Purpose: The purpose of this study was to develop a new diagnostic technique for measuring bone mineral density (BMD) for the assessment of osteoporosis, which improves upon the coherent to Compton scattering ratio (CCSR) method, which was first developed in the 1980s. To help the authors achieve these goals, they have identified and studied two new indices for CCSR, the forward scattered to backward scattered (FS-BS) and the forward scattered to transmitted (FS-T) ratios. They believe that, at small angles, these two parameters can offer a practical in vivo determination of BMD that can be used to overcome the limitations of past CCSR systems, including high radiation dosages, costs, and examination durations. Methods: In previous CCSR studies, a high-activity radioactive source with a long half-live (usually 241 Am) and an expensive and bulky cryogenic HPGe detector were applied to both in vivo and in vitro measurements. To make this technique more suitable for clinical applications, the possibility of using a standard diagnostic x-ray tube generating a continuous spectrum was investigated in this paper. Scattered radiation from trabecular bone-simulating phantoms containing various mineral densities that span the normal range of in vivo BMD was collected in this study using relatively inexpensive noncryogenic CdTe or NaI detectors. Results: The initial results demonstrate that a modified version of CCSR can be successfully applied to trabecular bone assessment using a diagnostic x-ray tube with a continuous spectrum in two variations, the FS-BS and the FS-T ratio. When FS-BS is measured, intensity spectra in the forward and backward directions must be collected while FS-T requires only the integral intensity of the scattered and transmitted (T) spectra in the energy region above 40 keV. For both of these methods, forward scattering angles less than or equal to 15 and backward scattering angles greater than or equal to (165 = 180 À 15 ) are needed. Conclusions: The authors determined that FS-T is more sensitive to changes in BMD than transmission or absorption alone and that the FS-BS method can yield an absolute measurement of the mean atomic number of the scattering medium, after a correction for path-dependent attenuation. Since this study determined that the FS-T ratio is independent of the incident energy over a broad energy region, it will be possible to apply FS-T to bone densitometry using inexpensive integral photon detectors. The authors believe that, by replacing the radionuclide source with an x-ray tube and the cryogenically cooled HPGe detector with a single solid state CdTe, NaI, or silicon detector or an annular array of detectors, as suggested in this study, the past difficulties of CCSR concerning high radiation exposure, costs, and durations as well as lack of convenience can be overcome and that CCSR could eventually become popular in clinical settings. V C 2012 American Association of Physicists in Medicine. [http://dx