Brain tumors are generated due to irregular development of brain cells. The life spans of the tumor-affected individual are complex to predict as they appear in different structures and are infrequent. Hence, brain tumor-based Magnetic Resonance Images (MRI) are utilized to recognize the tumors effectively. Moreover, the manual identification of tumors is determined as a complicated and time-consuming procedure and also generates some errors in the prediction outcomes. Thus, computer-aided techniques are generally used to resolve these constraints. Recently, deep learning and artificial intelligence techniques have been widely used to treat brain tumors by MRI images. Therefore, this research work plans to implement a Brain Tumor Segmentation and Classification (BTSC) model with a deep learning-based attention mechanism. At the start, the required MRI images are collected from the benchmark datasets. The collected images are segmented with the Fully Convolutional Network (FCN) approach to obtain the segmented images. Then, the classification of brain tumors is performed with the Convolutional Neural Network with Attention Mechanism (CNN-AM). Finally, the simulation analyses are carried out by comparing them with the conventional brain tumor classification models.
Efficient Deep Learning-Based Brain Tumor Segmentation and Classification Framework Using MRI Images
2023-11-22
1204497 byte
Conference paper
Electronic Resource
English
Multi-band Feature Images Concrete Crack Segmentation Framework Using Deep Learning
Springer Verlag | 2024
|Multi-band Feature Images Concrete Crack Segmentation Framework Using Deep Learning
Springer Verlag | 2024
|A framework of fuzzy information fusion for the segmentation of brain tumor tissues on MR images
British Library Online Contents | 2007
|