NAA has relative freedom from matrix and interference effects. For larger samples it may necessary to calculate wit the neutron shelf shileding. For biological samples the limit of detection is better han geological matrix.
An useful report: http://www-pub.iaea.org/MTCD/publications/PDF/te_1215_prn.pdf
Neutron Activaltion Analysis is very sensitive and highly acurate technique. No matter the type of the sample. the difference of samples kind, needs only different procedures for preparing them for irradiation and a little in calculation.
NAA works best where the matrix does not activate under neutron irradiation (or have a much shorter half-life than the elements to be analysed), so C, H, O, N, Si based matrices are very ideal. The presence of certain elements (Na, Mn, Cd, Gd) can seriously increase the detection limits of others or just render NAA as unusable. Of the examples given here, plant material and (powdered) milk are generally ideal matrices but blood is more challenging due to its relatively high sodium content. (Radiochemical NAA where Na is separated from the sample post-irradiation is often used with blood samples to decrease detection limits)