In metrology, the use of the concepts of decision limit and detection limit often poses many problems for metrologists in radioactivity analysis laboratories. It is usual to censor data when it becomes difficult to discern the presence or absence of activity, due to noise in the measurement data. This means that if the measurement results are insignificant (below the decision threshold), the analysis simply indicates that the actual value (signal) of radioactivity is below a certain limit called the detection limit. These problems are often due to a misunderstanding of the decision threshold formulas. In addition, it is not clear how to generate an appropriate and justified decision threshold in ISO (2020). In this research paper, we develop a statistical method for determining the most powerful decision threshold. Next, methods of statistical approaches are adopted to estimate the density, expectation and variance of radioactivity. The effectiveness and feasibility of these approaches are corroborated by applications on IRSN data sets.