Activity optimization method for nuclear medicine planar studies based on discriminant analysis: a comparison with ROC Curve

ABSTRACT

A method for optimizing the administered activity in Static Nuclear Medicine Studies is validated by comparison with ROC curve. Discriminant analysis of image quality is the applied statistical technique. The constructed linear function owns as dependent parameters, the differentiated levels of image quality obtained by observer's criterion. The percentage of cases correctly classified is analyzed. The minimum value of the administered activity, which permit good image quality, can be proposed from the best results of the parameters selected by the function. The method was tested using planar images from a Jaszczak phantom, acquired under four activities (1088 MBq, 962 MBq, 740 MBq and 562 MBq). Two of the six measured Background-to-Signal ratios (S/B) were the parameters able to construct the function with high correlation respect to the expert criterion. The value of 740 MBq was the optimum after method application. The results were coincident with the application of ROC-analysis.