adaptive speckle reduction. , 1993, Ultrasound in medicine & biology.

[13]  M P Moore,et al.  Classification of arterial plaque by spectral analysis of in vitro radio frequency intravascular ultrasound data. , 2000, Ultrasound in medicine & biology.

[14]  E J Feleppa,et al.  Determination of carotid plaque risk by ultrasonic tissue characterization. , 1998, Ultrasound in medicine & biology.

[15]  F Dunn,et al.  Ultrasonic propagation through fixed and unfixed tissues. , 1979, Ultrasound in medicine & biology.

[16]  J. Wilhjelm,et al.  Echo-lucency of computerized ultrasound images of carotid atherosclerotic plaques are associated with increased levels of triglyceride-rich lipoproteins as well as increased plaque lipid content. , 1998, Circulation.

[17]  M. J. Noruésis,et al.  SPSS-X advanced statistics guide , 1985 .