Figure 4: (a) Percentage difference between observed and calculated signals versus gadolinium concentration for a sample experimental run (SPGR TR 40 ms/TE 10 ms) for varying values of FA: 10°, 30°, 60°, and 90°. There is a systematic variation in percentage difference (PD) with increasing flip angle, with progressively decreasing ratios of observed to predicted signal. For a TR of 40 ms, the accuracyoptimal FA is 30°. At this flip angle, observed signal is only slightly greater than predicted signal. (b) Percentage difference between observed and calculated signals versus gadolinium concentration for another sample experimental run (SPGR TR 100 ms/TE 10 ms) for varying values of FA: 10°, 30°, 60°, and 90°. The same pattern of variation is seen as in (a), except that for a TR of 100 ms, the accuracyoptimal FA is 60°. 
