Minimum Detectable Activity

["Patrocinio, Horacio J." <patrocih@ERE.UMontreal.CA>]

Hello fellow RADSEFERS,

I came across an equation for calculating MDA based on statistical 
analysis that states MDA = (4.65 x N^0.5 + 3) / ET where N is the number 
of counts for the background in time T and E is the sensitivity of the 
detector in cps/unit activity. Our specific problem regards leak testing 
of selaed sources and this equation is suggested by the AECB (CANADA).

I understand how the MDA is based on a given confidence level assuming 
Poisson statistics but cant seem to see (or find) a good derivation of 
this equation.  Cember (1996) presents this equation without a thorough 
derivation.

Please let me know if there is a better reference.


Horacio J. Patrocinio
Hotel-Dieu de Montreal Hospital
Montreal, Canada