The goal of this study is to propose a system that combines artificial intelligence (AI) and neuroimaging to develop a prototype that serves as a predictive tool for the detection, localization, and sizing of abnormalities associated with brain tumors. Images from valid data repositories (3929 images) are used in the system. For this purpose, some training and validation processes were carried out through Google Colab, where the image formats are fundamental in proving some anomalies, pre-trained neural networks with deep search networks were also used, thus facilitating the training of the AI-based model and increasing the accuracy of the results. This results in a high percentage of success (86.67%). It is necessary to mention that the training was carried out with different formats, such as TIF, JPG and PNG. These are the most common formats for handling cranial resonances.