Diagnostic accuracy of a novel ultrasound-based methodology for spinal densitometry on a cohort of normal- and under-weight women

Currently, osteoporosis is mainly diagnosed through dual-energy X-ray absorptiometry (DXA). However, DXA cannot be used for early diagnoses through population mass screenings because of issues related to ionizing radiation employment. This paper describes the diagnostic accuracy of a novel ultrasound (US)-based method to perform spinal densitometry without employing X-rays. The proposed innovative methodology is based on a combined analysis of both echographic images and "raw" radiofrequency US signals. The diagnostic output is represented by the same parameters of DXA (bone mineral density (BMD), T-score, Z-score). The actual effectiveness of the proposed methodology was evaluated on a cohort of 350 normal-weight or underweight (body mass index (BMI) < 25 kg/m2) female patients in the age range 45-65 years by a direct comparison with DXA assumed as gold standard. The accuracy of US-based diagnoses ranged from a maximum of 90.5% to a minimum of 74.1%, corresponding to the youngest and oldest patient age category, respectively, with an average of 84.9%. A good correlation was also found between US-estimated BMD and DXA related values (r=0.69, p<0.001). Obtained results demonstrated the high accuracy of the proposed US approach to spinal bone densitometry compared with DXA. This technique has the potential to become a useful and effective tool in clinical practice improving the current approach to osteoporosis diagnosis.