Medical image fusion is the technique for consolidating and merging correlative data from two or more input pictures into a composite image to improve the diagnostic ability. In this work, Non Subsampled Contourlet Transform (NSCT), Stationary Wavelet Transform (SWT) and Discrete Wavelet Transform (DWT) based image fusion techniques utilizing distinctive fusion rules are performed on real time PET and CT images. For fusing low frequency coefficients, average and choose maximum fusion rules are utilized. For the fusion of high frequency coefficients energy fusion rule has been utilized on pixel level. The proposed methodology is performed utilizing eight sets of Positron Emission Tomography and Computed Tomography medical images. The performance evaluation of DWT, SWT and NSCT are analysed using four different quality metrics. From experimental analysis it is clear that Non-Subsampled Contourlet Transform (NSCT) performs superior than Discrete Wavelet Transform (DWT) and Stationary Wavelet Transform (SWT) from both subjective and objective estimation.