Measurement of ultrasound wave-speed in bone tissue can help assess bone mechanical strength and the risk of osteoporotic fracture. Several dedicated ultrasound devices have been proposed to measure the ultrasonic wavespeed in the cortex of different skeletal sites [1]. Bone exhibits anisotropic properties, allowing for the extraction of independent measures indicating its mechanical characteristics.

[2] introduced the first joint in vivo assessment of wave-speed in all bone directions using ultrasound imaging of bone. This method was initially tested on two healthy volunteers [2], yet there is currently no study on the precision of this method. This study presents the first in vivo evaluation of wave-speed precision in bone tissues using this technique. The aim is to determine whether it can detect significant changes in bone material properties using the estimated wave-speed and anisotropy form parameter. We provide wave-speed values in both radial and axial directions of the human tibia, along with an anisotropy form parameter.

Measurements were taken at the middle and proximal third of the tibia of 11 healthy male volunteeers with a mean age of 28 (2.5) years. Transverse and longitudinal ultrasound acquisitions were conducted using a singleelement transmission synthetic aperture imaging sequence with a research ultrasound system and a 2.5 MHz phased array transducer. Each acquisition was repeated five times with repositioning, guided by real-time visualization to ensure accurate image alignment.

The wave-speed in cortical bone is estimated using a weak transverse isotropy model with parameters including wave-speed in the direction of the material symmetry axis (V axial ) and in a direction normal to the material symmetry axis (V radial ), and an anisotropy form parameter χ [2,3] V radial and χ were estimated using an autofocus approach, while V axial was estimated as the headwave velocity. V radial is estimated with transverse acquisitions while V axial and χ are estimated with longitudinal acquisitions. Precision is assessed by calculating the root-mean-square (RMS) average [4] of the standard deviation (SD) from five measurements, expressed as a percentage relative to the mean group value.

The figures present wave-speed in the axial and radial directions and anisotropy form parameter (bottom) and the table gives precision on each parameter with 95% confidence intervals (top). Precision is less than 3% for V radial and less than 1.5% for V axial . With cortical porosity ranging from 2 to 25%, radial wave-speed can decline by up to 15% in total [5,6]. Given the precision of less than 3% found for V radial in this study, ultrasound imaging is able to discern multiple levels of cortical porosity. Furthermore, during adulthood, V axial declines on average by 5% [7]. Hence, given the precision of less than 1.3% found for V axial in this study, ultrasound imaging might be able to monitor longitudinal cortical porosity change.

In conclusion, our findings suggest that the precision with which the radial and axial wave-speeds are assessed is sufficient to identify and monitor individuals with low cortical bone mechanical quality at risk of osteoporotic fracture. Future work will evaluate precision in a clinical context.