Autocorrelation length of normal and myopathic human myocardium measured by acoustic microscopy: implications for clinical ultrasonic tissue characterization

Ultrasound Med Biol. 1990;16(8):793-9. doi: 10.1016/0301-5629(90)90043-c.

Abstract

The pattern of acoustic speckle present in ultrasonic images of tissue may be affected by pathological changes in the tissue. To investigate the possibility for such an effect in dilated cardiomyopathy, we examined acoustic micrographs of frozen sections of normal (autopsy) and myopathic (transplant recipient) human myocardium. The resulting images were digitized, and the major axis autocorrelation length calculated for each image. For the normal specimens (n = 75, four patients), the mean autocorrelation length was 37 +/- 18 microns, with the maximum value of 92 microns. The myopathic specimens (n = 64, four patients) had a mean of 52 +/- 19 microns, with a maximum of 100 microns. We conclude that the changes in structure between normal and myopathic myocardium are significant (p less than 0.001), but on so small a scale in relation to the wave length of clinical ultrasound that there will only be changes in the intensity of the backscattered signal, not the pattern of speckle.

MeSH terms

  • Acoustics
  • Cardiomyopathies / diagnostic imaging*
  • Echocardiography*
  • Humans
  • Image Processing, Computer-Assisted
  • Mathematics
  • Microscopy
  • Myocardium / ultrastructure*
  • Photography