The use of tonebursts as an alternative to broadband signals in themeasurement of speed of sound in human cancellous bone

Abstract

Speed of sound (SOS) measurements, typically made using 1 MHz broadbandpulses, are increasingly used in the clinical diagnosis of bone disorders.Previous in vitro studies indicate that broadband ultrasound pulsesare susceptible to distortion in cancellous bone, leading to imprecise arrivaltime and SOS measurements. We investigated the effect of bandwidth andfrequency on SOS by comparing measurements made using 1 MHz broadband with1 MHz and 300 kHz narrowband toneburst signals in 15 human proximal femurcancellous bone specimens.There was no significant difference in the value of SOS measured from theleading edge of 1 MHz broadband, 1 MHz toneburst and 300 kHz toneburstsignals. Values of SOS in later regions of 1 MHz and 300 kHz tonebursts fellsignificantly (p<0.001) when compared to earlier regions. This decrease inSOS levelled off by the third complete cycle of 300 kHz toneburst signals,reaching a plateau value of 1961±239 m s-1. No plateau SOSvalue was obtained in1 MHz tonebursts. The reproducibility of SOS, as measuredby the coefficient of variation, was higher for later regions of 300 kHztonebursts than for the leading edge of 300 kHz toneburst and 1 MHz broadbandsignals (p<0.005). The correlation between ultrasound measured modulus andcompressive Young's modulus improved when 300 kHz tonebursts (r2 = 0.83)rather than 1 MHz broadband (r2 = 0.77) signals were used to calculate SOS.The improved SOS reproducibility of later regions 300 kHz tonebursts suggestthat it may be beneficial to use such signals rather than 1 MHz broadbandpulses in SOS measurement. Since no reliable SOS measurements could beobtained from any region of 1 MHz tonebursts, the use of high frequencytoneburst signals in cancellous bone has little value.