Advancing Efficient Brain Tumor Multi‐Class Classification: New Insights From the Vision Mamba Model in Transfer Learning

ABSTRACT Early and accurate diagnosis of brain tumors is crucial for improving patient survival rates. However, the detection and classification of brain tumors are challenging due to their diverse types and complex morphological characteristics. This study investigates the application of pre‐trained models for brain tumor classification, with a particular focus on deploying the Mamba model. We fine‐tuned several mainstream transfer learning models and applied them to the multi‐class classification of brain tumors. By comparing these models to those trained from scratch, we demonstrated the significant advantages of transfer learning, especially in the medical imaging field, where annotated data is often limited. For each tumor class, the dataset was partitioned into training, validation, and testing sets using an 8:1:1 ratio, with data augmentation and early stopping techniques being implemented to mitigate overfitting. This study employed 1554 T1‐weighted MRI images categorized into six tumor types. Notably, in this work, we introduced the Vision Mamba (Vim), a novel network architecture, and applied it for the first time in brain tumor classification, achieving exceptional results. Experimental results indicate that the Vim model achieved 100% classification accuracy on the testing sets, outperforming other competing models, such as CNNs and Transformers, across all evaluation metrics, emphasizing its potential for tumor classification tasks. Results underscore the effectiveness of transfer learning in brain tumor classification and reveal that, compared to existing state‐of‐the‐art models, the Vim model is lightweight, efficient, and highly accurate, offering a new perspective for clinical applications. Vim learns visual representations through sequence modeling, enabling it to capture a data‐dependent global visual context. Concurrently, by leveraging Mamba's hardware‐aware design, it operates with lower computational complexity. Furthermore, the framework proposed in this study for brain tumor classification, based on transfer learning and the Vim model, is broadly applicable to other medical imaging classification problems.