Medical image fusion is the method of combining or merging complementary information from two or more source images into a single image to enhance the diagnostic capability. In this work six different fusion rules are performed for Discrete Wavelet Transform (DWT), Stationary Wavelet Transform (SWT) and Non Subsampled Contourlet Transform (NSCT) using eight sets of real time medical images. For fusing low frequency coefficients, average and choose max fusion rules are used. For the fusion of high frequency coefficients choose max, gradient and contrast fusion rules are used based on pixel based rule. The proposed technique is performed using eight groups of Positron Emission Tomography (PET), Computed Tomography (CT) medical images. The performance of DWT, SWT and NSCT are compared with four different quality metrics. From experimental output average, gradient fusion rule outperforms other fusion rules from both subjective and objective estimation. It is also observed that the time taken for the execution of images is more for Stationary Wavelet Transform (SWT) than Discrete Wavelet Transform (DWT) and Non Subsampled Contourlet Transform (NSCT).