BONE QUANTITATIVE ULTRASOUND IMAGING Pascal Laugier Laboratoire d’Imagerie Paramétrique CNRS-Université Paris 6, 15 rue de l'école de Médecine, Paris, France firstname.lastname@example.org Introduction Technol ogical as pects and signal processing Osteoporosis, the most frequent bone disease, is a Although bone properties can be assessed with systemic skeletal disease characterized by low bone mass ultrasound at several skeletal sites, the calcaneus is the and micro -architectural deteriorat ion of bone tissue, with most popular measurement site and the majority of a consequent increase in bone fragility and a susceptibility clin ical reports has focused on this bone. The calcaneus is to fracture. Over the past 10 years, osteoporosis has easily accessible, and the medio-lateral surfaces are fairly emerged as a major clin ical challenge, related to its flat and parallel, thus potentially reducing measurement prevalence, the financial burden placed on public health errors. In addition, several epidemiological studies resources and the morbidity and mortality of associated reported that the calcaneus appeared to be among the fractures. About one third of post-menopausal wo men optimal BMD measurement site for routine screening of will have a fracture related to osteoporosis before the end perimenopausal women to predict the risk of any type of of their life. In the European commun ity, it contributes to osteoporotic fractures. The transverse transmission causing more than 400,000 hip fractures annually. technique uses two transducers, one acting as a transmitter Prevention of osteoporosis has now been recognized and the second one acting as a receiver. Both transducers as a major priority in research and health promotion. are placed on each side of the skeletal site to be tested Effective prevention and treatment programs require along its med iolateral axis (i.e. widthwise). The method quick, inexpensive diagnostic methods suited for wide- has been applied to the measurement of the slope of the spread use. For that, it is necessary to evaluate the factors frequency-dependent attenuation the so-called broadband of fracture risk, the most important of which is bone ultrasonic attenuation or BUA in dB/M Hz) and the speed strength. Bone density accounts for most of the strength of sound (SoS, m/s) at the calcaneus. of bone tissue, and today reference methods for skeletal Ultrasonic measurements are performed using a status assessment are based on X-ray absorptiometry substitution method, consisting in the comparison of the techniques measuring bone mineral density (BM D). Fourier spectrum of the signal transmitted through bone to However, bone mass is not sufficient to explain bone the spectrum of a reference signal transmitted through a strength. Bone microarch itecture which is not captured by reference mediu m of known attenuation, e.g. water. The conventional X-ray-based densitometry also contributes to frequency-dependent attenuation of ultrasound bone strength independently of bone mass. transmitted through the skeletal site is derived from the The alternative to X-ray introduced about fifteen ratio of the magnitude spectra. As the attenuation varies years ago is represented by quantitative ultrasound (QUS) linearly with frequency, the slope of a linear regression fit methods. As several bone properties contribute to the frequency-dependent attenuation between 0.2 and independently to bone strength, mult iparametric 0.6 MHz y ields the BUA value in d B/MHz. SoS is approaches combining different ultrasonic parameters obtained from the difference of time -of-flight of the reflecting independent bone properties, i.e. BMD and reference signal and the signal transmitted through bone. micro -architecture, may offer a mo re co mplete assessment Given the small amount of soft tissue, the influence of of bone status, improve the precision of d iagnosis and soft tissue on BUA and SoS is generally considered as favor the dissemination of the technique for osteoporosis being small and consequently neglected. and other bone related diseases. Both these items have Transmission measurements only partially exploit been addressed in our laboratory and a QUS imaging the information related to the interaction between the based technology has been developed to assess BMD and elastic wave and bone micro-arch itecture. It is known that bone microarch itecture using a single device. It uses the ultrasonic backscattered signals depend on the transmission and the reflection of low frequency scatterering med iu m micro-structure, in this case the bone ultrasound (approximately 500 kHz) through the trabecular network micro-architecture. Ultrasonic calcaneus (heel bone). backscatter may be useful for the measurement of specific micro -structural properties such as mean trabecular adjustment to experimental backscatter coefficient allows thickness or mean trabecular spacing. In addition, skeletal the assessment of mean trabecular thickness, a micro - sites, which are difficult to reach by transmission, could architectural parameter. be evaluated by such reflection techniques. Measurements Currently available imag ing devices measure the are performed by means of a single transducer used in heel immersed in a bath. We have recently demonstrated pulse-echo mode. The simplest method is a substitution the feasibility of QUS imaging at the wrist (Fig. 1) with one, in which the signal scattered from the region under reasonable precision. Waterless contact imaging would test is compared to the signal fro m a standard reflecting represent substantial technological progress toward target. The backscatter coefficient is obtained by dividing quantitative ultrasonic imaging at mu ltiple skeletal sites. the average backscatter power spectrum fro m the The development of QUS techniques to estimate bone scattering volume by the power spectrum co mputed for mineral losses during weightlessness of space flights, the calibration echo. The frequency-averaged backscatter taking into account the constraints applying to space coefficient between 0.2 to 0.6 MHz has been termed equipment, has been supported by the European Space Broadband Ultrasonic Backscatter (BUB). Agency, based on the technical characteristics of the Most devices that use fixed probes take the UBIS device, and has led to the development of an measurement in a single point without accurate control of innovative waterless imag ing device using 2D arrays of the transducers’ position with respect to the individual transducers. heel anatomy. Bones are irregularly shaped and have an inhomogeneous inner structure. For that reason, accuracy and precision are affected by the location within the bone Conclusion interrogated by the ultrasonic beam. The ultrasound bone imaging system (UBIS) developed for the first time in our Innovative QUS imaging based devices meet the laboratory for the calcaneus has the potential to evaluate need for bone densitometry to assist in the diagnosis of standardized regions of interest in all patients and to osteoporosis and in the treatment of osteoporotic patients. reproduce the analysis of the same reg ion of bone in It may become an acceptable alternative to existing X-ray series of scans acquired over a follow-up period of long based densitometry machines at reduced cost and with no duration (Fig.1). QUS images are performed by using a ionizing radiation, thereby contributing to improve overall mechanically driven pair of single -element focused quality of life of osteoporotic patients, i.e. reducing transducers immersed in a water bath (pixel size 1 mm2 , fractures incidence, saving life and avoiding pain, field of v iew 60x60 mm2 ). The acquisition of backscatter fractures and disability. radiofrequency signals has been implemented on the UBIS device. A mean value of QUS parameters (BUA, SoS, BUB) can be calculated by averaging the values REFERENCES within a region of interest (ROI) placed on the image. The size, form and location of the ROI can be adjusted by the Laugier P, Fournier B, Berger G. Ultrasound operator. However, the contribution of the observer to the parametric imag ing of the calcaneus: in v ivo results with a coefficient of variat ion is eliminated if an algorithm is new device. Calcified Tissue Int 58:326-331; 1996. utilized which selects the ROI co mp letely automatically. Rou x C, Fournier B, Laugier P, Chappard C, Ko lta S, In the calcaneus, a region of lowest attenuation can be Dougados M., Berger G. Broadband ultrasonic attenuation found in the greater tuberosity. This property is used by imaging: a new imag ing method in osteoporosis. J Bone imaging devices to place automatically a circular ROI on Miner Res 11:1112-1118; 1996. images. Clinical studies using imaging instru ments have Laugier P, Droin P, Laval-Jeantet AM, Berger G. In shown that the use of the automatic ROI imp roves the in vitro assessment of the relationship between acoustic vivo precision. properties and bone mass density of the calcaneus by comparison of ultrasound parametric imaging and QCT. Bone 20:157-165; 1997. Experi mental and Clinical as pects Chappard C, Laugier P, Fournier B, Rou x C, Berger G. Assessment of the relationship between broadband Recent findings confirm QUS imag ing diagnostic ultrasound attenuation and bone mineral density at the sensitivity and clinical utility for the assessment of calcaneus using BUA imag ing and DXA. Osteoporosis Int fracture risk. The technique precision of QUS imag ing is 7:316-322; 1997. comparable to the precision of X-ray densitometry Fournier B, Chappard C, Rou x C, Berger G, Laugier techniques. Normal bone demonstrates higher attenuation, P. Quantitative ultrasoun imaging at the calcaneus using and is associated with greater velocity compared with an automatic region of interest. Osteoporosis Int 7:363- osteoporotic bone. As for X-ray absorptiometry, the 369; 1997. influence of age was observed with a more pronounced Laugier P, Novikov V, Elman-Larsen B, Berger G. decrease immediately after menopause. Various Quantitative ultrasound imaging of the calcaneus: short- retrospective studies have reported discriminating power term and long-term precision and longitudinal variations of QUS between osteoporotic and age matched control during bed rest. Calcif Tissue Int, 66(1):16-21; 2000 similar to other X-ray densitometry techniques. The Rou x C., Roberjot V., Porcher R., Kolta S., relationship between BUB and fracture risk has been Dougados M., Laugier P. Ult rasonic backscatter and demonstrated recently. Furthermore, theoretical models of transmission parameters at the os calcis in post- ultrasound backscatter have been developed and their menopausal osteoporosis J Bone Miner Res 16, 1353- 1362, 2001. Defontaine M, Certon D, Colin L, Yvon C, Vince P, Lacaze E, Dufait R, Camus E, Laugier P, Patat F. A prototype of a 500 kHz ultrasonic matricial device : beam scanner. Application to in vivo heel bone quantitative characterizat ion. IEEE 1999 Ult rasonics Symposium proceedings, Oct 17-20, 1999, Tahoe, Nevada (USA). Eds SC Schneider, M Levy, BR McAvoy, IEEE, Piscataway, pp. 1585-1588. Padilla F, Peyrin F, Laugier P. Pred iction of backscatter coefficient in trabecular bones using a numerical model of 3D microstructure. Submitted to J Acoust Soc Am (2001). Jenson F., Pad illa F., Laugier P. Estimation of the mean trabecular thickness from backscatter measurements in cancellous bone. Submitted to Ult rasound Med Bio l (2002) Fig. 1: UBIS device developed and commercialized by Diagnostic Medical Systems (Montpellier, France); BUA Fig. 2: BUA image of the calcaneus (left ) and X-ray image of the foot (right) (same subject) BUA BMD Fig. 3: BUA (left) and BM D (right) image of the wrist (same subject).