BRAIN TUMOR CLASSIFICATION USING CONVOLUTIONAL NEURAL NETWORK WITH PARALLEL PROCESSING

The recent decade has witnessed the rise of machine learning implementation in various domains. The processing and classifications of medical images such as the MRI images of the brain tumors is one of the disciplines where various approaches and methodologies using machine learning algorithms have been proposed. The paper compares the serial implementation with the parallel implementation of the machine learning model to classify brain tumor MRI images. The implemented model for serial processing provides an accuracy 90% and f1 score of 0.93. Whereas the implemented model for parallel processing provides 94% and f1 score of 0.95. As the mortality rate of the brain tumor is extremely high. Hence, discovering the tumor cells in the initial phases and at a faster pace is one of the best ways to eliminate cancer. Using machine learning algorithms with the combination of image processing methods is a powerful tool that not only improves accuracy, and precision but also speeds up the process to classify whether the tumor is malignant or benign.