Multi-focus medical image fusion is a technique for combining multiple images of the same region with different focus depths into a single image that is in focus throughout. Image fusion utilising multifocal pictures might enhance image perception. Fusing images with multifocal images may improve the visual representation in many cases and is widely used. The ultimate goal is to enhance the visual quality of image pairs. The proposed paper suggests a modified hybrid multi-focus image fusion approach that combines Discrete Cosine Transform (DCT) block-based Quaternion-singular value decomposition (QSVD) with the consistency variance (CV) method of fusion. The performance of the DCT-SVD and DCT-SVD-CV methods is compared with the proposed approach. This method detects focused DCT blocks using the mean of the best 5 blocks. Quality is assessed qualitatively and quantitatively based on measures of mean Brightness (MB), deviation (SD), and Entropy. As a result, this paper contributes to the field by offering an efficient and effective approach to multi-focus image fusion, with the potential for various applications where image quality enhancement is very important.