arious criteria are used to estimate TOF, for example the first arrival point,
the first zero-crossing point, or a fixed threshold on the rising front of the received electrical signal (see Fig. 3.2). Frequency-dependent attenuation and velocity
dispersion are acknowledged sources of bias when measuring velocity in the time domain [24, 35–38]. As the signal is distorted while propagating through bone, the
envelope of the received signal may differ considerably from the reference signal [38, 39]. Ambiguities in time-domain methods for velocity measurements in
cancellous bone were reported by several investigators [36–38]. For example, velocity variations from one zero crossing to the next one for a calcaneus with BUA
of 20 dB · cm−1 · MHz −1 can be of the order of 30m · s −1 [40], which is considerable compared to the difference between fractured and unfractured women [41].
This effect is a function of the frequency-dependent attenuation [42].
đang được dịch, vui lòng đợi..
