The identification of brain tumors is an important task in medical imaging. Manual classification of brain tumors can lead to inaccurate diagnoses due to various restricting factors. Brain tumors can be difficult to extract from images because they have diverse appearances and can look similar to normal tissue. This research study proposes a new way to identify brain tumors with brain pictures using convolutional neural network. The proposed method is tested on a dataset containing different tumor sizes, tumor shapes, tumor locations, and tumor image intensities. Simulation results shows that proposed classifier successfully identifies different types of tumors i.e. implemented a CNN with Keras and Tensorflow, which outperformed the traditional classifier with an accuracy of $\mathbf{9 7. 8 7 \%}$.
Classification of Brain Tumor based on Convolution Neural Network
06.11.2024
593318 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Travel Mutual Information Classification Based on Convolution Neural Network
Springer Verlag | 2022
|