Time domain analysis of ultrasonic backscatter signals from cancellous bone

Abstract

Ultrasonic backscatter techniques are being developed to detect changes in bone caused by osteoporosis. Most techniques analyze backscatter signals in the frequency domain by measuring quantities related to the power spectrum. Investigate the utility of two backscatter parameters determined from a time domain analysis of backscatter signals: the normalized backscatter amplitude ratio (nBAR) and the backscatter amplitude decay constant (BADC). A 3.5 MHz transducer was used to acquire backscatter signals from 54 specimens of bone prepared from 14 human femurs. nBAR was determined from the log of the ratio of the root mean square amplitude of two different portions of a backscatter signal. BADC was determined by measuring the exponential decay in the amplitude of a backscatter signal. nBAR and BADC both demonstrated highly significant (p < 0.001) linear correlations with bone density. However, the correlation coefficients were slightly stronger for nBAR (0.79 ≤ R ≤ 0.89) than for BADC (0.67 ≤ R ≤ 0.73). Parameters based on a time domain analysis of backscatter signals from bone may be sensitive to changes in bone caused by osteoporosis. Of the two parameters tested, nBAR demonstrated the strongest correlations with bone density. [Funding: NIH/NIAMS R15AR066900.]