It is possible to gather or collect medical images from a variety of modalities, which results in the need to handle a variety of distinct practical issues associated with medical image registration. In multimodal registration, intensity-based registration approaches are being applied; similarly, these algorithms will add up extra images that have shared the information in a single representation by utilizing image transformation. The transformation of photographs into their optimal form is required in a significant way. It is of the utmost importance that the optimization of the similarity metric is performed on each of the distinct photos. In recent times, an enormous number of optimization strategies have been suggested, each of which concentrates on the constructing measures of the optimization component. However, there is still a significant amount of room for development, both in terms of the efficiency of the processes and the quality of the end product. The brain MRI registration process is very important for recognizing the spot differences that exist between the many MR images of the brain that are used for diagnostic purposes. In this paper, a brain MRI registration using a fireworks algorithm with the improved adaptive transfer function for explosion generation spark (FWA-IATFGS) is proposed. FWA-IATFGS is a Meta heuristics-based Optimization algorithm that we have proposed to register brain MR images, for multimodal medical image registration. In this paper, the FWA-IATFGS algorithm is used to register brain MR images. To begin, we extracted 30 magnetic resonance images of the brain from the data set using three different modalities, including T1, T2, and FLAIR. Following that, the photos are registered with the help of FWA-IATFGS. In the end, the photos from the register are extracted. The outcomes of the suggested method for brain MRI registration are compared with the outcomes of a registration method that is based on particle swarm optimization (PSO). The results of the experiments show that the proposed strategy achieves statistically superior results to other methods when it comes to the registration of MR images of the brain.