[en] This paper is a review of current data on the risks associated with human exposure to ionizing radiation. We examine these risks for dose levels ranging from very high (atomic bomb survivors) to very low (background). The principal end point considered is cancer mortality. Cancer is the only observed clinical manifestation of radiation-induced stochastic effects. Stochastic effects are caused by subtle radiation-induced cellular changes (DNA mutations) that are random in nature and have no threshold dose (assuming less than perfect repair). The probability of such effects increases with dose, but the severity does not. The time required for cancer to develop ranges from several years for leukemia to decades for solid tumors. In addition to somatic cells, radiation can also damage germ cells (ova and sperm) to produce hereditary effects, which are also classified as stochastic. However, clinical manifestations of such effects have not been observed in humans at a statistically significant level

No abstract available

[en] The classical statistics approach used in health physics for the interpretation of measurements is deficient in that it does not allow for the consideration of needle in a haystack effects, where events that are rare in a population are being detected. In fact, this is often the case in health physics measurements, and the false positive fraction is often very large using the prescriptions of classical statistics. Bayesian statistics provides an objective methodology to ensure acceptably small false positive fractions. The authors present the basic methodology and a heuristic discussion. Examples are given using numerically generated and real bioassay data (Tritium). Various analytical models are used to fit the prior probability distribution, in order to test the sensitivity to choice of model. Parametric studies show that the normalized Bayesian decision level k_α-L_c/σ₀, where σ₀ is the measurement uncertainty for zero true amount, is usually in the range from 3 to 5 depending on the true positive rate. Four times σ₀ rather than approximately two times σ₀, as in classical statistics, would often seem a better choice for the decision level

[en] The problem of choosing a prior distribution for the Bayesian interpretation of measurements (specifically internal dosimetry measurements) is considered using a theoretical analysis and by examining historical tritium and plutonium urine bioassay data from Los Alamos. Two models for the prior probability distribution are proposed: (1) the log-normal distribution, when there is some additional information to determine the scale of the true result, and (2) the 'alpha' distribution (a simplified variant of the gamma distribution) when there is not. These models have been incorporated into version 3 of the Bayesian internal dosimetric code in use at Los Alamos (downloadable from our web site). Plutonium internal dosimetry at Los Alamos is now being done using prior probability distribution parameters determined self-consistently from population averages of Los Alamos data. (author)

No abstract available

No abstract available

No abstract available

[en] Over the course of several years of measuring in-vivo spectra (whole-body counts) in search of transuranic radionuclides, it has been observed that, relative to the count rate in a region around 80 keV (defined as region 3), large persons have relatively higher count rates than thin persons at around 60 keV (region 2, where an ²⁴¹Am line is expected), and relatively lower count rates at around 17 keV (region 1, where both ²³⁹Pu and ²⁴¹Am lines are expected). The observed data can be understood in terms of relative amounts of scattering and absorption of the 1.461-MeV photon from ⁴⁰K, which occurs naturally in the human body. For larger persons, increased scattering causes the Compton peak to shift to lower energies, thereby increasing the count rate in region 2 relative to region 3. Also, because of increased absorption of very low energy photons, the count rate in region 1 decreases relative to region 3. To test this hypothesis, we compute the Spectrum of photons emerging from cylindrical human phantoms of various dimensions, assuming a variety of distributions of ⁴⁰K within the phantom. The calculations are carried out using the Monte Carlo transport code MCNP. The results of these calculations qualitatively agree with the observations and support our hypothesis

[en] The classical statistics approach used in health physics for the interpretation of measurements is deficient in that it does not take into account needle in a haystack effects, that is, correct identification of events that are rare in a population. This is often the case in health physics measurements, and the false positive fraction (the fraction of results measuring positive that are actually zero) is often very large using the prescriptions of classical statistics. Bayesian statistics provides a methodology to minimize the number of incorrect decisions (wrong calls): false positives and false negatives. The authors present the basic method and a heuristic discussion. Examples are given using numerically generated and real bioassay data for tritium. Various analytical models are used to fit the prior probability distribution in order to test the sensitivity to choice of model. Parametric studies show that for typical situations involving rare events the normalized Bayesian decision level k_α = L_c/σ₀, where σ₀ is the measurement uncertainty for zero true amount, is in the range of 3 to 5 depending on the true positive rate. Four times σ₀ rather than approximately two times σ₀, as in classical statistics, would seem a better choice for the decision level in these situations

[en] Background radiation, including cosmic rays, can produce a hundred or more errors per day in a modern computer facility. Previous calculations indicated that, ''near sea level the neutrons and muons become important,'' in producing errors, and this has been the generally accepted position in the field. Pions, which are about ten thousand times less abundant than muons, have usually been ignored. Electronic components are now in the micrometer range and available microdosimetric spectra indicate that pions may be at least equally responsible for damage in electronic circuits as compared with muons. Initial Monte Carlo calculations for a typical geometry of a 4K RAM give an upper limit for errors per pion per cm/sup 2/ for 164 MeV/c pions of about 7 x 10/sup -6/. Present results are compared with earlier calculations and recent measurements of soft errors