Brain tumor detection using improved convolutional neural network

Brain tumors are a leading global cause of mortality, with only ۱۲% of adults surviving beyond five years. Early and precise diagnosis plays a crucial role in improving patient outcomes. MRI imaging is one of the most effective tools for brain tumor detection, but automated classification methods can significantly enhance diagnostic efficiency. This study proposes an optimized convolutional neural network (CNN) model designed to classify different types of brain tumors with improved accuracy and reduced complexity. By systematically refining network layers, the proposed model extracts deep features while minimizing computational overhead. In this study, an optimized convolutional neural network model is introduced for brain tumor detection. In this study, the feature of improving the layers of the optimized system and automatically reducing the complexity of the model has improved the complexity of the system by ۴۰% and brain tumor images have been detected. The results, validated through ۱۰-fold cross-validation, achieve an average accuracy of ۹۸.۹۲% and sensitivity of ۹۸.۶۲%, demonstrating the effectiveness of the approach. The findings highlight the model’s potential in aiding clinical diagnosis, offering a cost-effective and practical solution for brain tumor classification

کلیدواژه ها:

Brain tumor ، Convolutional neural network ، Optimization ، Classification