Abstrakt
Estimation of the effective half-life, i.e. the time during which the activity within a region of interest falls to half of its original value, is a frequent task in dosimetric/nuclear medicine applications. The estimation is usually based on the least-squares fit of a straight line in the semi-logarithmic coordinates.
Such a solution, even though susceptible to measurement errors, provides little information on the estimate reliability and neglects accessible prior information. The Bayesian solution presented here removes these drawbacks by respecting the well-justified probabilistic model and by deeper exploitation of the available information.
The positive contributions of the Bayesian solution are illustrated on real-life data related to diagnostics and therapy of thyroid diseases by I-131.