Absolute organ activity estimated by five different methods of background correction

Accurate absorbed dose estimates in radionuclide therapy require patient-sp ecific dosimetry. In patient-based dosimetry, estimation of absolute organ uptake is essential. The methods used should be reasonably accurate as well as easy to perform in routine clinical practice. One of the major sources of uncertainty in quantification of organ or tumor activity from planar ima ges is the activity present in the tissue surrounding the source. Methods: To estimate organ activity as a function of organ-to-background activity co ncentration ratio, a cylindrical phantom, filled with 5.6 liters of water w as used to simulate the abdomen of a patient. Two other cylinders of 150 ml each, representing the kidneys, were each filled with 19 MBq Tc-99m and we re positioned in the abdomen phantom. The phantom was imaged with a dual-he ad gamma camera with the kidneys placed at posterior depths of 1-, 5- and 1 0-cm at kidney-to-background activity concentration ratios of infinity, 10: 1, 5:1 and 2:1. The conjugate view geometric mean counting method was used to quantify activity. Five methods for background correction were applied: (1) no correction; (2) conventional background correction (simple subtracti on of the background counting rate from the source region counting rate); ( 3) Kojima method (background corrected for organ thickness and depth); (4) Thomas method (analytical solution); and (5) Buijs method (background corre cted for organ and total-body thickness). Results: Since the results were i dentical for both kidneys, only the left kidney activity measurements are p resented. The accuracy of the five background correction methods is given a s the percentage difference between the actual and measured activity in the left kidney. For Method 1,the percentage difference ranged from -2% with a n infinite kidney-to-background activity concentration ratio to +413% with a 2:1 ratio. For Method 2, these values ranged from -1% to -80%, for Method 3 from +11% to -18%, for Method 4 from -2% to +120% and for Method 5 from -4% to +39%. Conclusion: Even though quantitative SPECT is the most rigorou s method for activity quantification in conditions of low organ-to-backgrou nd activity concentration ratio, planar scintigraphy can be applied accurat ely if appropriate attention is paid to background correction. Using relati vely simple background subtraction methods, the quantitative planar imaging technique can result in reasonably accurate activity estimates (Methods 3 and 5). The use of Kojima's method is preferable, especially at very low so urce-to-background activity concentration ratios.