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

To test whether radio-frequency analysis of coronary plaques predicts the h istological classification, r.f. data were collected using a 30 MHz intrava scular ultrasound scanner. Two hundred ninety-nine regions-of-interest from eight postmortem coronary arteries were selected and identified by histolo gy as falling into one of seven different tissue types. These are loose fib rous tissue (n = 78), moderate fibrous tissue (n = 27), dense fibrous tissu e (n = 33), microcalcification (n = 14), calcified plaque (n = 55), lipid/f ibrous mixture (n = 51) and homogeneous areas of lipid pool (n = 29). On th e basis of a previous study, four spectral parameters were calculated for e ach of the regions-of-interest: maximum power (dB), mean power (dB), spectr al slope (dB/MHz) over the bandwidth 18-35 MHz and the intercept of the spe ctral slope with the 0 Hz axis (dB). A minimum-distance classifier using th e Mahalanobis (1948) distance was applied to the data. Following resubstitu tion of the training data into the classifier, the total correctly classifi ed was 54 %. The data were reclassified using three broader tissue groups: (1) calcified plaque, (2) lipid pool and (3) a mixed fibrous category, inco rporating loose fibrous tissue, moderate fibrous tissue, dense fibrous tiss ue, lipid/fibrous mixture and microcalcification. The total correctly class ified was 86 %. Using "leave-one-out' cross-validation, the classification rates were 48 % for seven tissue subgroups and 83 % for three broader categ ories of tissue type. (C) 2000 World Federation for Ultrasound in Medicine & Biology.