Comparative methods for quantifying thyroid volume using planar imaging and SPECT.

UNLABELLED: SPECT enables improved accuracy over planar imaging in the determination of the volume since it is derived from three-dimensional data rather than from a two-dimensional projection with a certain geometric assumption regarding the thyroid configuration. METHODS: By using the phantoms of known volume, it was possible to estimate the accuracy of three different methods of determining thyroid volume from planar imaging used in clinical routine: the standard method used at Malmö General Hospital; a modified version of this standard method; and the method used in Lund Hospital in combination with different ways of defining the regions of interest (ROIs), and to assess the accuracy of the adaptive threshold of gray level histogram method based on SPECT imaging which determines a threshold that maximizes the separability of two classes (object and surround). RESULTS: The correlation coefficient (r) and the regression equation between the true (x) and the calculated volume (v) were as follows: r = 0.99 and y = 0.98x + 3.6 using SPECT and the gray level histogram method for edge detection combined with attenuation and scatter corrections, while r = .97 and y = 0.67x + 3.2 using the standard method based on planar scintigraphy. The standard method as used in routine was found to produce large errors (24.8%). The error on the volume estimate was reduced to approximately 7% for volumes in the range 16-to-75 ml using SPECT. CONCLUSION: Compared with conventional scintigraphy, thyroid phantom volumes were most accurately determined with SPECT when attenuation and scatter corrections are performed, which allows accurate radiation dosimetry in humans without the need for assumptions on organ size or concentration.