Copyright © 2019 by SCITEPRESS - Science and Technology Publications, Lda. All rights reserved This paper executes a novel methodology of medical image fusion depending on two kinds of linear filters. The first filter is mean filter which is utilized to decay the source images to base and detail layers. The local linear filter (guided filter) is utilized to upgrade the source images from one another. Computed-Tomography (CT) images before and after freezing are being used to improve the corresponding data which lead to the right diagnosis. Likewise, examples from human magnetic resonance imaging (MRI) images are utilized to fused T1 and T2 of MRI images together. The fusion result is assessed dependent on standard deviation (Std), peak signal to noise ratio (PSNR), mutual information (MI), root mean square error (RMSE), and universal image quality indexes (UIQI). The combination result is contrasted and three existing combination rules. Hence, the numerical outcomes perform better execution among other combination rules.