Brain tumour (BT) is a dangerous neurological disorder produced by abnormal cell growth within the skull or brain. Nowadays, the death rate of people with BT is linearly growing. The finding of tumours at an early stage is crucial for giving treatment to patients, which improves the survival rate of patients. Hence, the BT classification (BTC) is done in this research using magnetic resonance imaging (MRI) images. In this research, the input MRI image is pre-processed using a non-local means (NLM) filter that denoises the input image. For attaining the effective classified result, the tumour area from the MRI image is segmented by the SegNet model. Furthermore, the BTC is accomplished by the LeNet model whose weight is optimized by the Golden Teacher Learning Optimization Algorithm (GTLO) such that the classified output produced by the LeNet model is Gliomas, Meningiomas, and Pituitary tumours. The experimental outcome displays that the GTLO-LeNet achieved an Accuracy of 0.896, Negative Predictive value (NPV) of 0.907, Positive Predictive value (PPV) of 0.821, True Negative Rate (TNR) of 0.880, and True Positive Rate (TPR) of 0.888.
Keyphrases
- deep learning
- magnetic resonance imaging
- contrast enhanced
- end stage renal disease
- early stage
- machine learning
- ejection fraction
- newly diagnosed
- chronic kidney disease
- convolutional neural network
- computed tomography
- peritoneal dialysis
- white matter
- resting state
- prognostic factors
- lymph node
- optical coherence tomography
- high grade
- magnetic resonance
- weight loss
- physical activity
- radiation therapy
- functional connectivity
- free survival
- neural network
- smoking cessation