In our recent work we developed a reconstruction approach for cardiac gated dynamic images based on B-spline modeling. The goal is to obtain a single sequence showing simultaneously both cardiac motion and kinetic tracer distribution change over time from one acquisition. In this work, we further develop and demonstrate the feasibility of this procedure for perfusion defect detection in gated dynamic images. We conduct Fisher's linear discriminant analysis and principal component analysis on the reconstructed dynamic image activities, and derive quantitative measures to differentiate defects from normal perfusion. The proposed development is demonstrated using a dynamic version of the 4D NURBS-based cardiac-torso (NCAT) phantom to simulate a gated SPECT perfusion acquisition with Tc99m Teboroxime.