Differences in MRI-measured contrast enhancement have been used for tissue characterization, particularly for the characterization of mammary tumours. T1 weighted spoiled gradient echo sequences have usually been acquired for this purpose and relative signal intensity increase (Srel) has been determined to quantify contrast uptake. The field strength dependence of this technique is evaluated in this paper by phantom measurements. Srel is compared with the recently introduced "enhancement factor" (EF) and "contrast uptake equivalent" (CE) as a quantitative assessment parameter for contrast uptake. Enhancing tissues were simulated with water phantoms doped with increasing concentrations of Gd-DTPA. Spoiled gradients echo images (FLASH-2D) were obtained on a 0.2 T and 1.5 T MR system. Srel, EF and CE were determined and the results of the phantom measurements were applied to typical types of breast tissues. Srel showed a strong dependence on the magnetic field strength and was greater at higher field strengths. This is because Srel correlates positively with the native T1 which is longer at higher field strengths. Conversely, the EF and CE almost eliminated the distorting influences of the magnetic field strength. CE provided a good approximation of the actual contrast uptake. In practice, the dependence of Srel on the magnetic field strength might cause malignant lesions to be misclassified as benign at low field strengths. Either Srel criteria should be adjusted to magnetic field strength or EF and CE should be used instead of Srel for tissue characterization with dynamic contrast-enhanced MRI.