Elevating Neurological Diagnosis: Unleashing Transfer Learning and Single Channel-wise Attention for MRI Brain Tumor Classification using Hybrid CNN Model

There is a pressing need for novel strategies to improve treatment results and diagnostic accuracy due to the rising incidence of brain cancers identified by MRI.This work investigates the relationship between neurology and artificial intelligence (AI), with a focus on MRI-based brain tumor categorization.Through combining the Mo-bileNetV3Large, InceptionV3, and EfficientNetB0 architectures, a hybrid model is developed that achieves an impressive 97.94% accuracy.The inclusion of a single channel-wise attention mechanism is noteworthy since it enhances the discriminative power of the model.A comparative study confirms the hybrid model's advantages over other combinations and standalone alternatives.These results highlight the revolutionary potential of attention processes and hybrid architectures in transforming MRI-based brain tumor categorization, offering improved patient outcomes and clinical judgment.