90 60 h=4cm, FWHM =8.5cm 50 RELATIVE ACTIVITY ——— Cc UNKNOWN Cc KNOWN —-— REAL DISTRIBUTION DISTANCE (cm) Fie, 72.—Variations in f(z) caused by previous knowledge of C, A = 4 em Integrated Activity on fix) - FWHM = 8.5 cm interval an fix) (cm) C Unknown Experimental Value of the Activity + 3% c= 85 th-3 /h-z4 (wc) 0-44 helfth=«=2];h*3Jhr4-helth=2 0.9 0.85 1.0 0.7 0.9 09 1.0 ll 0.9 14-44 3.0 2,85 2.9 2.6 3.0 3.0 2.9 3.1 2.9 44-54 45 4.6 4.6 42 46 4.6 4.6 43 47 0-54 &4 8.3 85 7.5 8.5 8.5 8.5 85 &5 Fic. 73—Integrated f(x) for various h’s. activities over various intervals the integrated quantities and the percentage err« the three previously chosen intervals. The tot: grated activity C is practically independent FWHMof K(z,x). (See above for dependenc on A.) The results of Figures 74 and 76 suggest th choice of the “best”? collimator must be based following :* (a) of Integrated Activity on fix) interval on f(x) (cm) oi FWHM = 15,0 cm h=2em i 4-44 3.2 tas 42 ~22 ~10 Percent Error FWHM = iL0cm h=2cm 1.0 2.7 ~ 7 4.8 ~10 FWHM= 85cm h=2cm 0.9 3 ~0 -35 4.6 ~2 Experimental Vaiue of the Activity +3% (b) {uci} Fic. 74—~Integrated activity with three different point response functions. the smallest interval Az, where the net grated) activity / f(z) dx = C; should have acceptable error. (If Avg > 29 47 curacy in the location of the peaks and magnitude of their slope; in our case Av, ~ FWHM; ttAXgs 09 ~2 the smallest interval Ax, (larger than A) in the shape of the computed f(z) should re the actual distribution, namely acceptal (ce) this condition is automatically satisfied.) ; the efficiency of the collimator should be enough to minimize the statistical errors \ feasible time intervals of measurement. * These requirements are similar to those needed to i and measure radioactivity in various organs of the body.