The feasibility of using triple photon absorptiometry (TPA) for the measurement of bone mineral mass about a hip prosthesis was examined. A theoretical expression describing the variance of TPA measurements was verified using a triple photon source and phantom materials which simulate the soft tissue-bone mineral-metal prosthesis system. The expression for the variance was used to determine an optimized set of photon energies. It was shown that a precision of 3% could be obtained for reasonable measurement times using this optimized set of energies, and that TPA should be a feasible approach for measurement of bone mineral about a hip prosthesis.
Scattered and transmitted spectra were measured for 103 keV photons using solutions of K2HPO4. The ratios of both the number of coherently scattered photons to the number of transmitted photons and the number of coherently scattered to Compton scattered photons were measured. The sensitivity of the coherent/transmission ratio to change in effective atomic number was almost twice that of the coherent/Compton ratio. The coherent/transmission ratio was measured in vitro for three human calcanei submerged first in water and then in corn oil to simulate bone marrow. The reproducibility of the bone-in-water measurements was 3–7%. The standard deviations for the calcaneal measurements in water and in corn oil were similar. A clinical technique based on the coherent/transmission ratio could provide a sensitive method for the measurement of trabecular bone mineral concentration.
The sensitivity of a technique for the measurement of trabecular bone mineral concentration has been examined theoretically and experimentally. The technique is based on coherent gamma ray scattering and corrections for attenuation are obtained from transmitted photons rather than Compton scattered photons. For an incident photon energy of 60 keV, the minimum detectable bone mineral difference is practically independent of scattering angle while for an incident energy of 100 or 122 keV the scattering angle must be less than 70 degrees to optimize the minimum detectable difference.
