Abstract:
Brain tumor is one among the most dangerous diseases in the world, patient’s life can be saved if the brain tumor is detected and diagnosed properly in its earliest stages. Since brain has the most complex structure in which tissues are interconnected rigorously. Thus makes the brain tumor detection a challenging task. Brain tumor detection and classification requires clinical experts to meet the standard level of accuracy.
This limitation is overcome by the use of Computer Aided Diagnosis Systems (CAD Systems) in the diagnosis of brain tumors. In this thesis propose an efficient method for brain tumor detection, also the thesis is interesting in determines which type of artificial neural network is the best for image recognition, Neural network must be able to determine the state of brain according to magnetic resonance imaging and determine whether it normal or abnormal state.
Data collection was from Harvard citations,cancer imaging archiveand figshare date base.
From each MR image texture features are extracted using Gray Level Co-occurrence Matrix to prepare training data which was introduced to neural network as input and target vectors. Three neural network are designed and trained using MATLAB feature nntool which are Cascade feed forward, Feed forward and Learning vector quantization, After testing, the Feed forward network achieved performance ratio equal 97.91 %, also Cascadefeed forward ratio was 96.88%, while Learning vector quantization performance ratio was reach to 56.25%.
