Answer to Question #13807 Submitted to "Ask the Experts"

Our human body consists of approximately 65% oxygen and 9.5% hydrogen by means of mass. If we use water as a standard medium to simulate our human body for the purpose of dosimetric measurements in the medical field, then it will simulate our body for about 60% as an adult human body consists of only 60% water. Then how can we compensate for remaining 40% composition of our human body? (That is, we have approximately 18.5% of carbon, 3.3% of nitrogen and 4% of other metals). My doubt is that without including the effect of radiation interaction with those 40% of other components, we are going only with the effect of radiation interaction with water alone. Hence, there is an uncertainty in simulation of our human body by water. How can we correct this?

Your inference that water may not always be representative of the whole body may be legitimate, depending on the conditions and requirements associated with the dose estimation that are being made. You note that you are concerned with the medical field, but you don't cite the specific exposure situations. If your concern is with doses to a patient from diagnostic medicine procedures—e.g., diagnostic x rays or diagnostic nuclear medicine procedures, or doses to attending personnel, the acceptable uncertainties in estimated doses may be quite large, possibly tens of percent. However, if you are concerned with doses to patients undergoing therapeutic radiation treatments, as I shall assume you are, acceptable dose uncertainties are typically much smaller, likely on the order of a few percent.

It is important to note that while the average human adult body contains, by mass, approximately 60% water, approximately 30% to 37% of the remaining mass consists of tissue containing relatively low atomic number materials, primarily C, H, N, O, associated with soft tissues in the body—e.g., muscle and fat. Considering this, there is a legitimate rationale for using water to simulate soft tissues in the body because the important radiological parameters that relate to radiation attenuation and energy absorption, which ultimately impact doses to selected tissues, are often not very different for water compared to such soft tissues. For example, if we review values of the mass attenuation coefficients and mass energy absorption coefficients in soft tissue (ICRU-44 composition) for ionizing photons that range in energy from 1 MeV to 20 MeV, we find that both the attenuation coefficients and the energy absorption coefficients for the two media differ by less than 1% at 1 MeV and slowly changing in a monotonic fashion such that the difference is less than 1.5% at 20 MeV. You can find values for these quantities at this National Institute for Standards and Technology (NIST) site. Resultant dose estimations would differ by commensurate amounts for water compared to soft tissue.

If your concern was for estimated dose to bone, however, or if you were attempting to account for attenuation through bone tissue, the substitution of water for bone may not be appropriate because the composition of bone is considerably different from soft tissue or water. In particular, the physically important parameters for photon irradiations. mass density, mass attenuation coefficients, and mass energy absorption coefficients for bone differ appreciably from those for water or soft tissue. The mass density of cortical bone is almost two times greater than that for water, and the mass attenuation and energy absorption coefficients are about 7% lower for bone than for water at 1 MeV, about equal a bit beyond 6 MeV, and about 11% higher for bone than for water at 20 MeV. Such differences would likely produce unacceptably inaccurate estimates of dose if water were used to simulate bone.

While I will not proceed much further here, I should note that similar conclusions regarding the appropriateness of simulation of other tissues by water in dose estimations would also apply if one were concerned with electron beam radiation therapy. The important comparative dosimetric quantity in this situation is the mass collision stopping power associated with incident electrons; values may be obtained at this NIST site.

Of course, there are other considerations that apply in terms of factors affecting the actual dose, including photon and/or electron scattering effects that impact dose, but these are beyond the current discussion. I hope this has been helpful.

George Chabot, CHP, PhD