A HYBRID CNN-LSTM APPROACH FOR BRAIN TUMOR CLASSIFICATION: A COMPARATIVE PERFORMANCE ANALYSIS WITH CONVENTIONAL CLASSIFIERS

Early diagnosis of brain tumors is vital for improving patients' quality of life and optimizing treatment processes. This study proposes a novel model combining CNN (Convolutional Neural Network)-based deep feature extraction and LSTM (Long Short Term Memory) networks in a hybrid structure for the classification of brain tumors. The CNN used for deep feature extraction processes spatial information obtained from Magnetic Resonance Imaging (MRI) images, while the LSTM network has the ability to analyze this information through sequential patterns. The performance of the developed hybrid model is evaluated comparatively with ANN (Artificial Neural Networks), KNN (K-Nearest Neighbor), SVM (Support Vector Machines), and Random Forest classifiers. Experimental studies show that the proposed CNN-LSTM model outperforms other methods with 97.15% accuracy. The findings show that the sequential data processing capability of the LSTM model increases image classification performance when used as a hybrid with CNN-based features, demonstrating the potential of hybrid architectures and providing a powerful and innovative solution that can be used especially in the field of medical imaging.