Brain Tumor Classification using Probabilistic Neural Network

Brain tumour classification is proposed in this work using probabilistic neural networks to handle images and data processing techniques for autonomous detection. Conventionally, the classification and detection of brain tumours are done by human inspection with a medical resonance image (MRI) of the brain. The manually operated methods could be more practical for massive datasets and non-reproducible. During MRI screening, noise is generated, and it leads to serious accuracy issues in classifying the disease. The real-time difficulties should be overcome with the help of artificial intelligence, which is a better solution for this field. Hence, this paper applied the probabilistic neural network. The proposed work was split into two stages: decision-making, performed in two phases; feature extraction using the principal component analysis; and classification using probabilistic neural network (PNN). The performance evaluation of the PNN classifier was based on the network's training performance and classification results. Probabilistic Neural Network provides better classification and is a promising tool for classifying tumours.