A Hybrid CNN-LSTM Network For Brain Tumor Classification Using Transfer Learning

• Published in: 2023 9th International Conference on Smart Computing and Communications (ICSCC) pp. 77 - 82
• Main Authors: Rajeev, S K, Rajasekaran, M. Pallikonda, Ramaraj, Kottaimalai, Vishnuvarthanan, G., Arunprasath, T., Muneeswaran, V.
• Format: Conference Proceeding
• Language: English
• Published: IEEE 17.08.2023
• Subjects: Analytical models; Brain modeling; Brain Tumor; Convolution Neural Network (CNN); Convolutional neural networks; Deep learning; Feature extraction; Long Short Term Memory (LSTM); Magnetic resonance imaging; Magnetic Resonance Imaging (MRI); Stochastic Gradient Descent Optimisation (SGDO); Transfer learning
• DOI: 10.1109/ICSCC59169.2023.10335082

Brain tumor analysis is a major area in medical imaging that necessitates precise and efficient techniques for early detection and diagnosis. Magnetic resonance imaging (MRI) is a popular diagnostic method for detecting and characterizing brain tumors. However, the accurate and reliable analysis of MRI images to diagnose brain tumors remains a challenging task, even for experienced radiologists. Deep Learning (DL) has shown remarkable success in medical image analysis tasks like tumor segmentation, classification, and detection. In this proposed method the MRI images are skull stripped and then pre-processed using Gaussian wavelet filter and a pre-trained model AlexNet is used for feature extraction. These extracted features are trained and classified into four classes using a hybrid CNN-LSTM deep learning model. The model's performance was analyzed for different parameters and it was found that the proposed model has provided better results when compared with the existing models. The validation accuracy obtained by the model is 97.94%.