Brain tumor detection using a combination of CNN an‎d LSTM neural networks

1/1/1900 12:00:00 AM

Brain tumors are among the most complex an‎d critical medical conditions, where accurate an‎d rapid diagnosis plays a key role in improving treatment outcomes. This research proposes a hybrid approach based on Convolutional Neural Networks (CNN) an‎d Long Short-Term Memory (LSTM) networks for detecting brain tumors from MRI images. The dataset used consists of 3,264 brain MRI images, which were preprocessed an‎d augmented before being used for training the model. The proposed method leverages the advantages of CNNs in extracting complex image features an‎d LSTMs in analyzing sequential data. Three different optimization algorithms (SGDM, RMSProp, Adam) were eva‎luated for model training. Results indicate that the combination of CNN an‎d LSTM using the Adam optimizer achieved the best performance, with an accuracy of 95.44% an‎d an F1-score of 97.42%. With its high accuracy an‎d efficiency, the proposed model can serve as a powerful tool for automated an‎d rapid brain tumor detection, paving the way for advancements in medical imaging technologies.

شبكه عصبي CNN , تركيب شبكه عصبي CNN و LSTM , تصاوير MRI , بهينه ساز

CNN neural network , combination of CNN neural network an‎d LSTM , MRI images , optimizer

Dr. Javad Vahidi, Dr. Seyedeh Mahboobeh Molavi Arabshahi

Dr. Javad Vahidi