Artificial Neural Network (ANN) and Genetic Algorithm (GA) Hybrid Method to Enhance the Prediction of Brain Tumors Cancer

Brain tumor is one of the most frequently diagnosed types of cancer, and understanding the intricacies of the brain is highly intricate and vital, leading to extensive research efforts. In the realm of artificial neural networks (ANNs), particularly in pattern recognition tasks, methods such as feed-forward neural networks (FNNs) are utilized due to their ability to learn complex patterns and handle intricate data structures. However, the efficacy of these networks heavily relies on their architecture and the optimization algorithms used to fine-tune their parameters. In this study, novel yet efficient methodologies for training artificial neural networks are introduced. The majority of contemporary training techniques for artificial neural networks are based on gradient descent algorithms. In contrast to conventional approaches, we propose an advancement by integrating the Evolutionary Algorithm for Brain Tumor Prediction into the training process. Additionally, we explore the utilization of other optimization algorithms such as Genetic Algorithm, alongside conventional methods like Adam. Our research involves the development of an artificial neural network model tailored for pattern recognition tasks, training the model using various optimization algorithms, and evaluating its performance through extensive experiments on the Brain MRI dataset.