Abstract:

Early detection of brain tumours is vital for effective treatment. Magnetic resonance imaging (MRI) is a fundamental tool for detecting brain tumours. There are many types of tumours such as glioma, meningioma, pituitary. Accurately determining the tumour type and making this determination is one of the most challenging aspects of classifying brain tumours. The use of artificial intelligence-based computer applications instead of traditional methods of disease detection can make significant contributions to experts in the detection of brain tumours. Especially deep learning methods are effective in disease detection based on the processing of medical images. In the literature, there are many deep learning-based approaches for categorising brain tumours. In this study, a model combining a CNN (Convolutional Neural Network) and a LSTM (Long Short Term Memory) deep learning layer is presented to detect brain tumours with MRI images. It is suggested that LSTM can support the feature extraction capabilities of CNN. In the experiments, it is found that the proposed LSTM-CNN model outperforms the standard CNN model. Using this model, an accuracy score of 98.1% was obtained in the detection of brain tumours. This result shows that it achieves a higher success compared to similar studies in the literature.

Keywords: